Home

LUCID User's Guide - Experimental Subatomic Physics

1. save Histograms Clear Outstanding Commands E all all all oH Histograms READER Input Variable List davey davey_cut spectrum spectrum_O spectrum 2d spectrum cut The READER needs a value for the variable zed Enter Z position for source run mm Q Save Format v xlucid o o zero suppress binary output O append to file Filename amp index by bin index at midpoint Control Data File Experiment Data file Save Selected Variables Changing the Input Data File Menu Buttons The Control Menu Figure 2 7 Sub windows from the Control Menu The User Commands menu option allows the user to issue com mands which were specified in the READER or LOOKER de scription files A sub window will be displayed which contains a button for every command given in a description file If the same command was listed in both the READER and LOOKER then selecting that button will cause the command to be sent to both The User Commands menu option will be made inactive if no user commands were defined The Change Input File option allows the user to change the source of data only when off line The file name should be a com plete path name meaning that it should start with a slash The user should keep in mind that LUCID might rebuild the soft ware if it encounters some change in the way data has been saved For example if
2. Local Varlable List E7777 d junkdata Filename vega dev nrst1 W SAVE O Event Filter L FailonError Program Event List E Figure 2 4 Sub windows from the Build Menu Remove Selected Variables tem administrator to determine the correct names of tape drive units The user is allowed to change the input file after the software is built when off line see Figure 2 7 under the Change Input File option for details The most important and time consuming step is the preparation of the experiment description files Chapters 3 and 4 describe Menu Buttons The Build Menu The LUCID Program 2 9 Using LUCID Check for Status or Error Messages Build Local Variables what the files may contain The description files can be prepared beforehand using any text editor The user should first select the LOOKER button if a LOOKER de scription file is to be used Recall that the READER is always present Next the user can select the appropriate Edit buttons to allow writing or modifying the descriptions When finished re member to get out of the editor completely by bringing up the menu from the top border of the edit window and choose Quit The user can select the button labelled Make at the bottom of the window Software will be generated and compiled If some mis takes were made in the description files error and status mes sages will be displayed look under the View menu to display messages
3. The Make button is actually a menu button having a menu which contains four options The user may have the software generated compiled loaded and started all in one operation which is the default choice The second choice is to simply gen erate the software to make sure that the description files are ac ceptable In this case nothing is compiled loaded nor started running When debugging a looker file the above two options are also available with array limit checking This causes the gener ated looker code to check that all accesses to arrays are within bounds Figure 2 5 shows the menu available with the Make button Build and Use Software i Build Software Don t Use it Build and Use with Array Checking i Build with Array Checking Don t Use it f Figure 2 5 The Make Menu Button The Build Options button brings up a window not shown that passes user values to the MANAGER for a greater degree of con trol over generating the READER and LOOKER To use this you should be familiar with the different programs that are used when generating LUCID software Other options are available from the main Build menu The Build Local Variable option allows the user to define a variable just as is done in a LOOKER description file and use it for tem porary storage of data The intention is that eventually arith metic involving histogram data may be done dynamically and local variables can be used to store temporary results
4. Bin Z Limits Adjust View spectrum Modes amp Options X Limits amp Rebin Y amp Z Limits Markers X Limits To 2048 Apply To View 2 Insert Dup View 4 Delete View Rebin Factor 1 Ir gv Adjust View spectrum Apply To X Markers Clear Markers Set To Limits Y Markers Insert Dup View Delete View Markers The View Menu Button Tea tae i E Insert Dup View Bin Delete View Figure 2 18 The Adjust View Sub window While it looks rather intimidating the sub window allows the user to go through simple and logical steps to adjust the display Depending on the selection made in the upper left hand corner one of the four sets of controls are displayed The user should keep in mind that the data is never being changed when these set tings are adjusted the view is changing and nothing else Al though this window can be called up by the View selection on any histogram window there is never more than one Adjust View window on the screen at any given time Before discussing features of this sub window it is important to understand markers Markers are simply two limiting lines drawn over bins in a histogram display Many of the options for The Histogram Windows 2 27 Using LUCID Histogram Sub Views The View Menu Button adjusting the display use markers When drawn on the display a marker occupies the same width as the bin over which it is placed Marke
5. LUCID users are allowed to read through the file and even make their own additions A user is allowed to supply his own data base file if he needs to access a module in a non standard way LUCID will read the user s file and then the system database us ing the first entry with a matching name When searching for module names in the database LUCID considers only alphanu meric characters There is no differentiation between upper and lower case characters For example the names LeCroy 2228 LeCroy2228 and lecroy2228 all match the same module LeCroy 2228 Time to Digital Converter The format of the module database file is described in detail in Appendix A of this manual The entries are in alphabetical order for easier reading When creating your own module names and aliases try and maintain a consistency with the database E g 4508 LeCroy 4508 3615 ks3615 ksc3615 Kinetic Systems 3615 The B and C keywords in our example are optional LUCID as sumes that the module is in branch 0 crate 1 unless told other wise Referring back to the original example define delta e 2249W N 1 Although this module is more complicated than the scaler of the previous example it s definition is smaller because all inputs are to be read when it is accessed The sub address code is optional and LUCID assumes all sub address codes are to be used if you don t list any in your definition The DEFINE Section CAMAC Variables Making a Reader Descrip
6. cept all occurrences of all events Both destinations should have identical Fail on error settings Having both fail on error in creases the chances of a tape write error aborting a run If both silently exit there is a remote chance that both destinations 5 3 o The Writer could stop writing with only a message to the view window If the experimenter misses both termination messages the experi ment would continue running but would not record any data The only other feedback is the Writer Bytes written this run box which would stop updating the number of bytes acquired opem 0 Chapter Conceptual Overview How LUCID Works Looker Writer When you tell LUCID to make an experiment it takes your de scription files and writes several computer programs Each of the READER LOOKER and WRITER are separate programs which LUCID may produce along with a separate program to access CAMAC if that was specified These programs are compiled linked and executed automatically when you say make but only if it s necessary For example if you change your READER de scription file slightly then only part of the READER program will be rewritten recompiled and fit into the system In any case the programs are started and after setting up communications with one another will wait for your command The data stream is on line in this case and the data is placed in shared memory The experimenter may place certain restrictio
7. low manipulation of the data or its representative image The window depicted in Figure 2 15 show the File View Edit and Print menu buttons which are associated with every histogram Many features within these menus are used quite often and LU CID understands several different keystrokes to perform the same functions as various menu selections Users will most probably use these simple keyboard shortcuts more often than The Histogram Windows 2 23 Using LUCID Save To File Read From File E Get Region Send Region Save Region spectrum Shortcut Key s The File Menu Button Figure 2 15 Menu Buttons in a Histogram Window menus and sub windows Before discussing the shortcuts we ll look at the menus The File Menu Button The File menu contains two options The first is labelled Save to File and allows the user to save histogram data for future ref erence When selected a sub window is displayed which allows different features to be used as shown in Figure 2 16 The user should select the desired options in the Save Data to File sub window as subsequent use of the save feature uses the same settings by default for all histograms The settings are not saved for the next time you run the lucid program The first field refers to the name of the file in which data will be saved By default it consists of the word Data followed by a pe riod then the name o
8. module LeCroy Model 4299 databus interface 4299 LeCroy 4299 shortword writesize 1 readsize 100 qstop channel length 0 value length 16 offset 0 read is f0 a1 readclear is f0 a0 write is f16 a0 clear is f9 a0 enable is f26 a0 disable is f24 a0 testlam is f8 a0 clearlam is f10 a0 module LeCroy Model 4300 16 channel fast encoding and readout ADC FERA 4300 LeCroy 4300 timeout 9 readsize 16 shortword read is f2 a0 15 write is f16 a0 test is f8 a0 clear is f9 a0 module LeCroy model 4413 16 channel 150 MHz updating discriminator 4413 LeCroy 4413 readsize 1 shortword read f1 a0 write f17 a0 enable f26 a0 disable 24 a0 module LeCroy model 4418 16 channel programmable logic delay fan out 4418 LeCroy 4418 shortword write is f16 a0 15 module LeCroy Model 4434 32 Channel 24 bit scaler 4434 LeCroy 4434 readsize 32 hold Sample CAMAC Module Database A 13 longword read is f2 a0 test is f8 a0 write is 16 a0 module LeCroy Model 4448 48 bit coincidence register pattern unit 4448 LeCroy 4448 readsize 3 shortword read 0 a0 2 readclear f2 a0 2 clear f9 a0 2 test f8 a0 2 module LeCroy Model 4508 Dual 8 input 8 output programmable logic unit module module 4508 LeCroy 4508 readsize 4 shortword writesize 256 hold shortword read is f0 a0 3 readclear is f2 a0 3 clear is 9 a0 1 write is 16 a0 3 24 bit word generator swit
9. timeout to trigger uninhibit using M wait when write 4 42 _ ae Chapter The Writer The third part of LUCID is called the WRITER and its sole func tion is to record the data produced by the READER and LOOKER LUCID runs the WRITER when an experiment is built and the user is responsible for informing the writer of the destination of the data Although the WRITER is only concerned with saving data it is capable of saving it in a variety of ways For instance you can tell the WRITER to write certain events on a tape using LUCID or to write only as many of certain event types as will keeping the acquisition from blocking Unlike the READER and LOOKER there is no WRITER de scription file This allows changing writer destination informa tion without requiring a rebuild of the experiment However destinations can only be changed when data acquisition is in a STOPPED state This helps to ensure the integrity of a run s data Data is saved only is a LUCID specific format There are func tions available to read this data into user written C or FOR TRAN programs see Reading LUCID Data Files on page 7 10 LUCID also reads this data when running an off line experi ment which is the easiest way to do more intricate calculations on your data The rest of this chapter assumes familiarity with the Build Writer window described on page 2 11 or with the Modify writer list option of the terminal oriented lucid com
10. All destinations are considered non criti cal If one fails the user can stop the experiment normally and all other destinations will have proper end run records and ap propriate file marks The drawback of the approach is the re quirement for constant vigilance The experimenter must monitor the message display window continuously watching for an error message from any of the writers The risk is that all the messages could be missed and once all writers have failed the experiment still continues merrily along sending all the data to the bit bucket The safest approach when acquiring data is to have only one de vice selected to fail on error This device should be writing all oc currences of all events When this device fails the acquisition will shut down and demand attention from the experimenter 1 At the time this section was written the WRITER software that distinguishes individual event re quirements was not completed It is hoped that the delay between documenting this software and writing it is minimal 5 2 m The Writer Changing Tapes Multiple Tapes Efficiency The WRITER will work most efficiently when writing every event to the same place When it has to look at an event and decide where it should be written the entire data stream will be forced to slow down It is best to save all types of acquisition events to tape or disk and then separate events later with a second off line LUCID data strea
11. In any case variables are defined in exactly the same way as in the READER file as shown in the following example define energy real 4 This defines a single precision variable called energy The same rules for defining READER variables apply in the LOOKER The DEFINE Section Types of Variables Making a LOOKER Description File 4 3 The Looker Defining Data Stream Variables The format for defining variables is somewhat standard the word define is followed by a list of names which is followed by the type of variable define delta e real 4 define hits ecounts fcounts int 4 define valid boolean Notice that the word integer can be abbreviated to int Variables which are expected to be seen in the data stream may be defined in the LOOKER Such variables have already been de fined in the READER and defining some of them a second time in the LOOKER allows it to check that its using the right data stream In fact new space for these variables isn t allocated as with other variables the definitions are used only for type checking To define these READER variables simply add the keywords previous data after the variable names For example define d a previous data define o tdc previous data int 2 The first example tells the LOOKER to make sure that some vari able named a exists in the input stream The second line makes sure that o_tdc exists but also makes sure that it is of type inte ger 2 Defining READER variable
12. coincidence 0b001 then trigger calibrate elif coincidence 0b010 then trigger hadron elif coincidence 0b100 then trigger electron else bad_coincidence bad_coincidence 1 endif Other statements are introduced in this example The trigger statement is a software trigger of the named event allowing dif ferent event generation from one initial trigger type The EVENT Section Making Decisions about Making a Reader Description File 3 25 The Reader Triggering Another Event Some event triggers may not provide sufficient information for the READER to trigger the desired event type directly In this circumstance a conditional trigger can be made which causes an other event to be triggered at the end of the current event even if the current event is rejected It is even possible to trigger multiple events to occur after the current event the events occur in the or der they are listed in the current event A conditional trigger ap pears as if coincidence 1 then trigger eventname endif Rejecting an Event After making certain tests you might determine that some data is in error or unsuitable for what you wanted to do In such a case you can use the reject statement to throw away the event When the READER sees the reject statement it simply stops what it was doing it backtracks over anything that you might have saved to that point and goes back to waiting for a trigger There is no indica
13. when implemented will allow a simple re map ping of displayed density blocks on scatter plots Currently the option brings up a density map window which affects nothing The final implementation of this window willlet the experiment er adjust the displayed density size for differing bin content val ues The Get Region Option Choosing Get Region starts a window that lists all the region vari ables in the looker having the same number of subscripts di mensions and bins per dimension as the displayed histogram as shown in Figure 2 34 Only one region can be selected or as signed to a histogram at any one time The DONE button avoids doing any assignment Pressing Assign Selected Region causes the named region to be associated with the current histogram and a request made to the looker to return the current region val ue Notice that if the region is already associated with another The Histogram Windows 2 44 w Using LUCID The Edit Menu Button Get Region Experiment demo glen skatter Variable List i zi j pokey Variable spectrum Name gumby 12 Type Region x axis 512 bins Y akis Assign Selected Region Figure 2 34 The Get Region Window histogram the region values are overwritten with the looker val ues The Send Region Option The region values associated with the current histogram are re turned to the looker Any further calculations made by the look er using the named region
14. 1024 bins The maxi mum number of bins in a one dimensional histogram is 4096 Two dimensional histograms default to 128 bins per axis The maximum number of bins in a two dimensional histogram are 256 bins per axis These limits attempt to keep the virtual mem ory requirements of LOOKER programs from becoming exces sive When you need to look at an area of a histogram in greater detail try restricting the X or Y range to just the area of interest or define multiple histograms with adjacent ranges incremented from the same variables Using Histograms In the event section of the LOOKER description file histograms can be used in expressions just as other variables would Let s skip ahead for a moment and see how you can actually get a his togram to count Using the histogram from the previous define example we could say increment pl using adcval This tells the LOOKER to check the value of the adcval1 variable and increment the corresponding bin within the p1 histogram For a complete description of the increment statement refer to Incrementing Histograms on page 4 16 Histogram Regions LUCID strictly enforces the rule that all bins in the same histo gram must be the same resolution Sometimes it is important to deal with a range of values which is larger than a single bin s ac ceptance but still keep the resolution of a single bin In other words we need the ability to deal with a group of bins which Histogram Regions Def
15. 200 marks the start of the last bin which counts the values 200 and 201 making 101 bins The next exam ple further explains this idea In all of the examples so far it is assumed that only integer values will ever be checked but it is also possible to count values con tained ina real variable LUCID makes no distinction as to what type of variable it is counting In many cases the range of accept able values in a histogram could be a very large or small interval For example define de 5 energy hist first 20 last 40 binsize 0 2 Here we have an array and a variable in which only values be tween 20 and 40 inclusive will be counted and which contain bins which are 0 2 counts wide In other words any value be tween 20 0 and 20 2 will be counted in bin 0 20 2 40 0 20 0 c 39 8 40 2 Bin 101 Strictly speaking the value would have to be less than 20 2 to be counted in bin 0 since 20 2 marks the first value accepted in bin 1 Due to the way the LOOKER calculates bins for incrementing Bin 101 will contain counts of values 40 and 40 2 which goes against the intuitive counting of only the value 40 Histogram Definition Defining Histograms LOOKER Histograms 4 31 The Looker How Histograms Count Counting Negative Values Default irst and last histogram values Restrictions This brings up a very important point in the LOOKER A vari able s value is always included in the bin
16. 40 Incrementing Histograms 4 33 Incrementing histograms 4 16 Keywords beginrun 4 11 bit 4 40 call 4 25 command 4 11 contains 4 22 decrement 4 17 endrun 4 11 if then else 4 21 increment 4 17 LUCIDbytes 4 13 LUCIDonline 4 13 LUCIDrun 4 13 LUCIDsequence 4 13 LUCIDtime 4 13 notregion 4 19 print 4 24 region 4 18 repeat 4 23 Keywords List 4 41 Operators 4 14 Printing values 4 24 Regions 4 33 2 dimensional 4 39 Arrays 4 35 Definition 4 33 4 34 Restrictions 4 36 use with histograms 4 18 Repeat statement 4 23 Subroutines 4 25 Summation operator 4 14 User specified commands 4 11 Variable array 4 6 Variable assignment 4 13 Variable groups 4 7 Variable types 4 3 Variable use 4 12 lucid 7 3 LUCID definition 1 1 lucid command 7 8 LUCID File Format 7 15 lucid library 7 10 lucidlog 7 3 lucidman 7 4 lucidview 7 4 M Markers Description 2 27 Modes Adjust View 2 28 Modules Use Hardware Database option 2 11 N netcamac 7 5 nethv 7 5 O Operators In Looker statements 4 14 In Reader statements 3 23 P Pause button 2 21 Playback Sequence button 2 20 Printing From Reader 3 24 Properties menu 2 15 Properties window 2 16 Q Quit button Xlucid 2 18 qview 7 7 R Reader read and clear statement 3 19 save data statement 3 18 Arrays 3 24 Index 3 ee Index Calculations and Variables 3 23 Calling Programs with Arguments 3 16 CAMAC see CAMAC camac keyword 3 28 CAMAC Modules Databas
17. Camac operations are performed by NetCamac where e b is the branch an integer between 0 and 7 cis thecrate an integer between 1 and 1 nis the station an integer between 1 and 23 ais the address an integer between 0 and 15 f is the function an integer between 0 and 31 data is the address of a long integer which is either the source or destination of the data for the operation NetCamacSetDebugLev int level sets the debugging level for CAMAC operations on the front end computer DIAGNOSTICS On error all routines send a message to stderr and return 1 If successful NetCamacOpen and NetCamacPing re turn 0 NetCamac returns the status bits X bit 0 and O bit 1 High Voltage Access The NetHv routines provide network access to the 4032 and 1440 high voltage controllers These routines are a standard part of the LUCID library NetHvOpen char name name is a string specifying the high voltage controller to con nect to This string is of the format bBRANCHcCRATEn SLOTtTYPE MACHINE where MACHINE is the front end computer BRANCH is the CAMAC branch to use CRATE is the crate in the branch SLOT is the module address and TYPE is the type of module either 4032 or 1440 that is in that slot a daemon process is started on the front end computer and it is this process that the NetHv routines communicates with 0 is re turned on success 1 on error NetHvFrame int frame Sets the frame entry for
18. Chapter7 documents the many different support subroutines and programs that may be helpful when running a LUCID experiment Where to Go from Here 1 17 A Tutorial Introduction Each section in this guide starts out with just enough informa tion to let you use a particular feature More details are given as you read further into the section so you can probably skip a lot if you re getting started When examples are given LUCID key words are printed in a typewriter typeface trigger scaler every 10 minutes Brackets around a word indicate that the named words are op tional We will assume that you re familiar with CAMAC and the variety of electronics that are available through it Where to Go from Here 1 18 _ e Chapter Using LUCID LUCID was designed to be easy to use It was also designed to help experimenters think about their work on a higher level As you read this manual try to forget the details of your experiment and think in more general terms Most experiments are quite complex in nature and the best way to solve the problems and get meaningful information is to de sign the experiment from the top down start with a very general statement of what you want then break the job into a few specif ic tasks Each of these tasks should then be divided into jobs which are more detailed and this process repeated until each task addresses a unique detail of the experiment LUCID fits into the middle of th
19. How to DEFINE Variables on page 3 6 A variable de fined as character can contain any valid ASCII character be tween 40 octal and 176 octal i e between the space character and the tilde character the return and newline characters and the tab character The null byte is reserved for internal use and marks the premature end of a string of characters in a larger vari able Boolean variables are numeric variables which may only contain a one or zero A value of one denotes a true state and zero a false one Making a Reader Description File 3 12 _ lemme The Reader Constants LUCID understands integer constants in binary octal or hexadec imal base two eight and sixteen respectively To specify a con stant in base eight for example you should make the first digit a 0 as in 0351 Similarly 0b11101001 and 0xe9 refer to the same number 233 in base two and hexadecimal Base 16 numbers can use either upper or lower case letters A through P Statements in an event specification can access variables includ ing CAMAC modules and arrays Character variables are used infrequently usually to record an operator s name or the name of an experiment within the data stream Recall that all variables must be defined in the define section of the description file When elements of an array are accessed an index must be given between brackets following the array name such as phi 3 Remember that all
20. Ideally we would like to keep 10 different histograms for delta e forevery value that energy contains we want to see how the delta e values change delta e s histogram built when energy contained 10 delta e s histogram lh A built when energy contained 11 Eight other histograms corresponding to the remaining possible values of energy One solution is to simply define an array of 10 histograms which would countdelta e values and update the appropriate histo gram by using the energy value as a subscript Of course we would have to subtract 10 from the energy value to get the cor rect subscript but it would work fine The ultimate goal from such a procedure is to determine how delta e values relate to energy values if at all For instance we might be able to deter mine that delta e values are higher whenever energy has a relatively low value LUCID allows an easier method of relating one variable s values to those of another A two dimensional histogram can be defined which allows two variables to be considered at the same time It is assumed that both variables will change at the same time be cause usually they are both part of the same event from the READER Functionally a two dimensional histogram will do ex actly what the array of histograms did in the previous example and we can visualize it as a matrix of bins rather than a simple list of them Defining Two Dimensional Histograms
21. LOOKER Histograms 4 37 The Looker Two Dimensional Histogram Arrays Possible values of tdc For example consider having to count values ranging from 0 to 18 in one variable named adc and a second variable named tdc having values from 0 to 8 A matrix of bins could be visualized like this 0 Possible values of adc 18 In this particular representation the histogram counted several instances where adc contained the value 5 when tdc contained a 2 Even more counts were made when adc and tdc both con tained the value 3 One advantage of using a two dimensional histogram is that both variables can have ranges of arbitrary values as a regular histogram does The following example defines this type of his togram define ta common hist from 0 to 18 and 0 to 8 The two ranges of values are separated by the keyword and which tells LUCID that t common is a two dimensional histo gram This particular example shows an alternative to using the array of histograms for delta e in the previous discussion which used energy values as subscripts Visually we can think of this histogram as a graph with an X axis from 0 to 18 and a Y axis from 0 to 8 Technically when new adc and tac values are found the corresponding bins in each axis are determined and the intersecting counter is incremented Arrays of two dimensional histograms are allowed just as with any other variable but one common problem to avoid is using too
22. LUCID defines it for you It can be used wherever a normal boolean variable could appear Is best suited to the if com mand described under Making Decisions about the Data on page 3 25 The qresponse variable will contain a one true value if the last CAMAC operation returned a Q response and it will be zero when there was no Q response You should not assign some val ue to the qresponse variable as you would other variables as it is considered read only Compressing Array Data In many experiments a sparse matrix of data may be acquired For instance a wire chamber could have hundreds of TDC chan nels most of which will contain irrelevant data when a trigger arrives When a computer goes to read and save the data most of itis unimportant LUCID allows you to compress an array by specifying what you consider to be unimportant The compression actually refers to discarding all unimportant data and adding an index for each good value For instance if an array of one hundred TDC values contained unimportant data except for the last two elements the compressed array would contain four elements the number ninety eight and the second last TDC value followed by the number ninety nine and the last TDC value 1 The definition of a timeout delay is discussed on page 3 6 The EVENT Section Compressing Array Data Making a Reader Description File 3 21 The Reader The compress Statement More General Software Dis
23. LUCID files the lines in the CAMAC database are free format The recommended format for the CAMAC database is that each module line start at the left hand margin and that all subsequent lines for that definition are indented by one TAB character This makes it easy to read the definition and find mod ule names quickly and it also allows you to print a list of all the modules very easily using the grep command in UNIX Com ments are encouraged inside the database file and are intro duced by an octothorpe f character except when it is inside a character string All characters from that point to the end of that line are ignored as in all other LUCID files You should also leave at least one blank line between subsequent CAMAC module def initions just to denote the end of one definition and the start of another After all the aliases for a module are given in the file all of the appropriate CAMAC capabilities should be listed CAMAC Capabilities There are twelve different CAMAC capabilities that LUCID un derstands each of which are selected by a unique keyword These keywords can appear in any order after the name aliases Readsize and Writesize The readsize keyword tells LUCID the maximum number of data transfers that this module can accept at one time In the ex ample of our 6 input scaler the statement would be readsize 6 Defining the Modules A 2 od mHEEER Similarly the word writesize tells LUCID how m
24. Looker 4 16 Built in Looker variables 4 12 Calculations in Looker 4 13 Calculations in Reader 3 23 Character variables in Looker 4 16 Defining in Reader 3 6 Groups in Looker 4 7 Looker groups 4 7 Use in Looker 4 12 Using variables Reader 3 12 View Data window 2 14 View Menu Xlucid 2 14 View menu button histogram 2 27 View Messages window 2 15 vme console 7 7 X Xlucid 1D Regions 2 37 2D Regions 2 38 Adjust View Stepsize 2 30 Adjust View sub window 2 27 Adjusting histogram markers 2 32 Change Input File 2 13 Control buttons 2 19 Control Menu 2 12 Eject button 2 22 Forward button 2 22 Histogram Density Map 2 44 Edit Data 2 44 Get Region 2 44 Print menu button 2 46 Save Region 2 45 Send Region 2 45 Histogram 2D markers 2 32 Histogram Autoscaling 2 30 Histogram Display Bin Numbers 2 30 Index 5 ee Index Histogram Display Options 2 29 Histogram Grid Lines 2 30 Histogram Markers description 2 27 Histogram menu buttons 2 23 Histogram Modes Marker 2 34 Pan 2 42 Pan temporary 2 43 Region 2 37 Scale 2 39 Scale temporary 2 43 Title 2 44 Window 2 44 Zoom 2 41 Zoom temporary 2 43 Histogram Rebinning 2 31 Histogram Shortcut Keys Autoscale a 2 32 Bin Numbers b 2 32 Delete DEL 2 34 Grid Lines g 2 33 Insert INS 2 34 Mode 2 34 Statistical Information S 2 33 Stepsize 0 9 2 33 Updating u 2 33 Histogram Statistical Data 2 30 Histogram Sub views 2 28 Histogram View m
25. Reader Description File 3 19 The Reader Read Save and Clear Operation The wait command The rule to remember is that when read and clear commands ap pear in the same statement and if the module supports the read and clear operation the module is issued a single CAMAC com mand containing usually F 2 If the commands are in separate statements or the module has no single read and clear opera tion then two separate CAMAC commands will be issued The clear command can be combined with the earlier read and save command giving read save and clear adc This is particularly efficient method for many modules Polling Camac Modules The term polling a module refers to the action of repeatedly ac cessing a module until it says that data is ready It is essential when reading from modules that have a lengthy conversion time but do not issue a lam when finished Note that a lam source signifies a unique trigger and LUCID can t perform a por tion of an event when a trigger arrives Therefore if several mod ules are to be read only one of them can issue a lam As mentioned on page 3 6 modules with lengthy conversion times should not be the ones to issue the LAM and they should be read later in the event giving them time to finish In such cases poll ing the module is required The keyword wait can be used to poll a module by giving it the name of an associated CAMAC variable For example wait adc will ca
26. The data storage requirements of this long style compression is the same as the previous example but the extracted value is treated as a long when doing data calculations channel value LVAR LONG channel bit offset 16 channel bit length 16 value bit offset 0 value bit length 16 This is similar to the old style LCMP_LONGLONG compres sion channel value LVAR LONG channel bit offset 0 channel bit length 32 value bit offset 32 value bit length 32 LUCID Data Record Format Declarations LUCID Data File Format 7 21 o Lue Utility Programs and Subroutines This example shows an extra bit which is treated as part of the value channel value LVAR LONG channel bit offset 17 channel bit length 15 value bit offset 0 value bit length 17 struct Lucidcompressoffset long lco begin Offset of first byte long lco end Offset of just past last byte hi Extra information stored in symbol table for compressed vari ables struct Lucidcompressinfo u_char lci_channel offset Bit offset of channel u_char lci channel length Bit length of channel u_char lci value offset Bit offset of value u char lci value length Bit length of value E Extra information stored in symbol table for histogram variables struct Lucidhistinfo u short lh nbins LHIST MAXDIMS number of bins per dimension float lh lolimit LHIST MAXDIMS lower limit values per dimension float lh hilimit LHIST MAXDIMS upper limit
27. The structure of an event record is determined by the order in which variables are saved in the event specification For those users wanting to write their own software the build command puts a definition of the event record into a LUCID file A C in clude file containing the structures describing the event data is also produced There are also subroutines available to read the event record so it is unnecessary to manually write code which knows the actual event record layout These features are further described in Chapter 7 A special form of the save command is save data The EVENT Section Saving Data Making a Reader Description File 3 18 eee The Reader Clearing CAMAC Modules Read and Clear Operation This statement causes any data buffered by the front end com puter to be transmitted as soon as possible For an experiment with a low acquisition rate this statement could be placed in an event triggered every few seconds or possibly in an event trig gered by a user command Alternatively the special save command can be entered as save data and wait As with the save data command this immediately causes the data buffer to be transmitted to the user s LUCID session How ever acquisition will not continue until the LUCID session has processed this data The effect is that none of the different soft ware sections of the experiment get too far ahead of any other section and responsiveness to user commands inclu
28. This describes a histogram in which the first bin starts at 0 and the last bin starts at 1023 Defining a histogram in this way en sures that the upper limit is included in the counting proce dure The following definition would result in exactly the same histogram define spectrum from 0 to 1024 1024 bins The keywords rom and to may seem more natural but might seem ambiguous in some cases since the upper limit is never re ally included in the counting Using the irst and 1ast key words introduce other conceptual problems however such as the acceptance of values within each bin This is best shown by introducing another keyword option described in the next para graphs Histogram Definition Defining Histograms LOOKER Histograms 4 30 OodmmEEE The Looker Specifying Bin Widths Counting real values Bin 0 MEER In many cases you aren t concerned with the number of bins in a histogram but rather with the acceptance of each bin In other words you d like to be able to say that each bin will count values in a certain range The following example shows how this can be specified define spec hist first O last 200 binsize 2 It may not seem obvious at first but this histogram actually has 101 bins It will accept values from 0 to 200 inclusive with the first bin counting values from 0 up to but not including 2 There fore 101 bins are set up because the range of 0 to 200 would rep resent 100 intervals and
29. a READER file called He3 r should have an accompanying LOOKER file called He3 1 As mentioned the file itself has only two distinct parts the de fine section and the event section As implied in the follow ing example they must appear in the file in that order but each section may contain any number of lines sample LOOKER description file The assumption is made that the variables adc value 12 and tdc value and events adcread and tdcread were defined in the READER TE Hb db db out define adc_value 12 tdc_value previous data define spectrum 12 histogram from 0 to 2048 define timings histogram from 0 to 1024 512 bins event adcread adc value 102 increment spectrum using adc value event tdcread increment timings using tdc value In this example the define section has only 3 statements the first one tells the LOOKER that an array and a variable will be present in the data stream which were previously saved by the READER The second line defines an array of 12 histograms and the third defines a single histogram a histogram is described un der LOOKER Histograms on page 4 26 Making a LOOKER Description File 4 2 wo The Looker The two event statements introduce instructions which will be performed when those named events are encountered in the data stream As the comments tell us these events are named af ter the original event names in the READER description file Fur thermore the variables
30. adc value and tdc value are not defined in the LOOKER description file because they are READER variables passed through the data stream As we ll see you are allowed to define regular variables in the LOOKER just as you could in the READER The DEFINE Section LUCID allows you to define variables in the LOOKER just as a general programming language would The LOOKER also has ac cess to every variable that was previously saved in the READER The LOOKER automatically makes those variables available to you when it starts or when they are changed during an experi ment LUCID will warn you if you define a new variable for the LOOKER which you ve also de ined and saved from within the READER Remember that only the names of variables that were saved by the READER are automatically defined for you If you don t want LUCID to automatically redefine the READER variables you may redefine them manually using the keywords previous data as shown below This forces the LOOKER to make sure it is working with the correct READER Types of Variables The LOOKER understands all types of variables that the READER knows about except for CAMAC variables When the LOOKER finds a CAMAC variable in the data stream from a READER event it is treated as an integer or array of integers It knows nothing about accessing CAMAC or other hardware On the other hand the LOOKER understands different types of variables which the READER can t handle
31. and y lt 100 Similarly the C language conventions can be used if x 10 amp amp y 100 The LOOKER accepts a mixture of them as well Parentheses are not required around the comparison The second form of the if statement is used to test the acceptabil ity of histogram data As described on page 4 17 any data which is beyond the limits of a histogram is silently ignored no error messages or diagnostics are printed If it is important to know when a variable had an acceptable histogrammed value the if statement can be used Consider this example define tdc 32 int 4 define eres 32 hist from 0 to 1024 event read data if incr eres using tdc print Acceptable value else print Unacceptable value endif The EVENT Section The If then else Statement Making a LOOKER Description File 4 21 The Looker The increment statement itself will be true or false depending upon whether the variable was within the accepted range of the histogram Similarly if a region is used and the value is not with in the accepted bins of the region the condition is considered false In this particular example an array of histograms and val ues are used if any one value of the array is outside the accepted limits of the its histogram the increment statement is considered false The LOOKER also allows checking data acceptance without ac tually incrementing the histogram The following example shows how a value can be passively check
32. area between em Er out in X axis markers fills the screen ae dM Pan right 1 page by the number of bins on screen Zoom in both axes by maximum amount Zoom in on Y axis same as ZOOM Figure 2 31 Keypad Functions in Zoom Mode dated rather the range changes to maintain the same percentage display of the available data Note the use of Page Up and Page Down buttons to scroll across the histogram once zoomed in The number of bins dis played on the X axis is taken as a page and the Page Down button will pan across the histogram a screenfull at a time The user should keep in mind that the mouse buttons have dif ferent meanings in zoom mode by dragging the mouse while the left button is pressed the display will zoom in or out depend ing on the direction of motion Pan Mode Shortcut Key F6 Pan mode refers to simple displacement of the histogram image and is only useful when zoomed into some area of the display Pan up across udin Pan left 1 page by the Pan over to include dede J mnb of bins on screen display start of histogram Pan s x ovet X axis Pan left over X axis z zt Pan to origin of marker area E M nu r Pan right 1 page by the Pan across to include number of bins on screen last bins on both axes om down over Y axis Figure 2 32 Keypad Functions in Pan Mode The View Menu Button The Histogram Windows 2 42 m Using LUCID Figure 2 32 describes t
33. at the menu buttons along the top of the window They can be used to bring up the menus as shown in Figure 2 2 Menu Buttons The LUCID Program 2 5 Lem Using LUCID Start Connect Search and then Connect Change Input File Add Comments Save Histograms Clear Outstanding Commands Build Experiment Build Local variable Peeters Change Directory Rie User Commands Defaults Permissions Save Window Positions Restore Window Positions 3 65 25 Jul 1995 Remove Local Variable Use Hardware Database Writer Playback Sequence Update Visible Run Lucid Messages Reader Stopped at End of File demo_glen virgo Menu Buttons The Experiments Menu Looker Stopped at End of File Writer Offline Figure 2 2 LUCID Menus The Experiments Menu The Experiments menu is shown in Figure 2 3 and is usually the first one to be used in a given session The default selection from it Start Connect must be chosen before most other choices will work It brings up a sub window which allows the user to name the desired experiment Giving the experiment name on the command line performs the same function and most experi enced users make use of this shortcut Experiment names are commonly of the form name computer and are not allowed to contain special symbols or blank
34. cept a list of values as well For example read compress and save timing discard 0 100 180 200 255 This statement might be useful when correcting for hardware problems such as a bad wire in a wire chamber i e when you want to ignore a particular channel A second type of compression is performed for Q Stop modules Again the READER is saving a variable number of entries for the module However only the values are saved The array s in dexing is assumed to go from 0 to the number of values saved As in previous examples the compress statement can be used with read clear wait and so forth Several variables may be specified for any of these operations read and save adc delta e and e prime The EVENT Section Compressing Array Data Making a Reader Description File 3 22 OdEmER The Reader LUCID Operators Each of the three variables will be read and then saved as if three individual read and save statements were used If the dis card feature of data compression is used it will apply to all data saved in that statement White space a comma or the and keyword may be used to sep arate variables which appear in the same statement just as oper ation keywords are separated For example wait read clear and save coincidence scalers adcs and tdcs Calculations and Variables LUCID allows arbitrary calculations to be performed on vari ables The most common form of this is variable expression T
35. comment E Definition of end of run record struct Lucidend struct Lucidheader le_head standard record header char le_name LUCIDNAMESIZE name of person ending run hi Definition of null padding record struct Lucidnull struct Lucidheader ln head standard record header E LUCID Data Record Format Declarations LUCID Data File Format 7 25 Lees Utility Programs and Subroutines LUCID Data Record Format Declarations LUCID Data File Format 7 26 oem 00 0 Appendix The CAMAC Module Database LUCID s knowledge of CAMAC modules comes from informa tion stored in a system database file which doesn t change very often The file is named lucid lib camac modules and al though anyone can look through the file only the system admin istrator is allowed to change it Chapter 2 is entitled Using LUCID and it describes how you can get LUCID to use your own module definition file which is nice when you want to test a new module or use an uncommon feature of an existing module Most users will ask the system administrator to add a definition to the system database since there is no practical limit on the number of modules that LUCID can know about and other ex perimenters will eventually want to use that module The module definition file contains simple descriptions of CA MAC modules which tell LUCID how to access the modules This includes commands for simple reading and writing but also how
36. defined in the LOOKER description file whether the system is processing data or not Select the View button and the View sub window will be displayed Recall that our example only has one histogram variable defined View Data Experiment demo dougGskatter Variables to List Source v All Types v All Sizes v Class spectrum LI s Variable List Name spectrum 12 Type int 4 X axis from 0 to 2048 1024 bins Y asis Display Selected Variables Select the name spectrum on the List then select the button la belled Display Selected Variables A small icon will appear somewhere on your screen which represents the data for that variable You can open the histogram s window by pointing the mouse arrow to the icon and pressing the key labelled Open located on the left side of the keyboard The Open key is actually a toggle so that if the mouse arrow is pointing to the window while you press Open again the window will close A Short Tutorial Session 1 12 OodmHEERE A Tutorial Introduction When the histogram window is open you will see something like this spectrum File v View v Edit v Print v Update zoom Rebin Rebin spectrum 0 X 0 to 2048 0 to 14799 AUTO If the Play button is still selected you might not have an up to date picture of the data since it is still being processed To u
37. directly to the parallel branch graded LAM lines For example the LAM line from station 1 must drive graded LAM line 1 station 2 must drive graded LAM line 2 and so on till station 23 and graded LAM line 23 Graded LAM line 24 is not used Camac Hardware Layout 3 1 The Reader Using Lams LAMS interrupts are often used to determine that an event has taken place If an event is made up of data from several ADC modules for example only one of them should send a LAM Most CAMAC modules can have LAMs disabled either by a CA MAC command or by setting a switch on the module s side pan el If a module s LAM disable or enable is controlled by CAMAC command then LUCID will enable and disable the correct mod ules based on your description file If a switch needs setting you must do that before trying to acquire data It is important to remember that modules such as waveform dig itizers or TDCs may have lengthy conversion times When using them in an experiment you should have some other module generate the event s LAM if possible because the READER can be reading other modules while these slow ones are still convert Making a Reader Description File Using Lams When you ask LUCID to build the programs for your experiment it needs to know exactly what you want It gets some of that in formation from a READER description file which is a text file that can be made with any text editor LUCID only needs to see this f
38. file or program output must contain the exact image of what is to be loaded into the module The EVENT Section Loading Data into a CA Making a Reader Description File 3 27 The Reader Specialized Camac Access LUCID allows you to specify arbitrary CAMAC commands For example define controls sal controller B 0 C 10 N 1 camac controls A 0 11 F 16 data cdat Here it is assumed that cdat is an integer variable In this exam ple the CAMAC operation is a write command and cdat con tains the data to write out Similarly in a read command the READER will put the data that was read into the named variable LUCID knows whether you are reading or writing data by check ing the value of the function code If the code specifies a control function then you can t have the data keyword or a variable name associated with the statement Although the data variable is optional its presence depends upon the type of function code used If writing the expression following the data keyword may also be a constant value The subaddress may contain a range of a codes in which case the READER will repeat the function code with each one Both the function codes and the subaddress code s must be integers as opposed to variables You can also specify a branch crate and station as in camac B 0 C 10 N 1 A 0 11 F 16 data cdat A missing branch specification is taken to be 0 and a missing crate specification is taken to be 1 Use
39. from a module the statement looks like disable is 24 a0 Testing for LAM Status If the user tells LUCID to wait for a module in his READER de scription file then LUCID simply waits for a LAM status bit to become true usually because that particular module will not cause an interrupt when a LAM is ready We often think of LAMs as being the same as interrupts except a LAM is essentially something that can be tested An interrupt is a trigger which is often caused by a LAM The test LAM function is recognized by most modules and the statement should look like testlam is f9 a0 LUCID s READER program will repeatedly issue the command until the OQ response from the module is true meaning that a LAM state exists This is used only for the wait command from the user Clearing LAMs When LUCID has processed an interrupt LAM trigger it will try and clear the interrupt state so it can accept a new one Mod ules usually respond to an F 10 command to clear the LAM state Keep in mind that this is different from the read and clear operation and from the clear operation It simply resets the in terrupt that has just arrived from the module The command is clearlam is f10 a0 5 Notice that individual channels may interrupt in some modules Specifying several A codes will cause LUCID to clear all of them Self Clearing Modules Certain modules will reset their interrupt state automatically af ter the last cha
40. g vega dev nrst1 accesses the no rewind raw scsi tape unit 1 on computer vega from any other computer 7 1 o Utility Programs and Subroutines demolucid demolucid The demolucid script installs in the current directory the neces sary files for running the demo lucid experiment Once the files are installed the xlucid program starts with the demo experi ment See Chapter 1 for more information on running the LU CID demo extractrun This program extracts a single run from the specified input and writes it to the specified output Usage is extractrun o outputfile inputfile run number If outputfile is not specified then the run is written to the stan dard output Inputfile must be a valid LUCID data file and run number should be a valid run number within that file It is recommended that copyrun be used instead as it allows the same functionality with greater flexibility For this very reason extractrun may be removed from the LUCID system in the near future findeot The findeot command positions the tape in the tape drive speci fied as the only command line argument between the two EOF marks at the end of the data If the name specified as a no rewind tape drive more data can then be added to the tape The mt eom command should be equivalent to the findeot com mand However early problems with eom on Exabyte drives at tached to Sun workstations brought about the findeot command Findeot
41. hand side of an assignment statement can use In other words an assignment statement which assigns a value to a single variable must contain an expression which also evaluates down to a sin gle value If one assigns something to an array then the expres sion may yield an array of values If an array of values is used in an assignment to another smaller array then the elements of the larger array which are beyond the smaller s size will not be used Consider the following defini tions define large 20 int 4 define small 9 int 4 The statement small large results in the last 11 elements of the large array being ignored and the first 9 being assigned to the corresponding positions within small The statement large small results in all elements within small being assigned to the corre sponding elements within large and the remaining 11 elements being assigned the value 0 If an array is assigned to a larger array the elements which are missing from the smaller one are assumed to be 0 Using the pre vious variable definitions consider this example large small 2 5 The EVENT Section Calculations and Assign Making a LOOKER Description File 4 15 The Looker Assignment Operators Using Character Strings The first 9 elements of large will be set to 5 more than twice the value of the corresponding elements of small The remaining 11 elements will all be set to 5 since the corresponding element s va
42. having a starting value which is lower than the variable s value In other words values are always rounded down to the next lowest bin s starting val we In reconsidering the earlier example of define spec hist first 0 last 200 binsize 2 bin 1 counts any values from 2 up to but not including 4 but the last bin would actually start at 200 Using the first and last keywords the histograms could actually count certain values which are larger than 200 a value greater than 200 and less than but not including 202 will be counted in the last bin Negative numbers can be used to describe the range of accept able values For example define posn histogram first 1000 last 1000 This defines a histogram which will count values between 1000 and 1000 inclusive thereby containing 2001 bins The only re striction on the limits for acceptable values is that the first val ue is numerically less than the last value LUCID will warn you if the limits are backwards Recall that by default LUCID will allocate 1024 bins in a histo gram if no bin or binsize values are given You may also omit the from or to keywords if you want The default values of limits for the first and last channels are 0 and 1023 respectively The fol lowing example will therefore define a histogram with 1024 bins the first one being 0 and the last being 1023 define el hist Some other restrictions should be mentioned here The binsize value can t be nega
43. i vd i ER HA DR UD REM 3 26 Lucid User s Guide ii ee Contents Writi g to Camac MOGHWIBS oap eco t Ua an UD M On tt tei hs 3 26 Loading Data into a CAMAC Module e rtt 3 27 Specialized Camac ACCESS errinitis 3 28 Camac C tate Operanons cessisse eier unna 3 28 User written Code soa cst suet ones ipte Quibus n aE ERE EEES 3 29 Stopping or Suspending an Experiment cernerent ent 3 30 Words Reserved for the Reader Description 3 31 Chapter4 The Looker How the LOOKER WOPDESsuicncaiteitifesteni inia ntes rust vie bn Ras 4 1 Making a LOOKER Description File so aisi uper tha idit tu raid 4 2 Th DEFINE Section sucreries 4 3 Ty eS OF Variables esirin 4 3 Defining Histograms sussies eras r ear ia Gnd aei pre asaya 4 4 Defining RECON S ais clcuaireecn Doble i ete eost Ros ab Docendi 4 5 Defining PUncbotte s semi eit o m iub linee EO d butt 4 6 Arr ys f Variables 4 6 Groups of Variables gemeint ni cont aliod eae rarae aeaieie ERa 4 7 The EVEN F SOC ELON au etticetin tiekeeibentiintueiket em ni EE ERRARE EES ERA ALARA 4 8 Event Types neperi eE r EEA acd dr vU EE 4 8 Using Variablesauiieusdeeuissridn akuten celeb dia t d Red 4 12 Calculations and Assignment of Values 4 13 Incerementing HistograM S so anion one niu ni ipt bran 4 16 Assigning Histograms To and From Arrays 4 18 Using Besions With IStOR TAINS ssec chen pre etie 4 18 Incrementing Two Dimension
44. making your software and compiling it a stop watch figure will be displayed instead of the regular mouse pointer When the pointer reappears LUCID will have finished making the software Most of the buttons in the main LUCID window will now be made usable and the word Offline will appear in the lower right corner Controlling the Analysis The main buttons in the LUCID window are meant to resemble controls of a simple tape deck or home CD player you re al lowed to rewind or replay the tape which can actually be a real data tape but also a disk file or any other item containing your data Actually all of the windows we ve seen here are used for online analysis and data acquisition too We see that the tape deck has the Stop button selected Also note that the Record button is masked out it s only available when we want to save data somewhere when a WRITER exists If you re ready to start the analysis just select the Play button This tells the READER to start reading and passing data to the LOOKER When the READER reaches the end of the data stream you ll see the Play button pop out and the Stop button will be highlighted again v LUCID Playback Sequence Run 8 Rewind ea Forward Reader Reading Data File ed Writer stepped demo dougGskatter Offline A Short Tutorial Session 1 11 A Tutorial Introduction The next step is to look at the data You re allowed to view any data
45. modules as variables such as delta e in this example Trigger specifications In LUCID a trigger is defined as something which causes an event to occur such as a LAM or some amount of time passing A trigger can also be generated by the occurrence of other events which allows you to implement software events This particular example has a LAM from the delta e variable being a trigger of an event named background User Code specifications LUCID allows the user to specify object files that are to be linked with the generated READER program or C code that is to be included directly in the front end reader Event specifications An event is a description of what to do when a trigger is received This might include what CAMAC modules to read or what calculations or tests to perform Our example has an event named background which will be performed when a delta e LAM occurs Four Sections of a Description File Making a Reader Description File 3 3 The Reader Define CAMAC Variables The four sections of the READER description file must be in or der The definitions must be first followed by the triggers then the user code if any and then the event section Each section may contain any number of lines and the statements within each of these sections are completely free format Notice that each statement begins with a name appropriate to its section define trigger load or program or event LUCID makes no assumptions abo
46. out for type eventno and is copied from buffer and is expected to be nbytes in size Nbytes is the unusual size of number of bytes rounded up to LUCIDALIGN boundary 2 The two bytes subtracted at the end are filled in with the event number The easiest way to calculate this number for a non compressed event is to define the structure for the event either using the generate program page 7 2 or by includ ing the on line reader include file and use the C operator sizeof to get the size of the event then subtract two The file opened for writing by LF open is closed Low Level CAMAC Access These routines allow camac operations to be remotely executed on a vme based cpu To call these routines it is necessary to in clude llucid on the compile command line NetCamacOpen NetCamacPing NetCamacSetDebuglev NetCamac LUCID Functions and Subroutines 7 12 o w Utility Programs and Subroutines NetCamacOpen char name To perform a camac operation first open a communica NetCamacPing tions channel a socket using NetCamacOpen The argument to NetCamacOpen is the host name of the target cpu e g palomar or hubble Only one connection can be operational at any given time Opening a second connection will cause the first to be closed can be used to determine if the connection is still active It is called by NetCamacOpen when the connection is established NetCamac int b int c int n int a int f long data
47. processing would contin ue at the beginning of the next run Requested Enhancements C 2 ew Changes to XLUCID e Add saving of histogram data directly to a user program e Add log scale display on axes e Add bytes seen this run information when forwarding over data e Add a routine callable by user subroutines that accesses data in a lucid data structure e Add backspacing of LUCID tapes to the beginning of a session e Add other information onto histograms for printouts e g date run number e Complete the implementation of the Density Plot win dow e Add the ability to modify histogram data that can be re turned to the looker Implement the Print menu selection Implement the Layout menu selection e Add simple histogram editing resulting in local variables Add more statistical functions for histograms e g first and second moments Requested Enhancements C 3 Requested Enhancements C 4 _ ee Appendix D Histogram Save File Format The original format used by LUCID for saved histograms was extremely simple It was assumed that histograms saved by the LOOKER would be re read by the LOOKER and histograms saved by XLUCID would be re read by XLUCID The new for mat makes the histogram files consistent and provides the user with better information about the histogram saved Overview The histogram file format constists of a header and data body for each histogram allo
48. rather generic title but these choices allow a variety of useful changes to be made The various choices may be made by selecting the box and having the check mark appear or disappear i isplay Options AutoScale Display Bin Numbers Grid Lines ORO RK Statistical Data O Zoom by Percentage Figure 2 20 Display Options The Histogram Windows 2 29 Using LUCID Shortcut Key a Shortcut Key g Shortcut Key b Shortcut Key S Zoom In or Out The View Menu Button AutoScale causes the current histogram display to be au tomatically scaled to allow the data to fit For 1 dimen sional histograms the Y axis is adjusted so the data will fit visibly within the window For 2 dimensional histo grams AutoScale will adjust the settings of the Z axis This option is selected by default Grid Lines may be selected to have dashed lines drawn across the axes at regular intervals This option is selected by default Display Bin Numbers will cause bin numbers rather than actual X or Y axis values to be printed on the display As mentioned under The File Menu Button on page 2 24 this option will also affect the values saved to a file Statistical Data will cause more information to be dis played under the title Regardless of this option the sum of all bins between the markers is printed when markers appear When the option is selected other information such as the maxima and minima values and the mean
49. subsequent NetHo calls frame must be between 1 and 16 inclusive NetHvOnoff int onoff If onoff is 1 the power supply is turned on A value of 0 turns High Voltage Access the power off Any other value results in an error return LUCID Functions and Subroutines 7 13 o Utility Programs and Subroutines NetHvSetRampRate int ramprate Sets the ramp rate for the power supply ramprate must be positive NetHvSetDemand int first int nchan int value Sets the demand to the specified value for the channels starting at first for the number of channels nchan NetHvGetDemand int first int nchan int value Finds the demand for the specified range of channels and places the values in the array value NetHvGetMeasured int first int nchan int value Reads the voltage levels for the specified range of channels and places the result in the array pointed to by value NetHvSetllimit int ilimit Sets the current limit to limit If the limit is negative the neg ative limit is set otherwise the positive limit is set NetHvGetllimit int pos int neg Finds the current limits and places them in pos and neg NetHvClose Removes the connection to the high voltage controller NetHvSetDebuglev int level Sets the debugging level for the net daemon software Reading O format Data The Oread routines provide access to events in files written by the Q software package A C definition for the Q structures used is found in
50. they normally would The name myhistogram can also be used in a limited fashion and refers to all of the histograms within the group You could assign 0 to myhistograms thereby clearing all of the histo grams in the group Histograms and simple variables can be mixed within a group as shown here define allvars group define x y z int 4 define elec specO 12 hist from 0 to 2048 define adc2 12 previous data endgroup Simple assignment of a value to the group allvars will set all variables all histogram elements and the current values of the event data adc2 to zero assuming the zeroing is occurring in an event where adc2 was previously saved The DEFINE Section Groups of Variables Making a LOOKER Description File 4 7 The Looker Some operations cannot be performed on allvars however such as incrementing it as a histogram If all variables within the group are histograms like myhistograms was declared then the increment operation works as if you were repeating the opera tion on each of the histograms in turn The rules for using group names are restricted in the same way as using any other names if an operation would fail on any of the members of the group then the operation will fail on the entire group You are allowed to nest group definitions within other groups For example the following definitions are acceptable define allvars group define x y z int 4 define histonly group define elec specO 12 hist fr
51. trigger constantly integer uninhibit contains integrate using data into wait dataway label when decr lam write decrement last The EVENT Section Words Reserved for the Making a Reader Description File 3 31 O The Reader The EVENT Section Words Reserved for the Making a Reader Description File 3 32 eC 0 Chapter The Looker The LOOKER is the part of LUCID which lets you analyze exper imental data It is a program which LUCID writes for you based on a description of what you want to do It will analyze data on line or off line depending only on where the READER is getting the data You might think of the LOOKER as a program that lets you look at your data In fact it doesn t produce any graphs or pictures at all It only interprets and arranges the data so that other pro grams can produce displays It is called the LOOKER because it looks at your data not because it lets you look at it The LUCID program itself displays the information as the LOOK ER is doing its work This implies that the LOOKER need not be concerned with the type of display screen you re using the type of graphs you want displayed or even the kind of computer you have Another program can be used to draw objects to make up a picture and do the work of drawing scatter plots or contour plots You should read Chapter 2 for more information on the LUCID command and the different ways in which it lets you in teract with your data How the LO
52. use F 1 A 1 to read the status value It also tells LUCID that within that piece of data the sixth and seventh rightmost bits must be zero and one respectively Octal base eight values are most commonly used here since they represent bit positions Decimal values can be used but should be avoided to prevent confusion In the this example the value 0140 repre sents 001100000 in binary The value 0100 represents the binary value of 001000000 The first value tells LUCID which bits are of interest and the second value tells what state those correspond ing bits should be in Thus the sixth rightmost bit must be a zero and the seventh rightmost bit must be a one 1 For more information refer to the last part of the Section entitled Reading Data on page 3 14 Defining the Modules A 3 If this test is performed and the result isn t what you ve told LU CID to expect then an error message will be displayed when you try and start the experiment Read All other capabilities in the database file are associated with CA MAC function codes and sub address values The read keyword tells LUCID what function code to use when reading the module and what sub address codes A codes are acceptable When the user tells LUCID to read this CAMAC variable it will repeatedly use the function code with each of the A codes to do it If a sub script is given for that variable then only that A code will be used In any case the statement m
53. values per dimension LUCID Data Record Format Declarations LUCID Data File Format 7 22 o w Utility Programs and Subroutines Symbol table entry for variables struct Lucidvar char lv_name LUCIDNAMESIZE name of variable u_long lv_asize array size u_char lv_type type of variable u_char lv compress non zero if data compressed u char lv strlen size of string variable u char lv ndims histogram dimensions used union struct Lucidhistinfo lv lh Histogram information struct Lucidcompressinfo lv lci LCMP VARIABLE information lv un define lv nbins lv un lv lh lh nbins 4 define lv lolimit lv un lv lh lh lolimit 4 define lv hilimit lv un lv lh lh hilimit Symbol table entry for events struct Lucidsym struct Lucidheader 1s head standard record header char ls name LUCIDNAMESIZE name of this event u short ls eventtype event I D code u short ls numvars Number of ls var to follow u char ls source Event source reader looker u_char ls internal Internal flags for Lucidread u_char ls pad 2 Pad to 0 mod 4 struct Lucidvar ls var 1 variables saved in this event lE Definition of Reader Looker Writer description record struct Luciddescribe struct Lucidheader 1d head standard record header char ld name LUCIDNAMESIZE user who installed this change char ld text 4 text of description file T Definition of new volume i e tape record The overall volume number i
54. when certain bits were set in a variable The total value of the variable isn t interesting because each individual bit might correspond to something different such as a bank of detectors firing When counting bits instead of variables a 32 bit variable integer 4 would have only 32 counters associated with it Note that several counters might be incremented at once when a variable is checked the value of 5 binary 101 would cause 2 dif ferent counters to count at the same time In any case a collection of counters is called a histogram and LU CID lets you define and use them as you would regular variables A histogram can count the occurrence of any value or count the changes to any bit position within a variable Ideally a histogram will count these different values whenever the variable changes Histogram Definition Using Your Own Sub LOOKER Histograms 4 27 Bins The Looker This action of determining the frequency at which different val ues occur is a fundamental feature of the LOOKER Each counter within a histogram is often referred to as a bin Usu ally a histogram won t contain more than a couple of thousand bins LUCID lets you decide how many bins a histogram will contain and therefore what range of values will be counted by each bin The rule to remember is that all bins in a LUCID histogram have the same width Each individual bin must therefore have the same range of val ues but the entire
55. will disappear when there is complete confidence that the majority of the free world has access to a properly function ing mt eom command generate Generate takes as input a LUCID data stream and produces a C include file from all the LREC SYMBOL records found This in clude file can subsequently be used when writing custom code For an existing on line experiment look for the already created buildname frontend h in the experiment directory Utility Programs 7 2 o w hv1440 Utility Programs and Subroutines hv1440 The hv1440 command communicates with the LeCroy 1440 high voltage mainframe Please refer to the manual page man hv1440 for complete details hv4032 The hv4032 command communicates with the LeCroy 4032 high voltage mainframe Please refer to the manual page man hv4032 for complete details intape An early program to copy a LUCID data tape to disk This is now obsolete and selected for removal Use the copyrun program in stead lucid The lucid command is a terminal based control program for a LUCID experiment This program pre dates xlucid and has not been kept up to date It does allow simple access for monitoring a running experiment Further discussion of this command can be found in The lucid Command on page 7 8 lucid_to_q The lucid_to_q command attempts to translate LUCID data files into Q format data This program has not been used to any great extent and the documen
56. will use the new value These values last until the next Send Region or the next build The Save Region Option Making the menu selection of Save Region brings up a window requesting a directory name and a file name see Figure 2 35 Two check boxes allow saving of only the selected area of the re gion and choosing to append to an existing file or overwrite the existing file At the current time the Selected Area Only option has no effect for 1D histograms The Histogram Windows 2 45 Using LUCID The Print Menu Button oH Save Region To File Region Name gumby Directory users3 wright xlucid File Name Region gumby O Selected rea Only O Append to File Figure 2 35 The Save Region Window The Print Menu Button Printing of histograms is intended to be done through the Print Menu The Print Option Currently not implemented The intention is to bring up a win dow which allows adding titles for printing with the histogram and selection of a printer The Layout Option Currently not implemented This option is a more involved ver sion of the Print option The experimenter will be allowed to se lect which histograms will be printed on a page and be allowed to position and size the histograms on the page All the intended features of the Print option will be available with the Layout op tion The Histogram Windows 2 46 bem 0 Chapter 3 The Reader The READER is that part of LUC
57. with a his togram An array of histograms can be incremented by an array of vari ables each variable within the array will be used to increment the corresponding histogram If a single variable is used then all histograms in the array will be incremented in the same way If they are arrays histograms and corresponding data must have the same array size Incrementing Two Dimensional Histograms Two dimensional histograms allow analysis of two variables at once For a description of two dimensional histograms refer to Defining Two Dimensional Histograms on page 4 36 Incre ment statements are used in the same way but two values must be given define adc 12 previous data define tdc 12 previous data define t vs e hist 64 from O to 1024 64 bins and from 0 to 256 64 bins event newdata incr t vs e using adc tdc Notice that arrays of histograms and data have been used in the same way as they were with one dimensional histograms the same rules apply The bits keyword may also be used in two dimensional histo grams with the same functionality In fact one variable may in crement the histogram with the bit keyword and the other use regular value increments In any case remember to put the word bits immediately after the expression for the corresponding axis See Histogramming Bit Positions on page 4 40 for a further de scription of how this works Using Two Dimensional Regions Two dimensional regions follo
58. words the data which you save with LUCID is completely self contained LUCID can rewrite your experimental software based on nothing more than the saved data The Record button in the Control window tells the READER to acquire data and make it available to the rest of the data stream This is only true when the WRITER uses its own data format it is not possible with Q s data format for example A Conceptual View of LUCID A READER must always be present for LUCID to do an experiment or analyze data but the LOOKER and WRITER need not exist The source of data coming into the READER solely determines wheth er it s on line or not If it s getting data from a tape that data must have been written by a previous LUCID data stream There is no limit as to the number of data streams through which the data may eventually flow but only one data stream is allowed to exist at one time while an experiment is running on line How LUCID Works 6 2 ee Conceptual Overview A more comprehensive picture of LUCID is shown in the follow ing figure The pieces that are required to exist have been high lighted front end processor Reader Shared Memory Looker Writer Data Pipeline Display Storage Storage Figure 6 1 Conceptual Data Flow in LUCID The READER may be used on its own when you re testing elec tronics The LOOKER may be added when you re ready to cali brate your detectors or the el
59. you re just learning about LUCID and don t care about the name just choose one that isn t already listed Finally if you want to create a new experiment to analyze exist ing data you should Find a computer which contains the data file or the tape drive which will read your data file Choose a unique name for your experiment as described above in the on line example One last point about getting started when a number of people are involved with an on line experiment it is sometimes easiest for all involved if an experiment account is set up This gives a common location for all LUCID user files and a common ac count for controlling and monitoring the experiment These are only allowed for on line experiments If this sounds useful to your situation you should see the system administrator for de tails The LUCID Program At this point you should be familiar with the tutorial given in Chapter 1 The OpenWindows graphical user interface was in troduced and at this point we ll assume that is what you re us ing and also that you re familiar with how to use it If not there is a good on line tutorial available as well as documentation If you don t have the OpenWindows interface on your worksta tion or you don t have access to the documentation or on line tu torial see your system administrator before continuing The remainder of this chapter is a reference rather than a tutorial The LUCID Program 2 3 Usin
60. zoom Rebin Rebin You can also press the Delete key to remove a view Note the corners of the window stand out in small wedge shapes These are resize corners and after the mouse pointer is positioned over one of them the entire window can be resized by dragging the mouse while holding down the select leftmost button The next step is to zoom in on a peak in the data The easiest way to do this is to draw markers which delimit an area of inter est Move the mouse arrow over top of the histogram and notice that it changes into the shape of a small plus symbol crosshairs When the left mouse button is clicked a marker will be drawn on the corresponding bin in the histogram You can actually drag the marker around by holding the mouse button down while you move the mouse It might not be apparent but the markers are always drawn to the width of the current bin A Short Tutorial Session 1 14 OodeEEER A Tutorial Introduction To put up a second marker you must click or drag the middle mouse button The second marker must appear on the right side of the first one Notice that a legend is displayed beneath the title telling the X axis values which are currently marked and what the sum of their Y values are Also the sum of counts in the marked area is displayed This total includes counts in the bins which contain the markers To zoom hold down the left meta key whic
61. 000000 integer 2 The DEFINE Section Defining Functions Making a LOOKER Description File 4 6 law The Looker When defining a variable the number of elements within the ar ray must appear between brackets after the name This example allocates room for a million integers each containing two bytes It also implies that you can have an unlimited amount of space There is indeed a limit and you should remember that using too much space can contribute to slowing down the computer which is not desirable if the analysis is on line A LOOKER that takes up too much space simply will not run Array subscripts always start at zero just as in the READER For example the array in the previous example can have subscripts from 0 to 999999 All variables are guaranteed to have a zero val ue after an experiment is rebuilt but they are not zeroed out au tomatically with every new experimental run Groups of Variables LOOKER description files may come to contain hundreds of vari ables as an experiment grows and it may become difficult to deal with such a large number To accommodate such experi ments the LOOKER recognizes groups of variables which can be treated as a common unit Such an aggregate of variables is easy to define as shown in this example define myhistograms group define tdcO 8 hist from 0 to 128 define tdc1 8 hist from 0 to 128 define tdc2 8 hist from 0 to 128 endif The histograms within the group can be used as
62. 2 Start run 9 Start viewer program 16 Offline Only 3 Stop run 10 Enter comment record 17 Offline Only 4 Suspend run 11 Create user list 18 Offline Only 5 Resume run 12 Set debug level 19 Offline Only 6 Change buffer size 13 Terminate 20 Modify writer list 7 Show status 14 Offline Only Command Experiment name demo glenGvirgo No READER currently active No LOOKER currently active No WRITER currently active vme console Figure 7 1 LUCID Menu The lucid Command 7 8 o w Utility Programs and Subroutines vme console Dn gg m o 10 LT 12 13 14 15 16 17 18 19 Builds an experiment prompting for build name to find source files and whether to build a reader and or a looker ON LINE After an experiment has been built this starts the acquisition ON LINE This stops the acquisition for the current run ON LINE The current run is paused ON LINE A paused current run is resumed ON LINE The buffer size used in communicating with the front end processor is changed It should not be necessary to use this as a save data statement in a user triggered reader command will flush the current buffer no matter the contents Prints the current status of the READER LOOKER and WRITER Prints the list of user defined commands in the reader and looker Starts the Sunview oriented viewer Because of conflict between this and xlucid this featu
63. 2 6 The Build Writer Window Apply Changes informs the WRITER of the net changes to the file list Store places the current list of output devices and their settings in your Xdefaults file This is automatically re read and used as the default settings the next time Xlucid is started from the same account CANCEL quits the window without applying any chang es The next time the window is started it will be back to the state that the WRITER is currently in The Event List shows the list of all events and the status of the event for the current output device Events can be saved com pletely only at the output device speed or never The Control Menu After software has been generated the user will probably start a run by selecting the Play or Record button All of the main con trol buttons are described in the next section but other less com 1 At the time of writing this section the Event List feature had not been implemented in the WRITER Menu Buttons The Control Menu The LUCID Program 2 12 m Using LUCID 9 mon features are found as options under the Control menu as shown in Figure 2 7 Control Pre defined Commands Commands from your LOOKER Description scalers counters channels status Save Histograms User Commands i Change Input File Suznhn Add Comments Experiment Variables to List Source Types Sizes Class 4
64. 29 User Code Section 3 11 Using Variables in Events 3 12 Variable Arrays 3 24 Variable Calculations 3 23 wait keyword 3 20 print 3 24 write keyword 3 26 program 3 11 Reading CAMAC variables 3 17 read 3 14 3 17 Reading File Data Reader 3 15 reject 3 26 Reading Xlucid Histograms 2 26 Index 4 ee Index readsara 7 7 Regions 1D in Xlucid 2 37 2D definitions in Looker 4 39 2D in Xlucid 2 38 Arrays in Looker 4 35 Defining in Looker 4 33 Definition in Looker 4 34 Get Region Xlucid 2 44 in Looker 4 33 Looker definition 4 33 4 34 Looker restrictions 4 36 Save Region Xlucid 2 45 Send Region Xlucid 2 45 Rejecting Events in Reader 3 26 Rewind button 2 21 S Saving CAMAC Data 3 18 Saving Xlucid Histograms 2 24 Software triggers 3 25 Software Triggers Reader 3 9 Starting an experiment 1 4 Status buttons 2 22 Stop button 2 21 T The Pause Button 2 21 The Rewind Button 2 21 The Stop Button 2 21 Trigger definition 2 2 Examples Reader 3 11 Trigger Section Reader 3 8 Trigger statement 3 9 3 25 Trigger types 3 8 Triggers Continuous Reader 3 10 Polling 3 10 Software Triggers in Reader 3 26 Triggers from Keyboard Reader 3 10 Triggers in Software Reader 3 9 Tutorial Session 1 2 U User Code in Reader 3 11 3 29 User Permissions 2 18 User Permissions window 2 16 V Variable types Looker 4 3 Variables Array assignment in Looker 4 15 Arrays in Looker 4 6 Assignment in Looker 4 13 Assignment operators in
65. 4 18 Arrays of regions may be defined just as arrays of histograms A region cannot be used by itself however and an array of regions is usually accompanied by an array of histograms Assume the following definitions define tspectrum 12 histogram define tvalid 12 region 1 Remember that bin numbers start at zero Histogram Regions Defining Histograms LOOKER Histograms 4 35 The Looker Region Restrictions Regions and histograms both have 1024 bins by default If we as sume that an array of 12 tac values exist we could increment all 12 histograms using the 12 corresponding tdc values but only if they are acceptable to the corresponding 12 regions This rather elaborate process is described in detail later but keep in mind that only one statement is needed to do all the work increment tvalidGtspectrum using tde Note the alternate method of naming a region and histogram pair using tvalidetspectrum is the same as using tspectrum region tvalid Basically regions make it very easy to determine if variables con tain acceptable values and can therefore let you cut unaccept able events from the data stream or increment histograms based on ranges of values Some restrictions should be mentioned at this point In particu lar a region must have the same number of bins as any histo gram with which it will be used The notion of binsize has no meaning to a region since it may be used to mask out histogram bins of a
66. 53002AAL define LREC READER 0x553003AAL define LREC LOOKER 0x553004AAL define LREC WRITER 0x553005AAL define LREC NEWVOLUME 0x553006AAL define LREC NEWSESSION 0x553007AAL define LREC BEGINRUN 0x553008AAL define LREC DATA 0x553009AAL define LREC ENDRUN 0x55300AAAL define LREC COMMENT 0x55300BAAL define LREC NOTHING 0x55300CAAL define LREC ERROR 0x55300DAAL Event source values define LSRC UNKNOWN 0 define LSRC READER 1 define LSRC LOOKER 2 The following structure is written near the beginning of a vol ume to allow software to determine the architecture of the writ ing machine It includes a constant in single and double precision format define LUCID FLOATCHECK 1 25 struct Lucidformat struct Lucidheader 1f head standard record header double 1f double double precision format of source float lf float floating format of source js Event definitions consist of the following Lucidsym structure which will contain zero or more variable definitions The order of the definitions dictates their order within the data record Zero variables indicates no data saved Different types of variables that can be saved define LVAR UNKNOWN define LVAR BOOL define LVAR CHAR define LVAR SHORT define LVAR LONG define LVAR FLOAT define LVAR DOUBLE define LVAR TYPEMASK ONUFWNEH O 77 mask for basic type LUCID Data Record Format Declarations LUCID Data File Format 7 18 o w Utility Programs and S
67. 6 6 _ e Chapter Utility Programs and Subroutines Utility Programs addclient There are a number of programs available for working with LU CID data A few of these programs are system administrator programs the rest are general use In the following list the pro grams for system administrators are flagged with system addclient system The addclient program sets up the system files on the current computer so LUCID can run This program can only be run by the super user No arguments are required and there are no harmful effects caused by running the command on a computer that had previously been set up copyrun Copyrun copies runs surprised from a LUCID data source to a destination Which runs should be copied are determined by the command line arguments The command synopsis is copyrun o outputfile i inputfile s session run number s session run number The inputfile and the outputfile are LUCID data files either may be a tape drive The session is the session number as shown by lu cidview The run numbers listed after a session number are the runs that are to by copied Runs from multiple sessions can be copied by adding additional session and run number specifica tions Copyrun supports the idea of remote tape drives Although not a regular feature of SunOS remote tape drives can be accessed by using hostname dev tape to access the device tape on ma chine hostname e
68. CID we re usually talk ing about the entire system shown in the previous diagram and not just about the LUCID program through which we interact with the data stream Using the Help Facility Before continuing with the experiment we should consider how LUCID will be able to help you after you lose this manual LUCID has a very simple Help facility Just point the mouse to something within any LUCID window and press the Help key on the keyboard 2 In LUCID the term on line refers to the act of acquiring new data and off line refers to processing data which was acquired in a previous LUCID data stream 3 Typically the Help key is located on the far left of the keyboard at the bottom Some older key boards don t have one you ll have to press the F1 key instead A Short Tutorial Session 1 6 Oo mHEERE A Tutorial Introduction For instance position the mouse pointer over the Experiments button Press the Help key and you ll get some information about it lucid Help Selecting this button use the left mouse button will allow you access or create a new experiment Use the right mouse button to get a menu of other options You ll be able to change directories or search for different experiments on Lab computers The important thing to remember is that pressing the Help key will give you help with any LUCID related object that you re pointing to This is usually true even with very small items such
69. Correspondingly a stop or suspend command will cause LUCID to finish whatever it s doing and issue those triggers last Sometimes housekeeping duties are required to be performed before a run is finished possibly after a certain amount of data has been collected LUCID allows you to specify that a certain number of processed events will act as a trigger For instance you could use the following statement trigger scaler every 1000 delta e This allows you to read the scalers after processing one thousand real data events ideally to guard against counters overflowing Note that the trigger is not really put into effect until after the one thousandth delta e event has been successfully pro cessed An event can trigger another event See Triggering Another Event on page 3 26 for a description of how this is done The TRIGGER Section Software Triggers Making a Reader Description File 3 9 The Reader Keyboard Triggers Module Ready Triggers Continuous Triggers Polling LUCID allows you to attach a command on your terminal to a trigger This is often used for testing various parts of your equip ment before the experiment actually starts For example trigger reset every reset command This attaches a trigger to an event called reset which will be performed whenever you type the command reset to the LUCID program Any sequence of keystrokes can be used including control characters Notice that the command nam
70. Currently temporary variables can only be used to read saved histogram files See The File Menu Button on page 2 24 or for temporary Menu Buttons The Build Menu The LUCID Program 2 10 w Using LUCID region variables when building a new region description for a histogram See Region Mode on page 2 37 The option to Remove Local Variable simply frees up the space associated with temporary local variables that have been created by any means The Use Hardware Database option can be used when acquiring data on line when a new READER is built to specify different CAMAC features than normal Typically a small system database is used to describe various features of CAMAC equipment If de sired one can change these characteristics by supplying their own database file The format of the file is described in Appen dix A near the end of this manual The name of the temporary database file should be set before the software is generated with the Make button in the Build Experiment sub window For convenience the modules specified in the system database are always read after the user s database LUCID allows the us er s definitions to take priority over existing ones The Build Writer window gives the experimenter direct control over the output devices being used in an experiment see Figure 2 6 The first part of this window lists all the output devices se lected for output with this experiment the first time this window is starte
71. ID State Triggers Software Triggers The trigger statement always starts with the keyword trigger The next portion tells the READER to perform some action called an electron arm event when the triggers arrive Details of thatelectron armevent would be described in the event sec tion of the file This particular example associates two different triggers with the event specifically a LAM from the CAMAC module named delta e anda time interval of 3 seconds When either of these things occur the trigger is performed LUCID al lows several triggers to cause an event If both things happen at the same time LUCID will see both of them nothing will be lost because triggers remain pending until serviced In many experiments the user will want to perform some kind of housekeeping duties at the end of a run such as reading scalers or disabling high voltage equipment LUCID generates a trigger automatically when you give a command to start or stop a run For instance it is acceptable to say trigger scaler every endrun There are four such internal trigger names called beginrun endrun suspendrun and resumerun As shown in the exam ple they are associated with events in the same way as a CAMAC LAM or a time unit would be These triggers will be issued when a start stop suspend or resume command is given to the LUCID program When triggering from a start or resume com mand LUCID will produce the trigger before doing anything else
72. ID which acquires data It is a program that LUCID writes for you based on a description of your experiment Having the programs written automatically from an experiment description relieves you of the task of writ ing all the low level acquisition routines yourself The first section in this chapter deals with the way the data ac quisition hardware must be set up The remainder of the chapter explains the details of writing an experiment description that LUCID will turn into the data acquisition programs Camac Hardware Layout CAMAC Requirements CAMAC Requirements If you re not concerned with CAMAC at this point you can skip ahead to the next section entitled Making a Reader Description File on page 3 2 LUCID is a multiprocessor data acquisition system The hard ware consists of one or more front end processors which acquire data from CAMAC FASTBUS and VME modules The front end processors forward the acquired data via an Ethernet link to an acquisition processor The acquisition processor writes the data to some storage device and makes the data available for further analysis and display CAMAC modules are connected to the front end processors by a parallel branch highway There can be up to eight parallel branches each with up to seven CAMAC crates Crate controllers in crates with modules that will be used to generate LAM s must be equipped with a LAM grader card This card must be wired to connect the crate LAM lines
73. OKER Works The LOOKER extracts data from the LUCID data stream after the READER acquires and saves it The original source of raw data could be CAMAC or a tape or disk file and the READER produces a data stream which is independent of that source The LOOKER simply watches the data stream and picks out the events that are important to it The LOOKER description file consists of two parts a de ine sec tion to create new variables and an event section which con tains the instructions that will be performed when specified events are recognized The Looker The LOOKER cannot define new event types and or save new events in the data stream It can when running off line filter out unwanted events or modify values in existing events for saving a reduced data set The WRITER is also capable of eliminating certain event types from a data stream as described in Chapter 5 Making a LOOKER Description File When you ask LUCID to build your experiment it produces a LOOKER based on a description of what you want to analyze Your description must be typed into a file very similar in format to the READER description file which was discussed on page 3 2 For instance the same rules for comments and free format statements apply The most notable difference is in the name of the description file itself It should have the same prefix as the READER description file but must end with a I to show that it s a LOOKER file Thus
74. R For instance entire arrays of variables can be used as single entities by simply leaving out the subscript The following segment would clear the first element within the named array define junk 100 int 4 beginrun junk 0 0 The following code segment would clear the entire array at once define junk 100 int 4 beginrun junk 0 The LOOKER has a few built in variables allowing easy access to status information The variables and the information they make available are The EVENT Section Using Variables Making a LOOKER Description File 4 12 m The Looker LUCIDrun current run number LUCIDonline non zero if performing on line acquisition LUCIDbytes number of data bytes seen this run LUCIDtime off line timestamp of data record LUCIDsequence off line sequence number of data record The LUCIDtime and LUCIDsequence variables only work when running off line The timestamp is a standard UNIX clock value as returned by time The number of bytes is a real 4 value and the other variables are integer values These variables can be used anywhere Calculations and Assignment of Values Arbitrary expressions are allowed in many places in LOOKER statements the most common being the assignment statement define new e 20 integer 4 define laststep 20 real 4 define spect hist event energy new e 0 10 laststep 0 130 2 2 2 This example shows several features mixed mode arithmetic is allo
75. Run 3 The number of bytes processed so far in the current session The current session includes all runs since the last Build operation 4 The approximate rate at which data is being acquired analyzed or recorded Figure 2 14 shows some examples having different status labels being displayed Run 8 Reader 1 019 8 Kbytes this Run Reader 3 059 3 Kbytes this Session Loo ker Writer 3 059 3 Kbytes this Session Looker Stopped at End of File Writer Figure 2 14 Different Status Labels After a status button has been selected three times it will cycle the label around to the initial status message The process of ap proximating the data rate is done every 5 seconds roughly so initially it might say Inadequate Statistics if it hasn t had enough time to make the calculation The Histogram Windows Status Buttons The most frequently used feature of the LUCID program itself is the display of histograms Histograms are described in detail un der LOOKER Histograms on page 4 26 and the user is encour aged to read that section if the topic is not familiar to them Histograms are displayed by selecting variable names from the View Data sub window as described under The View Menu on page 2 14 Each histogram is placed in a separate window and can therefore be individually resized or modified They can also be closed to an iconic state Each histogram window has a set of four menu buttons which al
76. a and perform the most basic operations as described later in this Chapter The Search and then Connect option causes a list to be generat ed of all LUCID computers in the Lab A list of all experiments on all the local LUCID computers can be found by selecting Search All The list of experiments on certain hosts can be obtained by selecting the desired hosts in the Hosts list then pressing Search Select If there is a specific experiment name to search for it can be entered in the Experiment field and the computers searched will only look for that particular experiment The Change Directory option causes a sub window to appear which allows different working directories to be used This op tion can only be used before the program is connected to an ex periment This is most useful when different directories in a computer account are used for different experiments and one needs to switch between them Remember that simply changing to the appropriate directory before starting the lucid program Menu Buttons The Experiments Menu The LUCID Program 2 7 Using LUCID Building the Software On line or Off line Mode and Using Data Files Menu Buttons The Build Menu achieves the same result at least for the first experiment connec tion The Disconnect option allows the user to disconnect from the READER LOOKER and or WRITER thereby allowing connec tion to another experiment The user is given the option of shut ting d
77. a program which can read settings of various mechanical components of a detector At your terminal you might normally type home wilson getvalue B302 The EVENT Section Reading Data Making a Reader Description File 3 16 o d The Reader When to use read Reading CAMAC Variables to find the setting of some component named B3 02 From within your LUCID program you can type read energy from program home wilson getvalue B302 The program can be any executable program including a com mand script Furthermore you can specify input redirection and use pipes between the quotes to start a pipeline of several pro grams just as you would from a terminal The programs are al lowed to access other computers or do anything that an ordinary program can do but it must produce output in the way mentioned above Also when LUCID runs the program s they aren t allowed to have any input from the user Parameters can only be passed to program s on the command line which starts them So far the read statement sounds like a simple but reasonable statement for a programming language to use However when we consider that LUCID is meant to acquire experimental data at high rates it seems out of place Actually the intended purpose is to allow certain parameters to be input at the beginning of a run for example This might include data which was not easily accessible on line such as a spectrometer angle or the mixture of gas in a wir
78. ady running experiments How LUCID Works 6 4 B Conceptual Overview prefix l the LOOKER program file which was copied from the user s directory when the experiment was built prefix lsymtab the looker symbol table created when the exper iment looker was built prefix r the reader program file which was copied from the user s directory when the experiment reader was built prefix ysymtab the reader symbol table created when the exper iment reader was built prefix offlooker the executable version of the off line analysis looker prefix offlooker c the C source generated automatically from the prefix l file when the user specified off line analy sis prefix offlooker o an intermediate file created when the prefix offlooker is generated logfile where many of the LUCID messages go When an obscure message is displayed elsewhere it s a good idea to look here for more information runno the next run number to be used with data acqui sition for this experiment seqno a summary of the number of LUCID records of each record type prefix frontend c the C source generated automatically from the prefix r file when the user specified on line acqui sition from an VME front end processor prefix rontend h the C source generated automatically from the prefix r file when the user specified on line acqui sition from a VME front end processor This file includes structure definitions fo
79. al Histograms 4 20 Using Two Dimensional BeptOns au asuscern coii V nente ink 4 20 The It the els Statement soe Capi Ribera cbe e dee 4 21 Th Repeat Statement eese cobran PR ER Feci bo d SD bci oer 4 23 Printing Values from the LOOKER 5 rhet tres 4 24 Loading Histograms from Files or Programs 4 24 setting Looker Variables Interactively aeree 4 25 Using Your Own Subroutines siirius bere redis 4 25 LOOKER SOS PANG NM 4 26 Histogram Denn ton ancudananncanaincanntnidunmeiiint 4 27 Defining EBSEOSEAEDS esseere vastinbieanicomeldendnnbaceleenecntontealy 4 28 Ini pec FAM calo P 4 33 Defining Two Dimensional Histograms sess 4 36 Defining Two Dimensional Regions see 4 39 EHistosramming Dit Positions eccessi 4 40 Lucid User s Guide iii ee Contents LOOKER IR Cy WOES HH 4 41 Chapter5 The Writer Saving Event LUAU iudiciis hlietiiald isi d di cake dades 5 2 Critical Data Destinations cccccccccccccesssesssceessscesssecessseeessesesssees 5 2 lois P 5 3 EXAMP m 5 3 Chapter6 Conceptual Overview How LUCID Works eeeeeeeene nennen 6 1 A Conceptual View of EUCID esesesesissaaitzinixk ri t th ak ra rapa R tinte 6 2 Data Communications in LUCID eee 6 3 The LUCID Experiment Directory siriar
80. an contain but for now lets just look briefly at this small example define demo adc 12 previous data int 2 define spectrum 12 histogram from 0 to 2048 event demo event increment spectrum using demo adc command reset spectrum 0 The first line tells LUCID that when the data was first ac quired an array of twelve integers was saved This state ment isn t required because LUCID can find the array in the data file It s a good idea to include it since it allows LUCID to check that it s got the right data file In our case this array represents the twelve inputs on a peak sensing ADC This tells LUCID that the LOOKER will need an array of twelve histograms each one accepting values from 0 to 2048 By default histograms contain 1024 bins This line instructs the LOOKER to perform the next two lines whenever an event named demo event is en countered in the data stream The appropriate bin in each of the twelve histograms is in cremented by using the values from each of the corre sponding twelve ADC elements Next LUCID is told to perform some instructions when ever the experimenter gives a reset command Specifically all bins of every histogram within the spec trum array will be set to zero A Short Tutorial Session 1 9 o A Tutorial Introduction These few lines are all that s needed to make LUCID write anal ysis software to histogram twelve ADC inputs It really is quite simple W
81. are displayed Zoom By Percentages is used in conjunction with Au toScale and Zoom mode See Zoom Mode on page 2 41 for a description of how this option affects zooming The Stepsize option allows the user to enter a scaling factor When a request is made to move a marker or adjust the viewing limits this step size will be used By default the step is 1 so that markers are moved 1 bin at a time and so forth The step size function is discussed again on page 2 33 The Subscript field allows the user to change the subscript of a displayed histogram array This field will be inactive if the asso ciated histogram is not an array The next choices in the sub window allow the user to change the limits of the viewed data and to enlarge or shrink the viewed ar ea Figure 2 21 shows some typical values in the various fields X Limits From 0 To 2048 Set From Markers Rebin Factor 1 aIv Figure 2 21 Adjusting Viewable Limits The Histogram Windows 2 30 w Using LUCID The X Limits may be set to any values within the extents of the histogram and the area delimited by these values will be re drawn to fill the entire window Bin values may also be set or ad justed by selecting the up down arrow buttons again the area enclosed between these values will be redrawn to fill the win dow These values apply only to the current sub view if more than 1 exists If a value is entered for X or Y axis values it will be rounded
82. arrays are indexed starting at zero so that this particular example is referring to the fourth element of the array called phi An index of a subscript may be any arbitrary ex pression that evaluates to an integer Arrays may be named without subscripts allowing operations to be carried out on every element For instance if we assume that the array theta contains fifty integers the statement theta 100 will set every element of the array theta to be one hundred Because most CAMAC modules have several inputs CAMAC variables represent arrays having subscripts which correspond to their sub address A codes All CAMAC variables are identical to arrays in that they are indexed starting at zero For example let s assume that the definition of a CAMAC variable adc indi cates that only sub address codes five through eleven of the CA MAC module are to be accessed define adc lrs2259B N 4 A 5 11 The EVENT Section Using Variables Making a Reader Description File 3 13 The Reader This means that adc can be used as an array of size seven with the sixth module input corresponding to adc 0 Module has 12 inputs numbered 0 through 11 Array has 7 elements membered 0 through 6 corresponding to module inputs numbered 5 through 11 adc 0 adc 6 Similarly the twelfth input numbered 11 corresponds to the last array element adc 6 If a particular CAMAC module has only o
83. as many experiments as you want An experiment can be thought of as a data stream as described in the first chapter and LUCID allows you to scan over the data as many times as you want In fact you can have several differ ent experiment names set up to look at the same data and sever al build names within an experiment if you wish LUCID maintains a list of trusted users for each experiment The person which initially creates the experiment is considered to be its administrator and is responsible for keeping the user list up to date See User Permissions on page 2 18 If you want to access an experiment which has already been set up and may be currently running make sure you know the fol lowing things The name of the computer on which the experiment was created Getting Started 2 2 m Using LUCID e The original name of the experiment as it was registered with LUCID e That the administrator of the experiment the person who first created it has added your name to the list of users If you want to create a new experiment to acquire data make sure you e Get the name of the computer which is connected to your electronics CAMAC equipment for example and make sure that the system will let you log in your login name and password will be the same on all computers Choose a unique name for your experiment When you first start up LUCID can give you a list of all existing ex periments on all Lab computers If
84. as titles or small buttons if you re not sure what something does just press the Help key Building the Experiment Now that we re successfully connected to the demo name experi ment the next step is to Build the software LUCID actually writes software for you based on a simple description of your experiment This is true if you re acquiring new or analyzing ex isting data The demolucid program has created two important files under your computer account before starting LUCID The first is named demodata and contains some data which was acquired beforehand with LUCID The second file is named demo l and contains a simple description of how one might analyze the data We ll see that LUCID can write the analysis software very quick ly A Short Tutorial Session 1 7 A Tutorial Introduction Select the Build button in the LUCID window The Build sub window will be displayed The first step is to select Offline then enter the name of the data file demodata as shown in this ex ample Build Experiment demo_doug skatter CAMAC Device my Online Data File Description Files dem Use READER Edit demor Use LOOKER Edit demo Next you ll have to tell LUCID that you want a LOOKER but not a WRITER simply select the looker button if it isn t already highlighted Keep in mind that the READER always exists so you won t be able to de select it The title above
85. ask which allows op erations to be performed on certain histogram bins but not on others Probably the most useful feature of regions is that they may be redefined dynamically as the experiment runs For more information on regions refer to Histogram Regions on page 4 33 Consider the following example define adc previous data define espect hist define rspect region event scint increment espect region rspect us ing adc This increments espect using the value in adc The rspect region has not been given a value in its definition and is therefore set to the entire range of the histogram Recall that by default histo grams and regions both have 1024 bins Consider if we were to change the region s definition define rspect region contains 0 to 300 400 to 800 The EVENT Section Assigning Histograms To Making a LOOKER Description File 4 18 m The Looker The histogram bins would still be incremented but only if the value was greater than 0 and less than 300 or greater than 400 but less than 800 If you want to use the bins of a histogram which are outside those accepted by a region the LOOKER allows you to use the notre gion keyword The increment statement from the previous ex ample could be changed to increment espect notregion rspect using adc This will increment bins which are outside of the rspect region s bins The LOOKER also understands the at symbol as a region specifier For instance the previou
86. at Declarations Output is always fixed length blocks LUCIDBLKSIZE Within these blocks are one or more records or parts of records Routines are provided to read and return the records to the calling routine The calling routine never sees the underlying block structure define LUCIDBLKSIZE 32 1024 Alignment restrictions for specified types must be powers of 2 since LUCIDALIGN_XXxX 1 is used as a bit mask define LUCIDALIGN RECORD 4 define LUCIDALIGN SHORT 2 define LUCIDALIGN LONG 4 define LUCIDALIGN FLOAT 4 define LUCIDALIGN DOUBLE 8 Number of bytes in variable names user names etc All names are null terminated so the maximum number of characters in a name is LUCIDNAMESIZE 1 define LUCIDNAMESIZE 32 LUCID Data Record Format Declarations LUCID Data File Format 7 15 o Utility Programs and Subroutines LREC_FORMAT LREC_SYMBOL LREC_READER LREC_LOOKER LREC_WRITER This header is contained within every record structure given be low The magic number ALWAYS comes first in EVERY record and will make the byte ordering of the writing machine apparent to the reader struct Lucidheader u long lh magic identifies type of record u long 1h time UNIX style time stamp u long lh nbytes number of bytes AFTER header u long lh sequence seq number of this record type hi The following is a list of values for the lIh magic field in the header structure This identifies the type of record that
87. ayed in the lower left corner of the histogram window currently being point ed to The mode remains in effect for all histograms until another mode is selected The user may also specify a temporary mode change so that only the current keystroke will be performed in the desired mode For example the user will most often work in marker mode but by performing an extra keystroke zoom mode can be enabled temporarily to perform one command Temporary mode changes will be discussed near the end of this section Some keys on the numeric keypad keep the same meaning re gardless of mode we ll consider them first Figure 2 23 shows the layout of the keypad The insert and delete keys at the bottom of the keypad will insert and delete sub views from histograms respectively The func tion of these buttons doesn t change from one mode to another they can always be used to change the number of sub views The mouse pointer must be pointing to the histogram sub view which is to be duplicated or deleted Marker Mode This is the most commonly used operating mode and is the de fault when the program starts The Histogram Windows 2 34 m gt a E a m me o Figure 2 23 Layout of the Numeric Keypad N T o Cie NR 7 d o VS 8 6 CA X A N la ol Al Ta A d Y Markers may be placed by simply selecting a bin in the histogram display the first marker must be placed by clicking the
88. b windows from the Properties Menu The Properties and Defaults options allow the user to tailor the system in a personal way The subwindow for the Defaults se lection allows setting values that only take effect when the LU CID program is started Menu Buttons The Properties Menu Experiment Name names the experiment to which LU CID will automatically connect when started Database File names the file which contains CAMAC module information when building This will be the de fault when starting but the user can change it Camac Processor Device fills in the appropriate field in the Build sub window Default Experiment Directory names the directory to which LUCID will change when it starts but before it con nects to an experiment Base Name for Description Files tells the system what the description files should start with when building soft ware File Containing Offline Data names the default data file to be used when building software The LUCID Program 2 16 pes Using LUCID The fields in the Properties Adjust window are effective imme diately when changed Menu Buttons The Properties Menu Maximum Number of Messages to Keep determines the size of the memory for the View Messages sub window It can be set to a very large number if the user wants to see lots of output messages Tenths of Seconds for double clicking can be set to a large value for users who can t or don t want to double click the mouse button
89. can arise where the READER description language is inadequate for a particular job Rather than giving the READ ER a full blown general purpose programming language it makes more sense to allow access to existing general purpose languages The program statement does just that by allowing user C code to be included at the current point in the event define data average compress inputs program home wright Lucid reader calc c The above example tells the reader that the C code in the file home wright Lucid reader calc c will place the values of av erage and inputs in the event record the latter as a compressed array The variables average and inputs must have been de fined earlier in the READER There are cases where it is more convenient to include the C code directly in the reader rather than in a separate file This is done simply by the sequence program divide average by 2 dp gt Vaverage 2 The EVENT Section User written Code Making a Reader Description File 3 29 The Reader To make coding easier the following definitions and declara tions are established by the READER variable names are Vname e g Vaverage above variables in the event record are dp Vname e g dp gt Vaverage each CAMAC module defined has the C defines Cmodule branch Cmodule crate Cmodule station Cmodule lowaddress Cmodule highaddress e g Cdoubles branch Saving compressed variables takes a special effort The l
90. ch register wgr200 longword read 0 a0 LeCroy Model 2415 High voltage power supply 2415 LeCroy 2415 shortword Block of 8 TDC modules associated with 4 SAL D flip flop modules See tagger template for an example of how this module is used module Block of 8 LeCroy Model 2228A 2229 Octal TDC 2228A Block 2229 Block LeCroy 2228A Block LeCroy 2229 Block timeout 100 shortword readsize 64 qstop channel length 16 offset 0 value length 16 offset 16 autoclear Sample CAMAC Module Database A 14 ec o Appendix Example Description Files The example shown here is the reader and looker description file used with the demodata data file discussed in Chapter 1 acquire data from Le Croy 2249A ADC define demo adc 2249A C 1 N 12 A 0 11 define scaler 3615 N 11 A 5 define total integer 4 define gates integer 4 trigger pulse every demo adc lam trigger quit every 40 seconds trigger count every 30 seconds trigger start every beginrun event pulse read clear and save demo adc Save demo adc gates gates 1 save gates event count read and clear scaler save scaler total total scaler save total event quit read and clear scaler save scaler total total scaler save total endrun event start total gates oll Figure 7 1 demodata READER File B 1 define spectrum 12 histogram from 0 to 2048 define increases number ga
91. clear is 9 a0 disable is 24 a0 enable is 26 a0 module LeCroy 2248 8 Channel Differential ADC 2248 LeCroy 2248 timeout 106 autoclear readsize 8 shortword read is f0 a0 7 readclear f2 a0 7 enable 26 a0 disable 24 a0 testlam 8 a0 clearlam 10 a0 module LeCroy Model 2249A SG 12 channel ADC 2249A 2249SG LeCroy 2249A LeCroy 2249SG timeout 60 autoclear readsize 12 shortword read is 0 a0 11 readclear is f2 a0 11 test is 8 a0 11 clear is 9 a0 disable is 24 a0 enable is 26 a0 module LeCroy Model 2249W 2259B 12 channel ADC 2249W 2259B LeCroy 2249W LeCroy 2259B timeout 106 autoclear readsize 12 shortword read is f0 a0 11 readclear is f2 a0 11 Sample CAMAC Module Database A 10 eee test is f8 a0 11 clear is 9 a0 disable is 24 a0 enable is 26 a0 module LeCroy Model 2256A 20 MHz Waveform Digitizers 2256A LeCroy 2256A readsize 128 hold shortword read is f2 a0 test is f8 a0 clear is f9 a0 disable is f24 a0 enable is f26 a0 NOTE the 2277 can be used to read out multiple values for a single channel This is not directly usable in the looker Instead the user must decode this manually To use automatic decoding change the readsize line to readsize 512 qstop channel length 5 offset 17 value length 17 offset 0 module LeCroy Model 2277 32 Channel TDC 22579 LeCroy 2277 readsize 512 qstop channel length 0 value length 22
92. crate is initialized nethv The nethv program communicates with a LeCroy 2132 across the network The command takes the form nethv i ccontroller pprompt file The controller should be specified as bNcNnNtN host Host must be a valid acquisition computer name N is a number rep resenting the branch crate station and type of high voltage mainframe either 1440 or 4032 Nethv commands are read from the files named on the command line The i option tells nethv to read from the standard input after reading commands from the named files help Produce a description of the rest of the commands mainframe N Make mainframe N the target of subsequent com mands The default 1440 mainframe address is 1 and the default 4032 mainframe address is 16 hvon Turn high voltage on hvoff Turn high voltage off set N V Utility Programs 7 5 o nethv Utility Programs and Subroutines set N M V Set the demand value for channel N or N through M to the value V The set keyword can be omitted Channel addresses range from 0 to 255 for 1440 mainframes and from 0 to 31 for 4032 mainframes For 1440 mainframes the sign of non zero values must agree with the polarity of the channel card For 4032 mainframes the sign of the value is ignored Channel ranges are handled differently for 7 kV pods in 4032 mainframes For these pods a range af fects every other channel For example assuming that two 7 kV pods are i
93. crimination Q Stop Compression Several Variables at Once Any CAMAC variable can be compressed In some cases it could be quite inefficient to do so as the saved data contains both the saved values and the saved indices For instance if less than half of the inputs to a module have unimportant values compress ing the data would result in an array larger than the original To use this feature simply use the compress keyword in the same way as read or save For example the following state ment is valid assuming that timing is defined as a CAMAC variable read compress and save timing In this default case LUCID assumes that zero values are unim portant and all non zero values are saved The compress fea ture only makes sense when used with the read and save command The compress keyword will be ignored if read and save is not specified on the same line It is often important to have some range of unimportant values for compression such as the case of a TDC module operating in common start mode In this instance all full scale values would be unimportant At this point we can think of the compress statement as a software discriminator The following example shows how this can be done read compress and save timing discard 200 255 This is similar to the previous example except any values in the range of two hundred to two fifty five inclusive are removed when the compression is done The discard keyword can ac
94. ctrum spectrum O spectrum 2d View Messages spectrum cut doug amp orion is now in charge of the experiment Disconnected from demo doug on skatter doug orion is now in charge of the experiment Name Type X axis Y axis Display Selected Variables Save Setup Restore Setup Figure 2 8 Sub windows from the View Menu The current setup of displayed histogram windows can be saved with the Save Setup button A file is written in the user s current experiment directory which records which variables are current ly displayed their position on the screen the window size limits of the histogram displayed and marker positions These settings can be restored by using the Restore Setup button Refer to the next section for information about using the histo gram windows The second option on the View menu is for the View Messages sub window LUCID displays several messages here as the ex periment proceeds and it is useful to examine them occasionally In particular errors encountered while reading the user s de scription files will be printed here along with corresponding line numbers in the description file A limited number of messages are saved and the scrollbar on the side of the window can be used to peruse back through them Typically the last 200 mes sages are saved The Properties Menu As the user works with the lucid program there are other ad ministrative options which are frequentl
95. d binary ascii variable to program destination Loading Histograms from Files or Programs The reverse of the write statement is the load statement which allows reading previously saved or software generated histo grams into the LOOKER The load statement expects input in the same format as that output by the write statement The EVENT Section Printing Values from the Making a LOOKER Description File 4 24 m The Looker Setting Looker Variables Interactively Users can set simple variables directly by using the read state ment A read statement looks like read variable variable prompt The user is prompted for input with the string prompt for each of the variables in the read statement The value input from by the user takes effect immediately Be warned that analysis stops while the LOOKER is waiting for input and this can cause the entire experiment to pause waiting for a response This is best used in a beginrun event for log information Using Your Own Subroutines One important aspect of the LOOKER is the capability to perform your own subroutines whenever an event is encountered Argu ments may be passed to the subroutine and may be variables arrays or histograms The following example shows how this is typically done define tdc 32 int 4 define adc 32 previous data event wirechamber call tracking tdc adc The word call must appear immediately before your subrou tine s name The par
96. d 2 44 Description files Looker 4 2 Reader 3 2 Display Options Xlucid 2 29 E Eject button 2 22 Event Rejection in Reader 3 26 Event Section in Looker 4 8 in Reader 3 12 Some events in Looker 4 10 Experiment Name property 2 16 extractrun 7 2 F findeot 7 2 Forward button 2 22 Function defines Looker 4 6 Functions awd dwa 7 10 LF open LF_event 7 11 LLucidopen Lucidread 7 10 Lucidwopen Lucidwrite 7 11 NetCamac 7 12 NetHv 7 13 Qread 7 14 Qwrite 7 14 User Functions in Looker 4 25 User Functions in Reader 3 11 G generate 7 2 H Hardware Database File Property 2 16 Histogram 2D in Looker 4 36 Acceptance tests Looker 4 21 Index 1 ee Index Arrays in Looker 4 28 4 29 Bit histogramming in Looker 4 40 Counts per bin 4 29 Definition in Looker 4 28 Density Map 2 44 Edit Data 2 44 in Looker 4 26 Incrementing in Looker 4 16 4 33 Looker Arrays 4 28 4 29 Print menu button 2 46 Region Save Region 2 45 Regions in Looker 4 18 Save Region 2 45 Send Region 2 45 Sub views Xlucid 2 28 Histogram 2D markers 2 32 Histogram Autoscaling 2 30 Histogram Bins in Looker 4 28 Histogram definitions in Looker 4 28 Histogram Display Bin Numbers 2 30 Histogram Display Options 2 29 Histogram Grid Lines 2 30 Histogram Markers 2D 2 32 Adjusting 2 32 description 2 27 Histogram menu buttons 2 23 Histogram Modes Pan 2 42 Region 2 37 Scale 2 39 Scale temporary 2 43 Tit
97. d the list is empty The second part shows the selected output device and its associated settings The third part are the control buttons The final part is the event list this is only appli cable if the current event is tagged as an event filter To add a new output file ensure that no entries are currently se lected then enter the name in the Filename field and press return The default flag setting of SAVE is displayed If there is only one output or this is the important output set the Fail on Error box If the filename entered is actually a program select the Program box Event filtering can be done by setting the Event Filter box The six control buttons provide the following features Change Tape shuts down the current writer ejects the tape and waits for the user to acknowledge completion of the tape change Delete removes the currently selected entry from the list The WRITER is not informed of this change until the Ap ply button is pressed UNDO puts the most recently deleted writer entry back on the list The WRITER is not informed of this change until the Apply button is pressed 1 Further discussion of the L Menu Buttons The Build Menu UCID writer abilities is found in Chapter 5 The LUCID Program 2 11 Using LUCID i WRITER OUTPUT FILENAME SAVE vega dev nrstl SL DJ Filename vega dev nrst1 Wi SAVE O Event Filter L Failon Error Program Event List co Figure
98. d an X arm event event electron increment elec using scin adc The scin adc array is used every time an electron event is found and is used to build up a histogram If an attempt was made to use scin adc when a hadron event was encountered an error message would be generated LUCID knows that the READER didn t save scin adc within the hadron event and that the array has no meaning when such an event is found in the data stream The EVENT Section Event Types Making a LOOKER Description File 4 9 The Looker Processing Some of the Events The LOOKER allows instructions to be performed at other times not necessarily based on a particular type of event being encoun tered For instance an event name is not required to appear be fore a specification event X 1 print Found SOME type of event These statements will be performed when any type of event is encountered Remember that you won t be able to access any READER variables since the LOOKER must know beforehand that the variable will consistently appear for the given event This can t be guaranteed when any event can trigger the state ments This type of event specification might be used if you want some statistics about all events for example see above the number of events processed If a LOOKER performs a lot of computation when finding an event which is quite common in the data stream the entire ex periment will slow down The READER won t process any more data
99. d buttons Clicking the left mouse button once on an area selects that area The selected area can be dragged left or right by holding the left button down after making a selection and then moving the mouse left or right If an area has been se lected pressing the middle mouse button adjusts the left edge of the area to the current bin and the right mouse button adjusts the right edge of the area to the current bin Double clicking the left mouse pointer does one of four different things depending on where the pointer is If the pointer is between the markers the bins between the markers inclusive are added to the region if not all the bins are currently in the region If all bins between the markers are already part of the region the bins between the markers are removed from the region If the pointer is outside the markers and the bin under the pointer is not in the region the bin is added to the region If the pointer is over an area out side the markers the area is removed from the region The Histogram Windows 2 37 Using LUCID spectrum spectrum 7 X 20 to 100 Y D to 2847 AUTO Region 40 to 52 Selected Area Unselected Area Limits of the Selected Area Figure 2 26 Viewing 1D Regions 2D Regions With 2D regions there are three types of objects to work with se spectrum 2d File v View v Edit v Print v Update zoom Zoom Rebin Rebin Unselect
100. d keypad to change the area of the histogram being viewed The user is allowed to easily pan across the data Window Mode or Border Mode is not currently imple mented Title Mode allows the user to adjust the position of the histogram title within the display or add new titles Using these different modes is described in more detail on page 2 34 Each histogram window can be split into several sub views of the same data Each sub view is independently controllable so that the user can view different parts of the same histogram as shown in Figure 2 19 As mentioned below the user can also view different subscripted histograms of the same array using sub views The Histogram Windows 2 28 ode Using LUCID The View Menu Button spectrum iv lew v Elis Chit v E zoom zem Retin Retin Figure 2 19 Multiple Sub Views of the Same Histogram The remainder of the fields in the sub window are applied to a particular sub view The next set of choices is titled Apply To and allows the user to determine which sub view to adjust The user won t be allowed to select a sub view which doesn t exist To add another view or delete an existing one the user can sim ply select Insert Duplicate View or Delete View located farther down in the sub window By selecting a current view first a du plicate copy of it will be inserted or when deleting the chosen sub view will be removed Display Options is a
101. date interval W number specify the writer debugging level Any argument typed on the command line that is not a flag is taken to be the name of an experiment to run as in the example above Because you can only look at one experiment at a time LUCID ignores any extra experiment names entered on the com mand line The LUCID Program 2 4 Using LUCID After typing the lucid command the LUCID window will be displayed looking like this v LUCID Lucid Playback Sequence Update Visible Run 8 Reader oe pU writer Siepred demo_glen virgo Offline Figure 2 1 The Main LUCID Window The window has three main sections Menu buttons located along the top of the window It is from these buttons or their menus that the experiment can be managed Control buttons located on the left side of the window In the example shown above they are inactive These but tons will become active only when software has been suc cessfully generated e Status buttons located on the right side of the window In the example shown above they are inactive These but tons will become active only when software has been suc cessfully generated The next three sections of this Chapter describe them in detail Menu Buttons In the example of Figure 2 1 the user has not yet connected to an experiment there is no experiment name displayed in the lower left hand corner of the window We ll first take a closer look
102. de The page up and page down buttons are also useful they move the Z axis limits up or down by the stepsize allowing the user to pan through the data in the Z direction Zoom Mode Shortcut Key F5 Although the user may run Zoom mode in the same way as other modes it is usually most convenient to simply use the tempo rary zoom mode described on page 2 43 The numeric keypad takes on different meanings when in zoom mode temporary or not as described in Figure 2 31 Again the center button is most useful the user may remain in marker mode to select a marker area then press left meta 5 to zoom in on it The user should keep in mind that the stepsize de termines the amount of zoom that is performed For instance shift left arrow will zoom in on the X axis by 5 stepsize bins With 1 Dimensional histograms zooming in and out on the Y axis can be done by either stepsize bin units or by stepsize percent increments Selecting the Zoom by Percentage and AutoScale options on the Adjust View window sets the autoscaling display to be a percentage of maximum The default displayed percent age is 100 this can be reduced to 1 The important point is that the percentage does not change when the histogram is up The View Menu Button The Histogram Windows 2 41 Using LUCID Zoom out on M N Pan left 1 page by the Same as ZOOM Y auma of bins on screen Zoom out to initial view em z Display entire MM Zoom until
103. ding stop and pause is maintained Because this statement can create large dead times depending on the amount of processing occur ing in the LUCID LOOKER it should not be used for produc tion runs Other CAMAC Operations To process some events other CAMAC operations must be per formed after the data is read This could include clearing an in put resetting a LAM or adjusting other module parameters Itis often important to clear a CAMAC module before proceeding with an event The following statement can be used clear adc In most cases this issues a CAMAC F 9 command which usually clears the lam status bit as well as clearing the module The sys tem CAMAC database is used to determine the correct function code Many CAMAC applications require that a module be cleared af ter itis read Most modules have an single read and clear opera tion in addition to a simple read function By using the statement read and clear adc the READER will perform the correct CAMAC function usually an F 2 command The READER will use a different function code simply because the read keyword appears The clear statement has a different meaning when used independently Using clear on a normal variable i e one that isn t CAMAC variable will simply set the variable or entire array to zero if defined numeric null if defined as a character or false if de fined as a boolean The EVENT Section Other CAMAC Operations Making a
104. e 3 5 clear keyword 3 19 Comments in Description File 3 4 compress keyword Reader 3 22 Compressed Array Data 3 21 Continuous Triggers 3 10 dataway keyword 3 29 DEFINE Section 3 4 Defining CAMAC Variables 3 4 Defining Variables 3 6 Description File 3 2 elif keyword 3 25 else keyword 3 25 endif keyword 3 25 endrun keyword 3 30 event keyword 3 12 EVENT section 3 12 if keyword 3 25 Keyboard triggers 3 10 Keywords save data statement 3 18 Arithmetic operators 3 23 camac 3 28 clear 3 19 compress 3 22 constants 3 13 dataway 3 29 discard 3 22 elif 3 25 else 3 25 endif 3 25 endrun 3 30 event 3 12 if 3 25 load 3 11 3 27 Operators 3 23 save 3 18 suspendrun 3 30 wait 3 20 write 3 26 List of Keywords 3 31 load keyword 3 11 3 27 Making a description file 3 2 Module Timeouts 3 6 print keyword 3 24 Printing Messages 3 24 Program Calling 3 16 program keyword 3 11 Read and Clear Operation 3 19 read keyword 3 14 3 17 Reading CAMAC variables 3 17 Reading File Data 3 15 Reading program output 3 16 Reading User Data 3 14 Arrays 3 15 reject keyword 3 26 Rejecting an Event 3 26 save keyword 3 18 Saving CAMAC Data 3 18 Sections of a Description File 3 3 Software Triggers 3 9 Stopping an Experiment 3 30 Suspending and Experiment 3 30 suspendrun keyword 3 30 Trigger Examples 3 11 Trigger Polling 3 10 Trigger Section 3 8 Trigger Statement 3 9 trigger statement 3 25 Trigger Types 3 8 Triggering another Event 3 26 User Code 3
105. e 4 34 The DEFINE Section Defining Regions Making a LOOKER Description File 4 5 The Looker 2048 Y 1536 A 7 7 7 7 A 1024 77 within defined 0 128 256 region Figure 4 1 Example Defined Region An array of regions is defined as above with the addition of a contains section for every array element If there are insufficient entries then the remaining array elements default to their entire range Extra entries cause a warning message to be produced and are otherwise ignored Defining Functions The LOOKER allows calling C library functions and user writ ten C functions Functions require a define statement to explic itly declare their argument and return types e g define log real 8 real 8 The above example lets the LOOKER know that the C library function log takes a single argument of type rea1 8 and the re turn value is also of type rea1 8 Any function must be defined before use and must have a return type The use of argument typing is optional but recommended the LOOKER will convert any argument of log that is not of type real1 8 to real1 8 before calling the function Functions that reside in the buildname c file can also be defined and used When writing these functions be warned that the ar guments are call by value not call by reference as in the user writ ten subroutines Arrays of Variables You can also define arrays of most types of variables as in the READER define calib 1
106. e a list of run number ranges has been made up the user can select the Start Programmed Playback button the sub window will disappear and the Play button will become highlighted Control Buttons The Record Button The LUCID Program 2 20 opdmEER O Using LUCID oH Playback Sequence Run Numbers for Processing Start Run Number 1 ale Add to List Runs Remove ALL Start Programmed Playbac Figure 2 12 The Playback Sequence Window The Pause Button If the user wants to suspend the run temporarily the Pause but ton may be selected A blinking message will be displayed across the window to remind the user that the run is suspended To resume the run the user need select the Pause button a second time or re select the button which was previously highlighted such as the Play or Record button The Pause button will work whether the system is on line or off The Stop Button To stop processing the current run on line or off the user may select the Stop button The Rewind Button If the current session is off line the Rewind button may be select ed It simply rewinds the input data file to the beginning This is true for tapes or regular disk files If a WRITER is present and ac tive a warning message is displayed Control Buttons The Playback Sequence Button The LUCID Program 2 21 Using LUCID Status Buttons The Forward Button The user may skip ahead to the next run in the data
107. e bin height When rebin ning the actual bins displayed will be adjusted to start the first bin on an even multiple of the rebin factor For example with a histogram displayed with no rebinning and the first displayed bin being 12 if a rebinning value of 5 were used then the histo gram would now start at bin 2 which would be made up of the original bins 10 to 14 The display is also adjusted so the last bin is made up of a number of original bins equal to the rebinning factor and the last displayed bin would include the last shown bin number from the original histogram display Currently when histograms are scaled panned or zoomed the cal culations for the start and end positions are based on the original The View Menu Button The Histogram Windows 2 31 Using LUCID Adjusting Markers Representing 2 D Histograms Visually AutoScale Display Bin Numbers The View Menu Button bin numbers Setting the step size to the rebin factor will make these functions behave as if they had a bin size of 1 Rebinning is not yet available for 2 dimensional histograms The final set of values in the sub window allows positioning the markers within a sub view Values are entered for them in the same way as for limits Any value within the extents of the his togram may be used but the system will round the value if nec essary to a value on a bin boundary Using the Bin fields will avoid confusion since they are always whole number
108. e chamber Remember that the READER simply stops and waits for input when it encounters this statement essential ly stopping the experiment If a user is not running the LUCID command when input is needed the READER will simply wait until one exists before prompting for input Now we ll consider something a bit more useful If the variable used in the read statement is a CAMAC variable then the corre sponding CAMAC module will be read If you attempt to supply a prompt such as the one in the previous examples LUCID will give you an error when you build the software Recall that CA MAC variables are considered to be arrays because they may correspond to modules with several inputs Let s refer back to our example of an adc module which was defined as using in puts five through eleven the following statement could be used to acquire that data read adc As mentioned in the last section referencing an array without using a subscript causes the entire array to be accessed Theoret ically one could replace this example with seven read state ments each using adc with a different subscript but achieving the same goal Such a technique would work but would be much slower than reading the entire array in one statement The EVENT Section Reading Data Making a Reader Description File 3 17 The Reader NOTE SAVE Restriction Format of Event Data The READER determines how to read the modules by checking the system CAMAC da
109. e cur rent experiment directory The default for Prog File Name field is the word Data followed by a dot and the name of the histo gram being read If the histogram is an array the current index is appended to the name Selecting a data source of Program ex pects the Prog File Name field to contain the name of a program The program will be executed to generate the histogram data to be read The Read button causes the histogram to be read Cancel re moves the window without making any changes to the named histogram Shortcuts for the File Menu The user can press s to save histogram data The Save To File sub window will not be displayed so the user should select the appropriate options first The user should remember that data for the histogram currently being pointed to will be saved The Histogram Windows 2 26 ee Using LUCID The View Menu Button The View menu contains a single option labelled Adjust View and allows the display to be tailored in a variety of ways The Modes amp Options X Limits amp Rebin Y amp Z Limits Markers Apply To Insert Dup View Delete View gy Apply To Adjust View sub window is shown in Figure 2 18 Adjust View spectrum Regions Pan Scale windows Display Options Stepsize W AutoScale 1 Display Bin Numbers Subscript 7 aly Grid Lines Statistical Data Zoom by Percentage Adjust View spectrum Y Limits To 2828
110. e defined Commands Commands from your LOOKER Description If you select the button labelled reset the spectrum histogram array will be set to zeroes as was requested in the LOOKER de scription file Remember that the histogram display itself won t change until you ask it to update the view How to Quit Lucid At this point we ve done many of the things that are needed to effectively use LUCID at least while offline But one last thing which is often important is to have the system process your data in the background You typically don t want to wait around if the computer is going to take a long time to analyze your data Select the Quit button on the main LUCID window and you ll be asked to confirm v LUCID c Playback Sequence Update Visible Run 8 wind Reader Stopped You can QUIT without affecting LUCID it will stay running without you STOP Looker Stopped You can SHUTDOWN the experiment You can change your mind and STAY with it Record Pause Eject Writer Stopped demo_doug skatter UI SHUTDOWN Stay A Short Tutorial Session 1 16 o o A Tutorial Introduction LUCID wants to know how much of the system you really want to quit By default the data stream will stay running even after you ve quit the LUCID program itself This is important when someone else takes over the experiment from you If the work station computer should crash
111. e has been applied The user must remember to use the digit keys along the top of the keyboard and not the numeric keypad since they have different meanings A temporary step size can be used by holding down either Shift key on the keyboard So long as it is held down a step size of 5 times the entered stepsize will be in effect The Control key can be held down to produce a step size multiplier of 10 If both the Shift and Control keys are held down at once then a multiplier of 50 will be used The Histogram Windows 2 33 Using LUCID Select the Current Mode Insert or Delete Sub Views Shortcut Key F2 The View Menu Button The most useful shortcuts involve adjusting the view by using the mouse buttons in conjunction with the numeric keypad on the right side of the keyboard As described on page 2 28 the current mode of the program will affect how certain keystrokes are interpreted The mode of the program may be changed quite easily by selecting various function keys located along the top of the keyboard None of the keyboard shortcuts mentioned thus far are affected but the shortcuts involving the view as men tioned below depend entirely on the current mode Rather than selecting the current mode from the Adjust View sub window the user may press the following function keys Mode Function Key Marker F2 Region F3 Scale F4 Zoom F5 Pan F6 Window F7 Title F8 When a mode is selected a reminder will be displ
112. e looker These must not be used for variable names or great confusion and an unsuccessful LOOKER build will result The LOOKER distin guishes between upper and lower case characters in keywords and variable names so f is a valid variable name whereas F causes an error as a reserved keyword Not all the keywords re served by the LOOKER are actually used by the LOOKER some of these keywords are part of the READER language Keywords marked are reserved words in other parts of LUCID and should be avoided Keywords marked 9 are keyword abbrevi ations and would best be avoided as they may disappear in a later version of LUCID Histogramming Bit Positions Defining His LOOKER Keywords 4 41 and ascii beginrun bin binary bins binsize bits bool boolean by call camac char character clear command compress compressed constantly g contains data dataway decr decrement Histogramming Bit Positions Defining His def define discard display elif else endgroup endif endloop endrepeat endrun event events every first from grou hist histogram if incr increment inhibit initialize int integer integrate into label lam X last LOOKER Keywords load microseconds m milliseconds minutes notregion Or previous print program read ready real rebin region reject repeat resumerun save seconds sizeof some suspendrun then
113. e must be en closed in quotes If the LUCID program is being used with a different interface such as a mouse on a workstation then it will install your com mand on a software button In that case you could simply point the mouse arrow at the appropriate title and click the mouse but ton See Control Buttons on page 2 19 for more information To enter a control character as part of the triggering command type a circumflex in front of the character s symbol For example trigger reset every R command or every highvoltage lam This will cause the reset event to occur whenever you type a control r to the LUCID program or if the highvoltage module gives a lam If you really want to have the circumflex as part of a command you should enter two of them in the command one after another The ready status of a module can initiate an event through the when module ready trigger LUCID then checks for a Q response from the module continuously and the event executes when a Q response occurs Finally there is the simplest of all triggers If you enter the key word constantly as part of the statement LUCID will trigger the associated event as often as it can but allow other triggers to take precedence when they arrive For example trigger tdc readout constantly This statement will cause the tdc readout event to be per formed as fast as the computer will allow All other triggers list ed in the file always take priori
114. e to re create exactly the environment an experiment had when it was run Code included in the reader in this manner should be printed and included in the experiment log book and all changes to the code should like wise be documented The EVENT Section The event section of the READER description file is usually the largest because it tells LUCID what to do after receiving a trigger The section consists of one or more event specifications one after another in the file Each specification may contain many state ments but the first statement is an event name similar to a sub routine name in a programming language For example the statement event electron associates all subsequent statements up to the next event speci fication with the event name electron Notice that the first word is event being similar to the first word of the define and trigger sections LUCID makes sure that every event name that you specify is trig gered by something If LUCID finds an event without a trigger it will warn you that the event will never be performed The rest of the statements in the event specification are similar to the statements in most programming languages allowing you to access any variables including CAMAC variables that were de fined in the first section of your file and perform tests or some arithmetic on the associated data Using Variables The acceptable range of values for numeric arguments was listed under
115. e until you ve assured LU CID that you ve seen it s message You can also just press return to confirm your request After a moment the Experiment Start Connect sub window will disappear and a label at the bottom left of the LUCID window will name the current experiment z LUCID DE y The experiment name appears Run at the bottom of the window when you re successfully connected Reader Looker Writer demo_doug skatter Offline 1 You can press return or select the Start Connect button after typing the name It is usually true that pressing return on the keyboard is the same as selecting the confirmation button the button with the double border A Short Tutorial Session 1 5 A Tutorial Introduction As the name demo dougeskatter implies part of this experi ment will use the computer named skatter Specifically we re going to do some simple offline analysis of data using a READ ER and LOOKER Both of these options will run remotely on skatter while we control them from the workstation This brings up an important point about how LUCID works The data stream including any of the READER LOOKER or WRITER options that were chosen is often started on a remote computer CAMAC or previously recorded data is read by the READER The User and the LUCID program need not be on the same computer as the data stream Keep in mind that when we speak of LU
116. e usable since LUCID wants to write an ordered sequence of runs to the output A Run is started when the Record button is selected the current run number is displayed near the top right center of the main window Every time the Record button is selected the run num ber will be automatically incremented If no WRITER is present the Record button will be inactive and the Play button can be used to start the next run The Play Button The Play button can be selected to begin a new run It will only be usable when data is riot being saved such as when analyzing data off line or testing electronics or detectors on line If the system is off line the Play button will remain highlighted until all data has been processed the Stop button will then be come highlighted and the Play button returned to normal If only certain runs are required to be analyzed off line the user may select the Programmed Playback button The Playback Sequence Button It is often useful to replay only certain runs while off line The Playback Sequence button allows the user to specify various run numbers which should be processed The example in Figure 2 12 shows the sub window which ap pears when the button is selected The user must enter the first and last numbers which make up a range of run numbers in the blank spaces provided then select the button to add that range to the list The range of run numbers can be a single run num ber Onc
117. e used with the ta common histogram from the previous example define protons region from 0 to 18 18 bins and 0 to 8 8 bins contains 7 to 12 1 to 6 6 to 14 2 to 5 5 to 15 3 to 4 The following diagram shows how we can imagine this region with respect to the t common histogram from page 4 38 Do Um PT 0 18 The region has the same matrix structure of the histogram but in this diagram the white bins represent the acceptable bins from its definition If this region was used with the ta_common histogram from page 4 38 then only the histogram bins corresponding to the white bins of the region would be accepted In short such re gions will allow more elaborate cuts to be made to your data Analysis can thus be limited to an area of particular interest Defining Two Dimensional Regions De LOOKER Histograms 4 39 The Looker Histogramming Bit Positions Histograms typically count the occurrence of different values within a variable The LOOKER allows histograms to count the frequency of bit values within an integer For example an exper iment might contain an input register which has 16 bits in a word corresponding to 16 physical actions taking place A common practice is to count the times each particular bit is being set ims Each bit corresponds to something of interest Each bin counts the number of times each bit has been set For example the least significant bit rightmost in t
118. eas Shift Polygon stepsize stepsize bins up points up iS Shift Selected Area stepsize bins to the right on the display Shift Selected Area J eee Shift Polygon stepsize left on the display disini P 4 5 points right Shift Polygon stepsize 7 Ss points left Toggle from markers M Shift Selected Area Shift Polygon stepsize stepsize bins down points down Toggle from polygon Figure 2 29 Keypad Functions in Region Mode 2D This represents the most difficult array of functions to use in the entire program because it is inherently difficult to visualize Z axis transformations The center 5 key is interesting because it allows the user to se lect an area with markers then set the Z axis data limits so that the marked data fits exactly within The View Menu Button The Histogram Windows 2 40 ew Using LUCID Reset the Z axis limits Increase the upper to their initial values Z axis limit by stepsize to include all data I 4 Increase both upper and lower Z axis limits by t E C stepsize pan up in Z A Zoom in on X and EM FgUp Y axes by stepsize bins diia Bann c 4 a Zoom out of X and 7 2 Yaxes by stepsize bins Set the Z axis limits to the iH maxima and minima of data PgDn Decrease both upper and currently between the markers lower Z axis limits by stepsize pan down in Z Decrease the upper Z axis limit by stepsize Figure 2 30 Keypad Functions in Scale Mo
119. ecesee zy PgUp Move right marker stepsize if an array Move right marker stepsize e as bins to the left bins to the right E to next subscript Draw or erase markers a Uv o Go to the last subscript in an array The View Menu Button C Move marker stepsize bins down if 2 dimensional Figure 2 25 Keypad Functions in Marker Mode If the right or left arrow keys are pressed so that the marker would be positioned outside of acceptable histogram limits the marker does not wrap around but stops at the limit The center key 5 is quite useful In most modes it means start with markers In marker mode the user can press it once to po sition the first marker over the leftmost bin currently shown on the display A second press will position the second marker over the rightmost bin being shown This is a quick way to sum all bins on the display at any time since the sum of counts between markers is displayed after the second marker appears Pressing the 5 a third time erases both markers To review for a moment the user could create a marker and po sition it ten bins from the left side of the histogram display by pressing the center 5 key on the keypad followed by control right arrow that is hold down the control key while pressing the keypad 6 The Histogram Windows 2 36 m Using LUCID Shortcut Key F3 1D Regions The View Menu Button The Home and End keys will cause the first a
120. ectronics and need to see what the data looks like You can incorporate the WRITER into the data stream when you re ready to record the data If you re running an experiment with a high count rate you might also want to leave the LOOKER out of the data stream after checking the ini tial setup It can be easily restarted between runs Data Communications in LUCID The inter process communication in LUCID can involve up to three separate computers and at least five different processes for an on line acquisition Figure 6 2 on page 6 4 shows a typical on line setup of processes and communications 1 The Front end computer is considered a single computer although it may contain multiple proces sors How LUCID Works 6 3 Lem Conceptual Overview COMMAND MESSAGES y XLUCID RESULT MESSAGES i Writer Supervisor 1 Writer CAMAC Crates and Modules Front end Figure 6 2 Processes and Inter Process Communication The LUCID Experiment Directory For each experiment run on each computer LUCID establishes a unique directory named lucid experiments experiment name Within this directory LUCID places copies of the prefix r and prefix l files In addition a number of other files exist We ll look at the list of files assuming that the experiment prefix prefix was used when building the experiment Lucid socket a special file that allows the display programs to communicate with alre
121. ed Area Selected Area Polygon Figure 2 27 Viewing 2D Regions The View Menu Button The Histogram Windows 2 38 w Using LUCID lected areas unselected areas and polygons An area is a group of one or more adjacent bins and the outline follows the border of the bins Polygons are user drawn lines and go from screen point to screen point ignoring bin boundaries see Figure 2 27 Polygons are drawn on the screen to fill or empty an area Draw ing a polygon is done by pressing the SELECT button when the pointer is outside of any region area then moving the pointer to the next location and pressing again and continuing until all points have been selected Pressing the 5 key on the numeric key pad causes the area within the polygon to become a selected area Pressing 5 a second time causes the area to be cleared If a mis take is made when drawing the ADJUST button erases the cur rent polygon Unselected areas cannot be modified directly but must be changed into a selected area by placing the pointer over it and pressing the SELECT button A selected area can be moved around the screen with the four ar row keys If the selected area is shifted beyond the limits of the histogram the edge of the selected area will fall off A selected area can also be dragged by selecting an area and not releasing the SELECT button then moving the mouse Releasing the SE LECT button stops the dragging Be warned that the screen up dates more m
122. ed without the histo gram being changed define tdc 32 adc 32 int 4 define teres 32 hist from O to 1024 64 bins and from 0 to 64 64 bins event read_data if teres contains adc tdc print Acceptable value incr teres using adc tdc else print Unacceptable value endif This example shows two variables being checked at one time for acceptance in a two dimensional histogram In fact arrays of variables are being checked if any of the values in either array are outside the limits of the teres histogram the if statement will evaluate to false Accordingly a single variable can be tested by simply using if nhist contains nvariable Similarly if the word bits comes after the variable the bits will be tested for acceptance rather than the value of the variable Keep in mind that the variables which are being used in these ex amples can actually be complicated expressions For example consider this if statement which checks two variables for accep tance in a two dimensional histogram if xyhist contains tdc 0 tdc 1 adc 0 0 98 adc 1 The EVENT Section The If then else Statement Making a LOOKER Description File 4 22 OodmmEEE The Looker The Repeat Statement The LOOKER allows an arbitrary number of statements to be re peated according to certain criteria The standard form of the re peat statement can be seen in the following example define x int 4 endrun repeat x from 1 to 100 by 10 print Yo
123. entheses must appear afterwards whether variables are being passed to it or not If arguments are given they must be separated by commas This example shows a subroutine with two arrays being passed as arguments The LOOKER currently understands subroutines written in C or FORTRAN and will call them with the appropri ate arguments Histograms may also be passed as arguments the size of the array will be equal to the number of bins in the original histogram All arguments are call by reference which is the normal method of passing arguments for FORTRAN In C all arguments must be pointers Subroutines may be called at any point in the LOOKER descrip tion file If it prints something using a write statement or printf in C the message will be displayed on the user s display screen or in the experiment s log file as described under Printing Val ues from the LOOKER on page 4 24 The EVENT Section Setting Looker Variables Making a LOOKER Description File 4 25 The Looker Your subroutines must exist in a file with the same name as your description file but having the appropriate suffix For example if your description files were contained within demo r and or demo l demo w your C language subroutines would be con tained in a file named demo c Similarly FORTRAN subroutines would be contained within demo f You are allowed to use both C and FORTRAN subroutines at the same time Unfortunately errors in your subroutine are not
124. enu button 2 27 Histogram Windows 2 23 Histogram Zoom buttons 2 30 Histogram Zoom By Percentages 2 30 Marker Mode 2 34 Modes 2 28 Pause button 2 21 Playback Sequence button 2 20 Properties menu 2 15 Properties window 2 16 Quit button 2 18 Reading Histograms 2 26 Regions 1D 2 37 2D 2 38 Get Region 2 44 Rewind button 2 21 Save Histogram 2 24 Splitting Histograms 2 28 Status buttons 2 22 Stepsize 2 30 Stepsize Adjust View 2 30 Stop button 2 21 Supply Input 2 14 User Commands 2 13 User Permissions 2 18 User Permissions window 2 16 View Data window 2 14 View Menu 2 14 View Messages window 2 15 Index 6
125. ere placed by the READER For more information refer to Keyboard Triggers on page 3 10 The EVENT Section Event Types Making a LOOKER Description File 4 11 The Looker Built in Variables In summary there are five different causes which are recog nized to start performing statements in the LOOKER event pion event pion some event event some beginrun endrun command calibrate Using Variables Variables in the LOOKER can be used in the same way as they would be in the READER For more information about what val ues are acceptable and how constants may be specified refer to the READER section Using Variables on page 3 12 Some notable differences from READER variables should be re alized here First double precision floating point variables are accepted in the LOOKER The following example demonstrates define d energy 6 real 8 Second histograms and regions are variables which the READER doesn t understand These rather specialized variables are known only in the LOOKER but they are treated as variables in most respects For a description of these variables refer to LOOKER Histograms on page 4 26 and Histogram Regions on page 4 33 Third the LOOKER knows nothing about CAMAC variables If the READER has saved values from CAMAC variables the LOOKER will only see their values as data without caring about their origin Many other features of using variables are the same as they are in the READE
126. erial could represent a trigger signifying that CAMAC equipment should be read out at that point The next chapter describes several different ways to trigger the acquisition of event data In many experiments it may not be possible to relate a unique trigger to every different event For instance you might not know what type of interaction occurred until a bit pattern register is examined LUCID handles this by letting you reject the current event and trigger a new event after performing some tests The best way to get started with an experiment is to list every dif ferent type of event the experiment will encounter give each one a unique name and describe how each event will be triggered This is certainly the first step in deciding how to set up your elec tronics and will make the job of describing the experiment to LU CID very easy In the next chapter we ll see that in order to acquire data LUCID just needs to know where to get the data and when to read it If you re interacting with existing data from a previous experi ment things are much easier because each event type and trigger has already been defined You won t have to prepare any kind of description because LUCID can extract the original description directly from the saved data You should at least be familiar with the names of the variables that were used originally on line If not LUCID can list them for you see lucidview on page 7 4 LUCID allows you to create
127. ertagger 2 SAL stu shortword read is f0 a0 clear is 10 a0 module Accelerator lab homemade tagger trigger module SAL supertagger SAL stm shortword read is f0 a0 clear is f9 a0 module Accelerator lab homemade output register SAL outreg SAL oreg shortword read is f0 a0 write is 16 a0 clear is f9 a0 module Interface Standards OD 48 48 bit output register Sample CAMAC Module Database A 8 Od mEEEE od A48 readsize 2 writesize 2 longword read 0 a0 1 write f16 a0 1 test is f8 a0 clearlam is f10 a0 disable is 24 a0 enable is f26 a0 check is f1 al data amp 017600 02000 module Jorway model 41 NIM output register J41 Jorway 41 shortword write 16 a0 module Kinetic Systems 3615 6 input 100 MHz Scaler 3615 ks3615 ksc3615 Kinetic Systems 3615 readsize 6 longword read is f0 a0 5 clear is 9 a0 5 readclear is f2 a0 5 enable is 26 a0 disable is 24 a0 testlam is f8 a0 5 clearlam is 10 a0 5 module Kinetic Systems 3924 LAM Encoder for Serial Highway 3924 ks3924 ksc3924 Kinetic Systems 3924 readsize 1 read is f1 a12 enable is f26 al disable is f24 a1 testlam is f8 a0 clearlam is f10 a0 clear is f10 a0 module LeCroy Model 2228A 2229 Octal TDC 2228A 2229 LeCroy 2228A LeCroy 2229 Sample CAMAC Module Database A 9 timeout 100 readsize 8 shortword autoclear read is f0 a0 7 readclear is f2 a0 7 test is f8 a0
128. evel CAMAC Access ssssscccccceseessesccececeessssstscceceeeeeeesees 7 12 Dish Voltage AGUBSE edid ti eii EG Pul GE FU rn Lni lr nM ME 7 13 Reading O format DOLO eges deecttetited tiat rt RiR ENRE ER 7 14 Writing Q format Data sssini 7 14 LUCID Data File Orin 20 secsussssaccsansaccecarnviecnanstaceieendwiemonnewiecnd 7 15 LUCID Data Record Format Declarations 7 15 Appendix A The CAMAC Module Database Defining the Modules ueniunt eta Ina i A 1 CAMALC Capabilitie Secciones A 2 Readsize and WritesiZe sucr crisprrisorii inaunda eene enne eene A 2 Shortword and Long word uisesoosaniis etui bv iret at A 3 Checking Module Identity i2 onsceima cote expe tre tdi A 3 Reddine haee a Ia NUI aa MIU IL A 4 HOLD Mode Transfers eeeeeeeeeeeee nme ener aa A 4 banum A 5 Read arid Clear deactivated Eee Drev dee UHR A 5 CNC AN E AAEE A E E E A E A A A 5 Enabling and Disabling LAMS suiscnniese et orbodeini dE eDim uad A 5 Testing for LAM Seat tassa tei ecrit uiboba utn ete id A 6 Clearing LANIS dg cii vi pianta belidULi as Ono eet t d rM RUE ER UAE A 6 Selt learning Modul s 52er ren coit d ve cri toli A 6 Q Stop Modules siinse e eni ees aia ai Herd ld A 7 Selt Compressme Modules uiae teeden etui dre ri rna ee seansie a A 7 Sample CAMAC Module Database cisonectt desertae inl A 8 Appendix B Example Description Files Appendix C Known Problems lare BL Pr e C 1 Problems When Runn
129. everal menu buttons located along the top of the window The first four correspond to the most common steps that you ll take during the lifetime of an ex periment In very general terms the process could look like this Gain Control of the Data Stream Have LUCID Build the Software Gain Access to Your Experiment View the data as the Stream is being Processed possibly make some changes to the experiment A Short Tutorial Session 1 3 A Tutorial Introduction Type demo yournameQskatter for this tutorial Buttons are drawn on the screen to allow choices to be made To make a selection move the mouse arrow over top of the button and click it that is press then release the leftmost button on the mouse the left mouse button is called the select button A menu button is slightly different it has a small triangle dis played to the right of the label This means that more options are available which you can see by clicking the rightmost mouse button Start Connect Search and then Connect Change Directory A menu is displayed containing all of the options for that button For instance the Experiments menu button contains four op tions The fourth one has been made inactive in this example and the first one has been highlighted with an oval In this case the first choice is the default one that s the one that will be used when you select the button wit
130. f the variable If it is an array another dot is The Histogram Windows 2 24 odmEEE O Using LUCID The File Menu Button oH Save Data to File Variable Name spectrum 7 File Name Data spectrum 7 mec E Fe format E ASCH f Einary Values to Save X V Values Start of Bin PH inici iin Figure 2 16 The Save Data to File Sub window appended to the name followed by the current subscript being displayed The next choice which the user can make determines whether the output file will be appended to or overwritten by the saved data By default the action of saving data will overwrite the ex isting file The next choice is concerned with the format of the data to be saved Currently all data is saved as printable ascii characters which can later be edited or used as input to other programs For 1 dimensional histograms the X axis value is printed fol lowed by a blank space then the number of counts and a newline character for every bin that is being saved For 2 dimensional histograms the X axis value is printed fol lowed by a space and the Y axis value then another space char acter and the number of counts in the bin and a newline character In either case the user may save bin numbers instead of X and Y axis values by selecting Display Bin Numbers for the histogram as described on page 2 30 The field labelled Values to Save allows the user to compress the amount of saved data by not sa
131. following example shows a very simple LOOKER description file which recognizes 3 different event types in the data stream define el e2 e3 int 4 event photon el 1 print Found a Photon Event event calib e2 1 print Found a Calibration Event event scaler e3 1 print Found a Scaler Event This is an overly simplistic example meant to show the layout of the description file Notice that the event section itself is quite similar in form to that in the READER description file refer to The EVENT Section on page 3 12 The rest of the statements in the event specification are similar to the statements in most programming languages allowing you to access any variables that were defined in the first half of the file or previously in the READER Histograms may be built up on an event by event basis since each event specification is repeated for every event of that type which is found in the data stream Keep in mind that variables which were defined in the READER are only accessible if they were saved in the READER Further more such variables will only be accessible within the specifica tion for the event in which it was saved For example let s assume that the READER had events named electron and had ron and also an array called scin adc which was saved within the electron event A LOOKER description file might look like this define elec 12 hist define scin adc 12 previous data int 2 event hadron print Foun
132. follows declares writing machine s binary representation defines layout and variables used in an event s data record See struct Lucidsym below This record will contain Lucidvar struc tures to describe individual variables written when user changes experiment s reader description file Essentially contains copy of the file itself See struct Luciddescribe below Padded with newlines to a 0 mod LUCIDALIGN_RECORD boundary written when the Looker description file changes similar to LREC_READER above See struct Luciddescribe below Padded with newlines to a 0 mod LUCIDALIGN_RECORD boundary written when the Writer description file changes similar to LREC_READER above See struct Luciddescribe below Padded with newlines to a 0 mod LUCIDALIGN_RECORD boundary LREC_NEWVOLUME marks the transition from one file to another typically a change of tapes in a stacker See struct Lucidvolume below LREC_NEWSESSION written at the beginning of each session See struct Lucidsession below LREC_BEGINRUN written at the beginning of each distinct run whenever the begin run trigger is processed See struct Lucidbegin below LUCID Data Record Format Declarations LUCID Data File Format 7 16 d Utility Programs and Subroutines LREC_DATA LREC_ENDRUN LREC_COMMENT LREC_NOTHING Contains zero or more event records as defined by the user s reader file definition of each event Each event has a similar over all structu
133. g LUCID Command Line Options Command Line Options To start LUCID you can just type the word lucid as a command in a window On some systems you might have to type xlucid to get the window version see your system administrator if you re not sure You can also type xlucid experiment _name computer name This is a shortcut for having the program automatically connect to the named experiment on the named computer In fact there are lots of options that are available on the command line all which will over ride the corresponding entry from the property list See The Properties Menu on page 2 15 for a more complete explanation of what each option does The valid command line options are n ignore the X11 default properties for LUCID b file specify the base description file name d number set the program s debugging level ename specify the text editor to be used i file specify the input data file for off line data m file specify an alternate CAMAC Modules database s file specify the device name for the frontend processor t number set the maximum number of displayed messages u directory specify the user directory to execute in F number specify the lucid finder debugging level L number specify the looker debugging level M number specify the manager debugging level R number specify the reader debugging level S number specify the debugging level for the front end pro cessor U number change the status up
134. h is located on the left side of the space bar on the keyboard it has a small diamond symbol 0 on it While holding that down press the 5 key on the keypad it s the key on the far right with R11 written on the edge This will zoom in to the marked area spectrum File v view v Edit v Print v Update zoom zoom Rebin Rebin Notice that the marked bins now appear on the edges of what is displayed Your markers might not have been placed on the same bins as in this example so the display might not look iden tical The leftmost meta key always means zoom so you can zoom in or out vertically or horizontally by using the arrow keys with the meta key depressed To zoom back to the initial view press meta Home 4 Some older keyboards have the meta keys labelled with the words left and right but are still located immediately next to the space bar on either side A Short Tutorial Session 1 15 A Tutorial Introduction Sending Pre defined Commands Recall that in our original LOOKER description file we had a in struction that was to be performed whenever the user issued a reset command Move the mouse arrow back to the original LUCID window and select the Control button A window labelled Control Pre defined Commands will ap pear and will contain a button for every command that you ve included in the description file 4H Control Pr
135. handled grace fully in this version of LUCID and the entire experiment will stop if your subroutine performs an illegal operation such as an array subscript out of bounds or an attempt to divide by zero The LOOKER has three user callable routines available Lucid savehist Lucidloadregion and Lucidloadvars All these routines must be declared as integer 4 within any looker pro gram that calls them Lucidsavehist varname filename flags is similar to the write statement The main differences are strings variables can be used to specify the variable name or the file name selection of type of save compressed binary vs ascii save to a program bin vs value numbering can be done at runtime saving a mid point value is optional The variable name is passed as a string either a string variable or a string constant so the function can obtain all necessary information for saving the histogram Lucidloadregion regionname filename is like a load state ment for a region The input format looks like a region definition within the looker Lucidloadvars filename is similar to a namelist read in FOR TRAN Each line in the input file is expected to have the name of a variable declared in the looker an optional subscript which must be a constant an equals sign and a replacement value The replacement values are converted to the type of the variable String variables expect their values to be placed within double q
136. he View Menu There are several shortcuts which the user can make allowing quick adjustment of the view First the four display options can be changed by pressing single characters on the keyboard An upper or lower case a will turn on the autoscale option Another one will turn the option off Pressing an upper or lower case b will enable or disable Dis play Bin Numbers in a similar fashion The Histogram Windows 2 32 Using LUCID Second Marker First Marker Grid Lines Statistical Information Update the Display Changing the Stepsize The View Menu Button c psdright 0 Figure 2 22 Markers on a 2 Dimensional Histogram 414 n Accordingly typing a g will draw a grid of dashed lines over the display and pressing it a second time will erase them A cap ital G will perform the same function A capital S will toggle the display of statistical information A lower case s will not do the same it causes data to be saved as explained under The File Menu Button on page 2 24 u n Pressing the will cause the display to be updated with new data An upper case U will do the same thing The Stepsize function can easily be changed by simply typing a number The step size is used as a factor for most operations as mentioned on page 2 30 For instance if the user normally press es some keystroke to move a marker 1 bin to the right on the dis play it would move 10 bins after the stepsiz
137. he diagram in the integer might represent the occurrence of a special hard ware trigger if the bit is set the trigger has occurred The sec ond bit could be set if a calibration signal was present somewhere In any case the LOOKER allows you to count the oc currences of these bits being set by using the bits keyword define inreg previous data int 2 define bitcount hist from 0 to 32 32 bins event readbits incr bitcount using inreg bits 1 If a bit is set it contains a 1 If cleared it contains a 0 Histogramming Bit Positions Defining His LOOKER Histograms 4 40 dmm The Looker Histogram The definition of the histogram is the same as any other histo gram but the increment statement is slightly different The bits keyword tells the LOOKER to increment bins using the bit posi tions which are set in inreg Notice that several bins could be in cremented at once since a particular value within inreg might have several bits set For instance if the integer contained the value 11 the binary representation would be 001011 and three bins would be incremented at once Bits 0 1 and 3 are set meaning that bins 0 1 and 3 will all be incremented at once The least significant bit in the integer will cause bin 0 to be incre mented The second least significant bit will increment bin 1 and so forth LOOKER Keywords There are a number of keywords reserved by th
138. he functions available in panning mode As with zoom mode panning across a display is most commonly performed with the temporary mode keys as described on page 2 43 The arrow keys will shift the view in the direction of the ar row by the stepsize amount not by the size of the pan window The Home button causes the lower left portion of the histogram to be displayed the End button causes the upper right portion of the histogram to be displayed The Page Up and Page Down buttons work the same for both 1 dimensional and 2 dimension al histograms the histogram is shifted left or right by they num ber of bins on the page The display can be dragged using any mouse button As might be expected the histogram will follow the movement of the mouse when a button is pressed inside the histogram and the mouse is moved left or right The user should keep in mind that the stepsize determines the amount of panning that is performed For instance shift left ar row will move the display 5 stepsize bins to the left Temporary Scale Zoom and Pan Modes The user is allowed to temporarily go into scale zoom or pan modes by holding down a modifier key as shown in Figure 2 33 regardless of the current mode A modifier key acts like a shift or control key the user must hold it down while some other key is pressed For example the temporary scale mode is identical to regular scale mode describe above except that it is in effect so long as the use
139. he user may define the events manually or the name of a LUCID data file containing a valid symbol table This symbol table gets copied to the output data stream Writing LUCID Data Records LUCID Functions and Subroutines 7 11 o Utility Programs and Subroutines LF define eventno eventname varlist count LF define creates a symbol table entry for the event named eventname numbered eventno Varlist is a pointer to an array of count Lucidvar variables that will be entered into the event description record LF open filename experiment description filename is opened for writing LUCID events The output starts with a NEWSESSION record and a LUCIDFORMAT record If an rsymtab file was given to the LF rsymtab filename then the symbol table is copied from that file to the output file If LF rsymtab was not called then the experiment character string is used to find a computer and experiment directory with an rsymtab file for buildname description An example of look ing for a description file is LF open testdata demo glenGvirgo demo which opens the file testdata for output and looks in the direc tory home lucid experiments demo glen on the computer virgo for the file demo rsymtab The description name and experiment name are used as part of the start session and start run entries Run numbering starts at 1 LF event eventno buffer nbytes LF close Low Level CAMAC Access An event record is written
140. he variable can also be an array name without a subscript as discussed in previous sections The expression can be a regu lar arithmetic expression using the following operators Symbol Function exponentiation multiplication division addition subtraction bitwise left shift bitwise right shift integer modulus remainder from division bitwise and bitwise or bitwise exclusive or one s complement negation boolean not The first part of the table contains binary operators requiring two operands and the second part lists unary operators Parentheses can be used for grouping The bit operators can be used on any integer variables The EVENT Section Calculations and Variables Making a Reader Description File 3 23 The Reader Using Arrays Without Subscripts An expression cannot contain a subscript less array reference unless it is used in an assignment to an array For instance it is legal to use an assignment like corrections corrections 4 where corrections is previously defined as some kind of nu meric array However it would be incorrect to use path 7 corrections 4 where corrections is the same numeric array In this exam ple note that path is an array containing at least eight elements indexed 0 through 7 so the entire corrections array can t fit into a single element of path A further restriction is that the array on the left side of the assig
141. hen you re finished looking at the description file quit the ed itor by getting the menu from the window s top label and select ing Quit from it Window y Edit Close define demo adc 12 previous data int 2 Full Size efine spectrum 12 histogram from 0 to 2048 x event pulse Back increment spectrum using demo adc Refresh command reset Quit spectrum 0 c3 LOOKER Description demo l At this point all you need is to point back to the Build sub win dow select the Make button and the software will be generated You ll be asked for a confirmation Build Experiment demo doug amp skatter CAMAC Device Data File demodata Confirm Make software for offline analysis having demo as the LOOKER description Data will be analyzed from demodata ption Files demo Yes Make the Software Edit demor Edit demo The resulting analysis software will be in the form of a program about four or five hundred lines long including comments and data structures Very large experiments might end up with LUCID generated programs containing ten thousand lines of C program source Typically we re not interested in looking through the resulting program but LUCID adds lots of comments in case you re curious Chapter 6 describes where the generated programs can be found A Short Tutorial Session 1 10 Oo mHEERR A Tutorial Introduction While LUCID is
142. histogram is allowed to count values in any range that can be contained in the variable The resolution of any histogram is defined to be the width of a single bin or the range of different values a bin will accept Defining Histograms The following example shows how histograms can be defined define el e2 e3 histogram from 0 to 1024 This allocates room for three histograms whose names are e1 e2 ande3 Each one can count values from 0 up to but not including 1024 If one of these histograms were given a variable with a val ue of 1029 for example it would simply ignore it By default LU CID gives every histogram 1024 different bins Therefore the default resolution is 1 1024 of the histogram s acceptable range The keywords rom and to name the limits of values ac cepted by the histogram Remember that the value of 1024 will never be counted in this histogram For now we will assume that only integer variables are being counted but real variables floating point can also be used as described on page 4 31 Let s look at another example define pions histogram from 100 to 126 26 bins Histogram Definition Defining Histograms LOOKER Histograms 4 28 dmm The Looker Imagine how this histogram would look after it has monitored some variable for awhile The height of each vertical bar corre sponds to the number of counts in each bin K 26 bins 10 last bin has counted 2 occurrences
143. home lucid include qformat h This software has not been extensively tested Qopen char name The file referenced by name is opened to be read by the Qread routine 0 is returned on success 1 on error Oread The next Q record is read and returned as a pointer to the Qtape header structure This pointer can be cast to point to the appropri ate structure type NULL is returned on error or end of file Oclose The Q data stream is closed Writing O format Data These routines were originally intended to allow Q data to be written directly from the writer Because the Q information is a subset of what LUCID records experimenters must now record their data in LUCID format and then use the lucid to q pro gram to convert their data This allows complete access to the ex periment data for independent analysis by different groups Reading O format Data LUCID Functions and Subroutines 7 14 _ eee Utility Programs and Subroutines Qwopen char name The file referenced by name is opened for writing of Q format data Owrite void lucidp The lucid record pointed to by lucidp is converted to the proper type of Q record and then written 0 is returned on success 1 is returned on error Owclose The output file is closed LUCID Data File Format This section describes the format of LUCID data Data defini tions are shown using C syntax and can be found in the LUCID include file lucidformat h LUCID Data Record Form
144. hout getting the menu When you select a button containing an ellipsis a sub window will be displayed To make a selection from the menu just click the rightmost mouse button over top of the choice you want Starting an Experiment To start an experiment select the Experiments menu button A pop up window will appear asking you to name an experiment LUCID Experiments v Build v Control v View v Properties v Quit Playback Sequence Update Visible Run Looker demo doug Gskatte Start Connect Experiment Name demo_doug skatter Start Connect A Short Tutorial Session 1 4 m A Tutorial Introduction You re going to create a new experiment using your own name as a base Make sure that the pointer is pointing into the Experiment Start Connect sub window then using your login name instead of name type demo name skatter and press re turn First you ll be asked to confirm your choice then to con firm that the experiment should be created oH Experiment Start Connect Experiment Name demo doug amp skatter There is no demo_doug experiment on skatter Do you want to CREATE one GD GD Use your own name but make sure that it starts with the word demo so it won t be mistaken for a permanent experiment Select the Create button if LUCID is displaying the correct name You won t be able to do anything els
145. i e IER a Er eds 2 20 THE Play BUON T 2 20 The Playback Sequence BUON s aera reni quaereret e o ai 2 20 The Pause Button t de i aee dens 2 21 The Pop BuUMOi enti mrad inca tute uie medie ramadan 2 21 The Rewind Button sss enne enne 2 21 The Forward Button sees 2 22 The Eject DUIOT eese ete labia oerodicbo elc OR tee eR 2 22 The Update Visible DUI OD un easinu nti coded ie 2 22 SAUS BUON S siete utes A E eee eas 2 22 The Histogrammi V BOOWS cilia boictieditdbrs dace idi bor dandus 2 23 The File Menu Pt OA ches oot rear rhe eani adu A 2 24 Shortcuts for the File Menu essere 2 26 Lucid User s Guide 1 ee Contents THe Vie We MENU BUON M 2 27 Shortcuts for the View MODE iere uberes mbH ei sismis 2 32 luli iato s P M 2 34 PROP TOT i oro e R 2 37 velimus M M 2 39 ZOOM mie MM 2 41 PP NIBUS vas cee tute ELE e iq e saan e D et RAD 2 42 Temporary Scale Zoom and Pan Modes cette 2 43 Wind w MOG giclee arabika da eo ENDE 2 44 Tile MOG T H 2 44 Jie ECDERIERGCDURCO Loeb a aet trenta idt PERDIDA DE HE 2 44 Th Edit Data ODDO es tones vobi vitu pre aede bt dro ede tetti 2 44 Ihe Density Map ODBOT scasassbtcttuimeniex n d cres CUR EE ED RR 2 44 The Get Region CISUQDI a teni donare dii eksen aie an ite 2 44 The Send Region DEO eus ona irre ta nate onto iri reus Ora 2 45 THE Sa
146. ie LUCID User s Guide LUCID Copyright 1990 1995 by The Saskatchewan Accelerator Laboratory Permission to use copy modify and distribute this software and its documentation for any purpose and without fee is hereby granted provided that the above copyright no tice appear in all copies and that both that copyright notice and this permission notice appear in supporting documentation and that the name of Doug Murray and the Saskatchewan Accelerator Laboratory not be used in advertising or publicity pertain ing to distribution of the software without specific written prior permission Neither Doug Murray nor the Saskatchewan Accelerator Laboratory make any representations about the suitability of this software for any purpose It is provided as is without ex press or implied warranty The sale of any product based wholly or in part upon the technology provided by LU CID is strictly forbidden without specific prior written permission from the Saskatchewan Accelerator Laboratory LUCID technology includes but is not limited to the source code executable binary files specification language and sample specifi cation description files Comments bug reports and changes to this software or documentation should be mailed to lucid skatter usask ca Bug reports should be accompanied by sample in put or description files if appropriate UNIX and OPEN LOOK are registered trademarks of Novell Inc Sun Microsystems and Sun Wor
147. ient use of computer memory when the pludata array is only used once in the entire program taking valuable space away from data areas An alternative to this solution is found with the load statement Loading Data into a CAMAC Module The write statement is useful for loading values into CAMAC modules but is inconvenient for loading large amounts of data To handle such cases LUCID provides a load statement which transfers data from a file directly into a module For example the following statement would perform exactly the same function as the last example in the previous section load plu from home norum plu datafile This is an efficient way of getting values into a module and it doesn t waste any valuable data space In many cases an experimenter might load a waveform genera tor or a programmable logic unit with numbers produced by a program LUCID allows you to load the output directly into the module without making an intermediate file For example load wavegen from program home pywell bin makewave This is identical to the read from program statement previ ously discussed Keep in mind that the exact contents of the file are loaded into the module no conversions are done so that modules requiring binary information must be loaded from bi nary files It s also a good idea to check the CAMAC module da tabase for the number of bits that the module needs for a valid transfer either sixteen or thirty two Your data
148. ile once at the very beginning of your experiment when it ac quires the raw data You won t have to remake the description in the future to look at the data off line The READER description file should be named appropriately for your experiment but must end with r for Reader You should not use any special symbols such as asterisks or question marks in the filename but it may contain digits and upper case letters such as in He3 r or carbon14 r Remember that LUCID only needs a description of your experi ment when raw data are being acquired from CAMAC If you want to get data from some other place such as a tape or a disk file then you can skip ahead to the next chapter you don t need a READER description file to read previously acquired data Making a Reader Description File 3 2 lee The Reader Example READER Description File The contents of the description file tell LUCID what data to read when to read it and what decisions are to be made before the data is considered acceptable Here s an example of a simple READER description file acquire data from LeCroy 2249W charge integrating ADC define delta e 2249W B 0 C 1 N 5 trigger background every delta e lam event background read and save delta e The first four lines are comments which LUCID will ignore The remaining lines can be grouped into four sections Four Sections of a Description File 1 Definitions of variables LUCID treats CAMAC
149. in all bins in the pion histo gram define sum int 4 define pion hist from 0 to 512 512 bins event moredata sum Spion The summation operator may also be used on regular variables the result will be the value of the variable itself More useful however is that the operator may be used on an array The fol lowing example will add all of the elements within the array named junk define tot int 4 define junk 100 int 4 endrun tot junk The EVENT Section Calculations and Assign Making a LOOKER Description File 4 14 m The Looker Assigning Arrays of Different Sizes Notice that the array has no subscript If it did the summation operator would evaluate to the value of that single element The summation operator can be used wherever a single value is re quired An expression may be used to yield several results at once when entire arrays are used at one time As mentioned in a previous section you may clear arrays with one statement you may also assign entire arrays with only one statement Using the previous example we could add the following statement new_e laststep 1 05 This statement assigns each element of the laststep array in creased by 5 percent to corresponding elements of the new_e ar ray This seems to contradict a previous statement which said the an expression is expected to yield a single result A better statement is that an expression is allowed to yield as many values as the left
150. ing Off Line uiieniseaen iun roe bretekeeiciad C 1 Problems When Running OneLiftie assdqenienetdotsi irte bebe cori C 2 Requested Enhancements xisaincsvsiaiesnmasdesinsasaioonaneniiniastenniadincesnetans C 2 Changes to READER o ooaaocptie pop pipi pietate dA aan C 2 Changes to WRITER ciosxioR ii aiat eu iunc pu rue ten bon bU REDE C2 Changes to LOOKER uii ERR AARE IDA NIMM INA MEM MUR C 2 Chansesto NEUCID criteiretenienieniantcg rduananhninieinsin ns C 3 Lucid User s Guide V Appendix D Histogram Save File Format Overview cccccccccccececececececececececececececececececececececececececececececececececececececs D 1 Header oaeee haud e TAM Rae Rae UEEUE D 1 Dur qe D 2 Lucid User s Guide vi Re Chapter A Tutorial Introduction LUCID is a system which provides a way to acquire and analyze experimental data It sets up a data stream that has three distinct parts the READER the LOOKER and the WRITER 1 The READER gets data from some source such as CAMAC It can perform simple software cuts then make the data available to the rest of the system 2 The LOOKER is optional and allows analysis of the data as it is being acquired Although its name implies passively view ing the data the LOOKER can perform elaborate tests and mathematical operations on the data 3 The WRITER is also optional and can save data on magnetic tape in disk files or send the data to other programs as in
151. ining Histograms LOOKER Histograms 4 33 The Looker Region Definition could make up just a small piece of the total histogram To do this LUCID allows you to define regions of a histogram These re gions can be treated as histograms unto themselves but actually use the bins of other histograms Regions cannot have any values on their own and can only be used in conjunction with a regular histogram It is best to think of a LUCID region as a mask through which we can see certain bins of histograms This will be further discussed but we ll first consider how regions can be defined The following example shows how to define a region define tdc rtdc hist from 0 to 1024 define tvall tval2 region from 0 to 1024 The first line defines two regular histograms but the following line describes two regions the keyword region is simply used instead of histogram All other options to define histograms can also be used to define regions and as mentioned under Defining Histograms on page 4 28 the default number of bins is 1024 if no other indication is given Consider another example with fewer bins define rh from 0 to 26 26 bins Let s assume that only a few bins should be considered accept able bins 2 through 5 10 through 12 and 20 through 22 We would define a region to mask out the unacceptable bins define rh reg region from 0 to 26 26 bins contains 2 to 6 10 to 13 20 to 23 The following diagram shows how we can best v
152. instance an appropriate statement for the previously men tioned waveform digitizer would be readsize 1024 hold Hold mode can therefore be used when reading or writing Write The write keyword tells LUCID what function code to use when writing to this module The statement is essentially the same as the read statement except that LUCID will complain if the func tion code is not between 16 and 23 The same rules for hold mode transfers also apply For example write is 17 a0 9 Read and Clear When the user wants to read and clear a module with one oper ation LUCID uses the CAMAC operation specified with the read clear keyword Normally the function code is F 2 but LUCID warns you if it is not between 0 and 7 The hold mode transfer for reading applies to this operation readclear is 2 a0 15 Clear A simple clear operation may be specified by the user which is different from the clear in a read and clear operation In this case the entire module is cleared reset to it s initial state Nor mally the command is F 9 clear is 9 a0 Enabling and Disabling LAMs LUCID will try to enable LAMs in those modules which use LAMs for triggers It will also try to disable LAMs from all other modules To specify the enable LAM command the statement is enable is 26 a0 Defining the Modules A 5 LUCID will complain if the function code is not between 24 and 31 Similarly to disable LAMs
153. is hierarchy it only needs to know about a few general tasks and it will worry about most of the details for you For example you should think about the physical occurrence which will signify an ADC event and not worry about how to access the ADC in the most efficient way Getting Started Events The first thing to do is decide what you want to measure when to make those measurements then determine how those mea surements will flow through LUCID and be recorded If you re starting a new experiment data will probably originate in some CAMAC equipment You must determine how many dif ferent event types you ll need an event is simply information that is grouped together to describe something Some events may only contain a single ADC measurement while others can contain as many as several thousand distinct values The current version of LUCID limits an event to less than approximately 32 000 bytes Furthermore the rate at which data can be acquired is currently limited to about 245 000 bytes per second LUCID al lows 65535 types of events to be used in the same experiment Using LUCID Triggers Administering an Experiment In addition to the primary data for your experiment other event types might be necessary for example you might want several scaler counter values to be recorded at regular intervals A trigger simply determines when to make some measurement A particle interacting with scintillation mat
154. isualize this re gion and its relationship to the histogram named rh rh histogram rh reg region EBLTITEEEELTIEEEEEEEI TT EIS Resulting Histogram Histogram Regions Defining Histograms LOOKER Histograms 4 34 m The Looker Using Regions Region Arrays The dark areas in the region represent masks through which bin values will not pass Using this region with the rh histogram produces a second histogram containing only certain bins of in terest A region can be used wherever a histogram is used by simply giving the original histogram name the region keyword and the region s name Recall the example with the increment state ment on page 4 33 we can give a similar example using a region increment rh region rh reg using adc This statement tells the LOOKER to check the value within the adc variable and increment the correct bin in the rh reg region of the rh histogram In other words the rh histogram is incre mented as it would normally be but only if the adc value is within the range of acceptable bins in rh reg Histogram will count a value only when the value is acceptable to the region s mask H H For a complete description on how to use the increment state ment refer to Incrementing Histograms on page 4 16 for more information on how to use regions refer to Using Regions With Histograms on page
155. it can re The EVENT Section Printing Messages Making a Reader Description File 3 24 eee The Reader ally slow down the data acquisition Printing can take up a lot of computer time and even worse a lot of disk space if log files are being used On the other hand it can be very handy to know when an error occurs during the experiment or when some in teresting but infrequent event is found A variation of the print statement exists to help specifically dur ing setup or test periods The display statement has exactly the same form as the print statement but it will print the name of each variable given followed by a colon a space and the vari able s value Character strings are displayed in the same way as the print statement prints them Making Decisions about the Data It is important to perform tests on the data when processing an event This could include checking for a certain bit pattern in a coincidence register or checking the result of a calculation This is done with an if then else statement much as is used in other languages The following example could be used to check the value of an in put or coincidence register The variable coincidence has been read from a CAMAC module which has three input signals each which is associated with a different event type There will be one bit in the register corresponding to each event and the electron ics doesn t allow more than one bit to be set at one time if
156. kstation are registered trademarks of Sun Microsys tems Inc XView SunView and OpenWindows are trademarks of Sun Microsystems Inc DIGITAL MicroVAX and ULTRIX are trademarks of Digital Equipment Corpora tion X Window System is a trademark and product of Massachusetts Institute of Technol ogy ae Contents Chapter1 A Tutorial Introduction A Short Tutorial SesSsion ccccccscccscccscccccccscccccecenecscecanscecenecenenenees 1 2 Starting an ExperiMeNt sssrin isisisi 1 4 Using the Help FAC Uys ae veni diuini tr bitu ost eb pee 1 6 Building the EXpEPINIBDE sigo eR GERM RETIRER Ie tero tee rset 1 7 Controlling theAnalysis accuantium tte aaan it itd 1 11 Sending Pre defined CommandS sssseessesrissiississsesissivosesasssesiissis 1 16 Ew to QUe EUC deenen E 1 16 Where to Go from Here nenennsesesesesererererererererererererererererererere 1 17 Chapter2 Using LUCID Getinge Started MN adanadaki ias 2 1 The LUCID Prog rai reres rin REEE E EERE 2 3 Menu ButtOns ccccccccsccccesssscceesssseccesseecesssseecessseeeesesseesessseeeessenees 2 5 The Experiments Mete ipe aa ctr a a 2 6 The Build Men 0 tone s Ces t eee DRE UTR DR 2 8 Ihe Control Menu pedcs 2 12 The View Menu sese nennen 2 14 The Properties MODE oder ppt neo coco ni dE da piiss 2 15 THe Quit B tton e rtete tete ettet i Deren Rede 2 18 Control BULL eee ttt eu od itn cdi ERRARE EARRA RRA 2 19 The Record B ttOon
157. le 2 44 Window 2 44 Zoom 2 41 Zoom temporary 2 43 Histogram Modes temporary 2 43 Histogram Rebinning Xlucid 2 31 Histogram Shortcut Keys Autoscale a 2 32 Bin Numbers b 2 32 Delete DEL 2 34 Grid Lines g 2 33 Insert INS 2 34 Mode 2 34 Statistical Information S 2 33 Stepsize 0 9 2 33 Updating u 2 33 Histogram Statistical Data 2 30 Histogram View menu button 2 27 Histogram Windows 2 23 Histogram Zoom buttons 2 30 Histogram Zoom By Percentages 2 30 hv 1440 7 3 hv4032 7 3 I intape 7 3 K Keyboard Triggers Reader 3 10 Keywords increment Looker 4 17 Looker see Looker Keywords Reader see Reader Keywords L Limits Acquisition Rate 2 1 Event Size 2 1 Event Types 2 1 Looker 2D histograms 4 36 2D regions 4 39 Array assignment 4 15 Arrays 4 6 Arrays of histograms 4 28 4 29 Arrays of regions 4 35 Assignment 4 13 Assignment operators 4 16 Bit histogramming 4 40 Built in variables 4 12 Call statement 4 25 Character Strings 4 16 Character Variables 4 16 Character variables 4 16 Comparison operators 4 21 Defining functions 4 6 Defining Histograms 4 28 Index 2 ee Index Defining regions 4 33 4 34 Defining variables 4 4 Description files 4 2 Event Section 4 8 Processing some events 4 10 Function defines 4 6 Groups 4 7 Histogram acceptance tests 4 21 Histogram arrays 4 28 4 29 Histogram Bins 4 28 Histogram definitions 4 28 Histogram regions 4 33 Histograms 4 26 Histrogramming Bits 4
158. left most mouse button while pointing at the appropriate bin While hold spectrum Markers 36 to 68 f Left mouse button Middle mouse button places the first marker places the second marker Figure 2 24 Placing Markers with the Mouse ing the mouse button down the pointer may be dragged through various bins until the correct one is highlighted Once positioned the title of the histogram will display the position of the marker If bin numbers are being displayed see DISPLAY OPTIONS on page 2 29 the bin number containing the marker will be shown instead The View Menu Button The Histogram Windows 2 35 Using LUCID Accordingly the middle mouse button is used to place the sec ond marker It can be dragged as the first one Once a second marker is positioned the histogram s title will show the sum of counts of the bins surrounded by the markers Double clicking the left mouse button will erase both markers Figure 2 24 shows an example of placed markers in which the markers are drawn to the current width of the bin on the screen this particular ex ample has been zoomed in The keypad may be used to fine tune marker placement Figure 2 25 shows the additional functions available on the keypad while the program is in marker mode Recall that Stepsize is nor mally 1 and the up and down arrows have no effect on 1 dimen sional histograms Move marker stepsize E Go to previous subscript up if 2 dimensional n a oa
159. lled from a workstation running X11 There are occasions during ex periment setup and extended running when a workstation is not at hand The lucid command provides simple access when only a terminal available Options available to the lucid program are d N Set the debugging level to N The default no debug ging is 10 the maximum debug level is 0 x The current experiment is off line Default is on line i FILE The current experiment is off line with input com ing from FILE m FILE The file containing module descriptions as used when building the front end reader is set to FILE s TYPE NAME The acquisition computer to be used is set to NAME and it is treated as a TYPE computer Valid types are m147 and m167 The s stands for starburst when it was believed that was the only front end this program would ever deal with lucid s m167 hubble demo gleneGvirgo starts up connects to the acquisition experiment demo_glen on the machine virgo using the front end computer hubble The menu options shown in Figure 7 1 on page 7 8 are for an on line acquisition The commands change slightly for an off line acqui sition Familiarity with using xlucid for controlling an experi ment makes many of the command here straightforward MANAGER glen skatter is now in charge of the experiment MANAGER Using directory users3 wright Lucid xlucid 1 Build Run experiment 8 Show user defined commands 15 Offline Only
160. lues in small are defined to be 0 The LOOKER understands several variations on the assignment statement based on binary operators For example the follow ing statement adds 5 to the variable i i 5 This is exactly the same as using roe 5 Similarly to increase a variable by 7 percent you could use x 1 07 To shift the bits in an integer 3 places to the left you could use j ses 3 Character string constants cannot be involved in calculations and may only be used in a simple assignment statement into a character variable or passed as an argument to a subroutine or function For example define name char 40 beginrun name Low Energy call histdisplay Histogram 1 hist1 Character variables may used in print statements read in from the user assigned to other character variables or used as argu ments to functions or subroutines Further manipulation of strings can be done using C string functions although this ap proach should only be taken if you are already familiar with the routines and their limitations Incrementing Histograms Histograms are used to count the frequency of values within a variable and the increment statement is used to do this Con sider the following section of code define adc previous data define freq hist from 0 to 1024 event adc event increment freq using adc 1 For a complete discussion of histograms refer to LOOKER Histograms on page 4 26 The EVENT Sectio
161. m The advantage of using the LUCID data file format is that your data file will be completely self contained Variable names event descriptions and your original description files are saved to the data stream whenever an experiment is started and then every time you change a description file and rebuild This means that you could tell LUCID to generate software to read an old data tape by simply telling it which tape drive to use It will check the tape extract the original description information then write the software to read the rest of the data Examples The most common interaction with the WRITER is changing tapes in the destination tape drive This is accomplished in the following steps e Stop the current experiment but don t shut down e Terminate the tape drive writer In XLUCID toggling the SAVE state of the destination and APPLYing the change is sufficient In lucid the tape drive must be deleted from the list and the list sent to the manager e After the writer process for the tape drive exits this is in dicated by a diagnostic message remove the tape from the drive and insert a blank tape Add the tape drive back to the list of destinations In XLUCID toggling the SAVE state on and APPLYing is sufficient In lucid the tape drive must be re added to the list of output devices The most common use for multiple tapes is Data Shadowing It is only necessary to specify two output tape destinations that ac
162. mand described brief ly in The lucid Command on page 7 8 5 1 o The Writer Saving ALL Events Data Shadowing One Fail on error Saving Event Data Although it is very easy to save different events in different plac es the most common requirement of the WRITER is to write every event that it finds in the data stream Every event that the READ ER can produce is saved thereby writing all data to tape You re allowed to mix events and destinations in any fashion For example all events could be written to tape but certain scal er events could also written into a disk file Critical Data Destinations The WRITER allows data output files to be flagged as critical des tinations as set by the Fail on Error setting in the writer window When this is set for a data destination the LUCID experiment will stop when a write error occurs on output This could be due to filling a tape or disk partition to an error on the output media or an output program terminating After such a failure the ex periment must be restarted The resulting output files are possi bly difficult to use end of run summary records are missing as are the LUCID supplied end of run marks When there is only one output device Fail on Error should be set With multiple devices there are trade offs Fail on Error can be set according to the results required The idea behind data shadowing is to write the same data to two or more destinations
163. mple allocates two arrays of histograms which will count values only when between 0 and 2047 inclusive There are actually 20 separate histograms defined in this statement mak ing up two separate arrays Remember that arrays of variables are accessed with subscripts that begin at zero and arrays of his tograms are treated similarly In this same example each histogram will contain only 512 bins as specified with the keyword bins Therefore every bin will count the occurrence of four consecutive values For instance the first bin bin 0 would be incremented when a value of 0 1 2 or 3 is found Accordingly the last bin will be incremented when a value of 2044 2045 2046 or 2047 is encountered Also notice that the word histogram can be abbreviated to be hist In the following example we define a histogram with 32 bins counting values from 0 to 31 inclusive This would probably be used to count changes in bit positions define pattern hist from O to 32 32 bins Whether counting bit positions or different values within a vari able the histogram is defined in the same way refer to page 4 40 for a description of counting bit positions For convenience LUCID also recognizes definitions based on the first and last bins used in a histogram rather than the limits of variable values accepted This is better shown in the following example where the keywords first and last are used define spectrum first O last 1023 1024 bins
164. much memory If we use the example of a 1024 bin histo gram on each axis of a two histogram the result is over a mil lion counters Usually such histograms should have very few bins in each axis to conserve memory space The next example shows an array of 2D histograms using differ ent bin sizes for each axis define xsect 10 hist from 103 4 to 110 6 64 bins and from 150 to 250 binsize 1 1 The LOOKER does not allow two dimensional histograms to have more than 256 bins per axis See page 4 33for further discussion on this restriction Defining Two Dimensional Histograms LOOKER Histograms 4 38 m The Looker We see that the definition can become quite long Remember that the lines are free format so you can wrap long statements onto the following lines in your description file This example defines an array of 10 two dimensional histograms and has the first range of values divided into 64 bins the second divided into 100 The definition of a 2D histogram must have both ranges of val ues defined The bin and binsize keywords are optional and the default is 128 bins if not otherwise specified Defining Two Dimensional Regions Two dimensional histograms may also be used with two dimen sional regions as regular histograms can They re a bit more dif ficult to describe because for every region on one axis there must be a second region specified for the other axis Let s define a region which could b
165. n Incrementing Histograms Making a LOOKER Description File 4 16 dmm The Looker Corrections to Measured Data Decrement instead of Increment This demonstrates the most common use of the increment state ment A variable which was saved in an event within the data stream is checked and its value is used to increment the appro priate bin of a histogram A more elaborate example can be seen using arrays of histo grams Consider another example which uses an array of tdc val ues from the data stream define tdc 32 previous data int 2 define tdchist 32 histogram from 0 to 2048 2048 bins event tdcfired incr tdchist using tdc This example shows that an entire array of histograms can be in cremented using the corresponding values from an array of tdc data in a single statement All 32 TDC inputs will be used in sep arate histograms For more information about histogram arrays refer to page 4 29 The increment statement can use an arbitrary expression when histogramming For instance the following increment statement is acceptable to the LOOKER define n adc previous data define n adchist hist 512 bins event adcfired incr n adchist using n adc 1 09 This type of expression could be used to make corrections to de tector inputs or normalize measurements to account for unique geometry Remember that the expression can be arbitrarily com plex If the resulting value is not within the acceptable limits of the hi
166. n exact time period For instance you could specify that a special event is to be triggered every 60 minutes This special event could con tain a single endrun statement The EVENT Section Stopping or Suspending an Making a Reader Description File 3 30 eee The Reader Both statements will produce results identical to the correspond ing commands from the LUCID program In any case a user must manually restart or resume the experiment by issuing the appro priate commands from the terminal Words Reserved for the Reader Description LUCID will understand the following words as keywords when checking your READER description file It will complain at length probably cryptically when you try and use any of them as variable names LUCID considers upper and lower case char acters to be different so that n is not a keyword and can be a val id albeit confusing variable but defining N as a variable would cause a build time error message A def load B define microseconds C discard milliseconds F display minutes N elif notregion all else or and endgroup previous ascii endif print beginrun endloop program bin endrepeat read binary endrun ready bins event real binsize events rebin bits every region bool first reject boolean from repeat by group resumerun call hist save camac histogram seconds char if some character incr suspendrun clear increment then command inhibit timeout compress initialize to compressed int
167. nd last subscripted histograms in an array to be displayed respectively Pressing the Page Down PgDn key will cause the next subscript to be dis played For instance if spectrum 0 is being displayed the Page Down key will cause spectrum 1 to be shown The user can think of this as going down through the array since we usual ly visualize an array as a vertical stack of subscripted items When the last subscript is reached pressing the Page Down but ton again will cycle the subscript back up to 0 Correspondingly the Page Up button PgUp will go up in the array the previous subscript will be shown until 0 is reached then wrap around to the bottom of the array again Region Mode The arrow keys can be used to position limits of regions in much the same way as markers The most apparent difference between regions and markers is multiple areas for regions requiring a concept of a selected region area The selected area is the only part of the current region that is affected by keyboard or mouse com mands Region mode is only useful if a region has been attached to the current histogram by the Get Region menu selection The bins included in the region of a 1D histogram are indicated by shad ing the background of the histogram The region bins in a 2D his togram are outlined The selected area is ghosted to provide visual differentiation Adjusting a region can be done completely with the mouse pointer an
168. ne input or responds to only one sub address code it is still treated as an array The user need not be concerned with this however since using array is identical to using array 0 in such a case Reading Data Most programming languages allow input of data to variables in a program The read statement can be used to read a value into a LUCID variable just as the read statement in FORTRAN does The statement read angle The EVENT Section Reading Data Making a Reader Description File 3 14 OdEmER The Reader Reading Array Data Reading from a File will cause the READER to print a message on the user s terminal in the form Enter angle The READER will then stop and wait for the user to enter a value that is consistent with the definition of angle We are assuming that angle is defined as a true variable and not a CAMAC vari able or an array When prompting for input the READER simply prints the word Enter followed by the name of the variable If you would like to give the operator more information about what s expected for input you can specify a message in the read statement after the variable name For example the statement read angle the spectrometer angle 20 60 degrees would display Enter angle the spectrometer angle 20 60 degrees to the operator If the variable is actually an array element then the default prompt would indicate that fact to the user For instance the statement
169. nment must be the same size as an array on the right side If there is no array mentioned on the right side of an assignment then each element of the left hand side array will be set to the same value CAMAC variables are not accessible after a read and save oper ation If such data are to be used in calculations the read and save operations must appear in different statements Printing Messages During Event Processing It is necessary to display messages usually during an initial set up period and occasionally when an unusual type of event is en countered Because printing can slow event handling significantly message printing should be limited to special cir cumstances LUCID allows you to print messages with the print statement Variables and character strings may be printed in any combina tion and are separated by commas as shown in this example print Signal from scintillator scin value was tdc scin In this example the variable scin contains some input register value and tdc is an array of time values from a CAMAC mod ule The resulting message might say Signal from scintillator 3 value was 511 The message is sent to every screen which is currently using the LUCID command and will be displayed in their message areas If there are no LUCID programs running the messages are placed in a log file As mentioned the print statement is not meant to be used when common events are encountered because
170. nnel is read out For instance the LeCroy model 2259 Peak Sensing ADC has twelve inputs and will issue a LAM when it has finished conversion and is ready to be accessed When the last input has been successfully read out the module Defining the Modules A 6 od mH clears it s own LAM it essentially sends itself and F 10 com mand To make use of this feature you may use the autoclear statement within the CAMAC module database file This tells LU CID not to clear the module if the user causes the last channel to be read out Q Stop Modules Some modules make a variable amount of data available at a time and indicate the end of the data transfer by a change in the module s Q state The keyword qstop indicates this The event data saved from a Q stop module is automatically compressed as described in the following section Self compressing Modules Modules such as Q stop modules that provide a variable amount of data each event are considered self compressing mod ules The CAMAC Module Database provides a good deal of flexibility in defining the format of the data for the two most common cases a simple list of values and a list of channel value pairs The statement used for both these cases is channel length CL offset CO value length VL offset VO The lengths and offsets of each field are specified as bit positions counting from the least significant bit The offset bits are counted starting with the least significan
171. ns on analysis in this case for instance he is allowed to have the LOOKER ignore some of the data in the interest of speed The Play button tells the READER to replay data which has been acquired earlier The data stream is off line in this case and the READER will pass the data through a pipeline from which the LOOKER and or WRITER can examine all of the data When you ask LUCID to Make the software it determines if the data stream is on line or off line and will only allow you to use one of either the Record or Play buttons Whether on line or not the LOOKER will build up histograms and perform other tests and calculations when new data arrives possibly calling external subroutines that you ve written before hand You can interact with the LOOKER via the lucid program which allows you to view any or all data and change the LOOK ER s data tests dynamically New data such as histograms can be included in the data stream for the WRITER to save The WRITER records the data after everything else has finished Data can be converted to an alternate format and saved to a disk file magnetic tape or passed as input to another computer pro ee Conceptual Overview Reader gram or any combination of places at the same time For in stance data can be written to a tape in LUCID format When data is saved by the WRITER using its default format the names of all LUCID variables and data types are also saved In other
172. nstalled in the first sockets the command set 0 6 5000 will set voltage channels 0 2 4 and 6 to 5000 volts The command set 1 7 300 will set the corresponding current limit channels Oc 2c 4c and 6c to 300 microamps display backup N M display demand N M display measured N M ilimit ilimit value Display the backup demand or measured value of channel N through M a single channel number N is also acceptable The display keyword can be omit ted The demand and measured keywords can be abbreviated to their first character If no channel ar gument is specified the value or values from the previous display command will be used If no value is specified this command displays the positive and negative current limit settings If a val ue is specified the command set the appropriate cur rent limit according to the sign of the value The current limit values are in units of approximately 10 microamps This command has no effect on 4032 mainframes save filename Save the demand and current limit values in the specified filename The name from the previous save command is used if no filename is specified The default file name is hv save Utility Programs 7 6 OdEHMBI qview Utility Programs and Subroutines quit Stop processing this file Forces termination if the end of the file arguments has been reached or if the program is in interactive mode qview Qview is similar to lucidview but instead work
173. ntire array each value should be separated by blank spaces tab char acters or appear on different lines As with single variable files LUCID will ignore the rest of the file after it has retrieved enough information to fill the array Notice that this option contains the keyword rom which tells LUCID that the name in quotes refers to a file rather than a prompt for interactive reading Only one array or variable can be read from a file The second option allows you to use the output from some pro gram as the value to use for the variable For instance the state ment read adjust from program home wright bin getadjustments when executed tells LUCID to start the home wright bin getad justments program running on the computer which is running the READER LOOKER and WRITER LUCID takes the first piece of information output from the program and places it in the named variable If the variable is an array name then LUCID will try and get the correct number of values from the output and as sign each one to successive array elements The program will output the values separated from one another by blank spaces or tabs or place each value on a separate line As with filenames the program name must refer to a complete filesystem pathname and you must have permission to run that program Within quotes you specify the program name just as you would if you were sitting at a terminal For example a col league might have
174. ocal variable cmpvar offset contains the offset in bytes from the start of the event record dp where compressed data can be recorded At completion of the user code cmpvar_offset must point to the first byte after the end of the saved compressed data that is on a LUCIDALIGN RECORD boundary The event record variable dp Vaverage in the above example is not an array but rather a structure of type Lucidcompressoffset The 1co begin field must point to the original cmpvar offset value and the 1co_end field must point to the first byte past the end of the compressed data Any subroutines that were included in the User defined Code section can be accessed only through a program statement Stopping or Suspending an Experiment An experimental run is usually stopped by a command from the user at a computer terminal He might decide that there is enough of a data sample to stop then start a new run or might want to temporarily suspend the run for some reason LUCID al lows a run to be stopped or suspended from inside an event For example the following event statement will cause the current ex perimental run to stop after this event is processed endrun No further triggers will be handled and no other events per formed Similarly the following statement will cause the experi ment to be suspended suspendrun These statements allow an experimental run to consist of a pre cise number of events or to contain data acquired over a
175. of the value 125 Sji SN a first bin has encountered the value 100 4 times Number of values encountered The definition says that only the values between 100 and 125 in clusive are important and values greater than or equal to 126 will not be accepted It seems apparent that most of the values that were checked so far were closer to 100 than 126 because those bins have higher counts We can also tell that 98 acceptable val ues were encountered because that is the total number of counts in all bins Number of Counts Bins can count as high as 4 billion before they wrap around to in each Bin 0 If you think that your experiment might be very fast or lengthy such that it will check billions of values then you may redefine your histograms to count to a higher range Typically a bin is implemented as a 32 bit counter but the LOOKER also al lows the bins to be real numbers The following example shows how this is done define pions real 8 histogram from 100 to 126 26 bins Keep in mind that the most efficient counter is the default which the LOOKER provides Any other type of bin is rarely needed Histogram Arrays You re also allowed to have arrays of histograms define d1 10 d2 10 hist from 0 to 2048 512 bins 1 The default bin is of type int 4 Histogram Definition Defining Histograms LOOKER Histograms 4 29 The Looker Specifying Histogram Bins Instead of Limits This exa
176. of this form of the camac command is discouraged as it is harder to track down and change the statements when a module is moved Camac Crate Operations Certain CAMAC modules allow you to stop taking data momen tarily by using a signal on the CAMAC dataway The inhibit I signal is passed to every station in a CAMAC crate and is normal ly in an off state Some modules will suspend their actions when this signal becomes true and restart when the signal becomes false again LUCID allows you to set or clear this inhibit line from your READ ER description file Before using it you should carefully check the hardware manuals for every module within the crate Some modules react differently to the inhibit signal which is sent to every station The EVENT Section Specialized Camac Access Making a Reader Description File 3 28 OdEmER The Reader To set the inhibit line use the following command dataway inhibit If the branch or crate are not specified the branch is assumed to be 0 and the crate is assumed to be 1 A more general form of the command allows the inhibit signal to be set in different crates dataway C 4 inhibit Specifying crate 0 gets the entire branch at once Similarly to clear the inhibit signal and resume module opera tion use dataway uninhibit In addition to inhibit and uninhibit the dataway clear and dat away initialize operations are also permitted User written Code The situation
177. offset O0 longword read is f0 a0 test is f8 a0 clear is f9 a0 disable is 24 a0 enable is f26 a0 module LeCroy 2323A Dual Gate and Delay Generator 2323A LeCroy 2323A readsize 2 shortword read is f1 a0 1 clear is f9 a0 1 write is f17 a0 1 module LcCroy Model 23418 16 fold coincidence register pattern unit 2341S LeCroy 2341S readsize 1 Sample CAMAC Module Database A 11 shortword read 0 a0 readclear f2 a0 clear 9 a0 2 The LeCroy 2373 MLU must be placed in INHIBIT mode and the CAR register must be written before the memory pattern can be written or read e g dE db H dt camac mlu A 2 F 16 data 0x50 camac mlu A 1 F 16 data O load mlu from tmp foobar mlu camac mlu A 2 F 16 data 0x40 module LeCroy 2373 Memory Lookup Unit LeCroy 2373 TEURS 2373 23 3 readsize 65536 hold writesize 65536 hold shortword read is f0 a0 write is 16 a0 module LeCroy Model 2551 12 channel 24 bit scaler 255 LeCroy 2551 readsize 12 longword autoclear read is f0 a0 11 readclear is f2 a0 11 test is f8 a0 clear is f9 a0 disable is 24 a0 enable is f26 a0 module LeCroy 3511 3512 Single Channel Spectroscopy ADC 3511 3512 LeCroy 3511 LeCroy 3512 autoclear readsize 1 shortword read is f0 a0 readclear f2 a0 write is 16 a0 enable 26 a0 disable 24 a0 Sample CAMAC Module Database A 12 eee testlam f8 a0 clearlam 10 a0
178. om 0 to 2048 define tdcspec 96 hist endgroup endgroup All variables would be accessible by using allvars Only the his togram variables would be accessed when nistonly was used Although unlikely to be useful one could increment all histo grams at once with the same variable increment histonly using x Another convenient use of variable groups is in the load state ment Rather than reading a single variable at a time or listing out all the variables to the load statement a group name could be used Remember that all variables are still accessible by referring to their names regardless of which groups they are contained in The EVENT Section The event section of the LOOKER description file is usually much larger than the define section because it tells LUCID what to do when a particular event is found in the data stream Event Types The section consists of one or more event specifications one after another in the description file Each specification may contain many statements but the first line must contain an event name similar to a subroutine name in a programming language For example the statement event electron The EVENT Section Event Types Making a LOOKER Description File 4 8 w The Looker associates all subsequent statements up to the next event speci fication with the event name electron The statements which follow in the file will be performed whenever an electron event is encountered The
179. or stop the data stream will still keep running without it when the experiment itself is running elsewhere Selecting the Quit button will cause the LUCID program to stop but keep the data stream active Selecting Shutdown will make LUCID shut down all of the data stream processes and then quit The Stay button will cancel the alert window and let you continue using LUCID Where to Go from Here Now that we ve seen how simple it is to use LUCID you should go ahead and set up a real experiment You re certainly welcome to play around with the demo experiment that was just set up but take note that all demo experiments are periodically re moved If you re just getting started with LUCID you should find out what kinds of statements the description files can contain These are described in Chapters 3 and 4 Chapter2 describes the XLUCID program in more detail in cluding how to start up and administer an experi ment It explains how to control the READER LOOKER and WRITER and monitor the progress of data through the data stream Chapter3 introduces each of the statement types that you ll need to describe your experiment to the READER mainly in terms of the equipment you have Chapter4 explains how to set up a LOOKER description file and describes histogramming and analysis in gener al Chapter5 talks about WRITER considerations Chapter6 gives an overview of how all the pieces of LUCID fit together
180. or type When off line the name of the file containing the data must be entered Tape drive units are named as files and are typically of the form dev nrst0 When reading from tape remember to include the letter n after the second slash mark which indicates that the no rewind mode should be used The n ensures that when finished the sys tem will not rewind the tape thereby saving a lot of time Also LUCID understands remote tape drives you can specify a tape drive on a remote workstation by using a name of the form reg ulus dev nrst0 The computer name and a colon precede the tape drive name In any case the user should check with the sys The LUCID Program 2 8 od mE Using LUCID Build Local Variable Build Experiment demo glenGvirgo Name junkdata CAMAC Device mJ67 hubble Data File Copy From Type v int 4 Size Array Description Files demo Use READER Edit demor Use LOOKER Edit demo l Build Options Class v 1 D Histogram L1 Region X Limits From 0 To 1024 Bins 1024 4 v Y i nits fronb gins Re Build Custom Hardware Database Experiment pion_gw virgo Database File 4 ght Lucid camac dbase Create Cancel Build Experiment Build Local Variable Remove Local Variable Use Hardware Database Writer kel WRITER OUTPUT FILENAME j SAVE vega dev nrsti kel Remove Local Variable
181. ossibly months after the experiment is finished You must also avoid using keywords for your variable names such as read or define This can really confuse LUCID so make sure you read Words Reserved for the Reader Description on page 3 31 A variable name should not be longer than 32 letters and the first character cannot be a digit An underscore can be used as part of a variable name and lowercase letters are considered to be different than uppercase ones There is no preset limit on the size of arrays or on the number of variables that you use other than the amount of memory avail able on the associated computer Large amounts of space could affect data throughput since buffers share the same memory as your variables In practice however the definition section con The DEFINE Section Defining Variables Making a Reader Description File 3 7 The Reader Definition Examples Trigger Types sists of as many definition statements as you care to include Keep in mind that although an array may be very large you ar en t allowed to save more than approximately 32 000 bytes in any one event A point to remember is that all LUCID variables are guaranteed to be zero when the program starts This doesn t mean they will be reset to zero before every run but they will be zero before the first start command is issued A few sample definitions are listed here define energyl integer 2 define energy2 real 4 define pil pi2 pi3
182. outines simplify the mapping of directory paths on different computers awd pathname fills in pathname with the name of the computer and the path on that computer re quired to get to the current working directory dwa rempath pathname takes a remote path specification as returned by awd and changes it into the proper path name for the current machine to reach the original path Awd stands for Absolute Working Directory and dwa stands for Directory from Working Absolute The fact that one name is the reverse of the other is entirely coincidence Reading LUCID Data Files On occasion it may become necessary to write data manipula tion software that reads a LUCID data file The following rou tines simplify this Opens the named file for reading An error is returned if the file name is not valid but no checking is done to ensure that the file contains valid data Accessing Networked Directories LUCID Functions and Subroutines 7 10 d Utility Programs and Subroutines Lucidfdopen fd Initializes for LUCID I O on the argument file descriptor No checking is done to ensure that the file descriptor references a LUCID data file or that the file is positioned at a proper location Lucidread Returns a pointer to the next LUCID record On an error a record of type LUCIDerr is returned On end of file a NULL pointer is returned Lucidskipdata Returns the next non data LUCID record Error and EOF condi tions are the
183. oving an area can be quite slow and it can take many tens of seconds for XLUCID to catch up with drawing the dragged area The CUT key deletes all selected areas The COPY key creates an unselected area at the same location as each selected area The PASTE key unselects all selected areas Send Region Changes made to a region do not take effect in the looker until Send Region is selected from the histogram menu Regions can also be adjusted using the keypad although this currently provides less functionality than the mouse Keypad functions for 1 dimensional histograms are shown in Figure 2 28 and keypad functions for 2D histograms are shown in Figure 2 28 Scale Mode Shortcut Key F4 Scale mode can be used to zoom in on the data This mode is the same as zoom mode for 1 dimensional histograms For 2 dimen sional histograms scale mode allows the user to adjust the scale of the data that is to adjust the Z axis Figure 2 30 describes the functions available from the keypad while in scale mode The View Menu Button The Histogram Windows 2 39 Using LUCID Select First Select Previous ic Area Aa CN Shift Selected Area 7 9 stepsize bins to the Home PgUp right on the display Shift Selected Area 4 5 6 stepsize bins to the gt 4 gt left on the display Toggle from markers 1 3 End PgDn Select Last m Select Next Area Area Figure 2 28 Keypad Functions in Region Mode 1D Shift Selected Ar
184. own the experiment after it is disconnected that is terminating any of the READER LOOKER and or WRITER piec es that may exist As shown in the previous diagram Disconnect is inactive if LUCID is not connected to an experiment The Build Menu Once connected to the desired experiment the user will typically go ahead to use options beneath the Build menu This won t be the case if the experiment is already in progress when the con nection was made In such a case someone has already built the experiment and the control buttons will all be active Sub windows which originate from the Build menu are shown in Figure 2 4 The default option under Build is named Build Experiment A sub window is presented which allows Editing of description files naming of the input file or hardware to use if on line and ultimately generation of the software This window will be used every time description files are prepared or used to generate software The user must first decide whether the system is on line or off line and select the corresponding button near the top of the window as shown in Figure 2 4 On line or off line a filename is required to name the source of data When on line the filename refers to the location of the pre processing computer see Chapter 3 and is of the form processor type amp hostname This indicates that the preproces sor is the VME processor named hostname which is a proces sor board of type process
185. ows the general format define x int 4 beginrun X 1 print Now starting Run x Each item in the print statement must be separated from the next item by a comma and character strings constants and variables can be printed in any order Remember that using an array with out a subscript in a print statement will print the entire array If the user is not accessing the experiment when the print state ment is performed it will not be displayed but saved in the ex periment s log file The print statement should be used with caution since each event specification will be performed when the appropriate event is encountered in the data stream There could conceivably be thousands of events processed per second which implies thousands of messages printed per second Messages can also be printed directly to a file or to a user pro gram define time int 4 int 4 print Run LUCIDrun at time 0 to runlog print LUCIDrun to program home exp000 bin exparam Histograms can be saved to files using the write statement There is little different between the write and the print state ment write requires that the output be directed to a file or pro gram and closes that file immediately after writing variables saved by write are in a fixed predefined format print allows constants and expressions to be saved write only allows one variable per statement to be written The syntax of the write statement is write compresse
186. p date the view move the mouse arrow into the histogram win dow and press u The mouse arrow should always point to the window that you want to adjust Recall that the spectrum variable was actually defined as an ar ray of 12 histograms The title in the upper right corner of the window shows the name but also the range covered by the X and Y axes The word Auto appears beside the Y axis indicating that the Y axis is being redrawn automatically so that the largest data value will be visible This particular view shows the first histogram in the array spec trum 0 Press the Page Down key on the right side of the key board and the next element in the array will be displayed Correspondingly Page Up will cause the previous element to be displayed and Home and End will display the first and last ele ments respectively Most of the interesting data for this tutorial is located in the eighth histogram that is spectrum 7 Press the Page Down key a few times until it is displayed A Short Tutorial Session 1 13 A Tutorial Introduction A common requirement is that several different elements be seen at the same time In fact it is sometimes important to view the same element several times in slightly different ways By press ing the Insert key also located on the right side of the keyboard another view of the histogram will be inserted spectrum File v View v Edit v Print v Update Zoom
187. pion 40 integer 2 define timel LeCroy 2229 N 10 A 6 11 define counts lrs4448 N 11 The TRIGGER Section Normally we think of an event as some happening or occurrence which causes a result LUCID defines an event as a collection of data items When data is recorded on tape for example it is basi cally a list of thousands of events one after another A trigger is defined as some occurrence which causes the event to be pro cessed If the source of the data is a CAMAC module then a trig ger is often a CAMAC LAM Triggers must be described in the second section of the READER description file The section contains statements which associate some occurrence with the action of acquiring the data There are several types of triggers that LUCID can respond to e CAMAC LAMS interrupts CAMAC modules which set a ready status bit This im plies polling for a data ready state e Some amount of time passing Acertain number of valid events occurring Executing a trigger statement in an event specification A change in LUCID s state such as the start or end of a run A request from the experimenter e Some external source making data available The syntax of the trigger definition statement is shown in the fol lowing example trigger electron arm every delta e lam or every 3 seconds The TRIGGER Section Trigger Types Making a Reader Description File 3 8 lime The Reader The Trigger Statement LUC
188. ptions to Save and Restore Window Positions can be used to tailor the look of lucid on the screen when it is started Sub windows should be moved and resized as desired and then their positions saved They will be restored to these positions and sizes when lucid is next started This does not save the char acteristics of histogram windows The View Menu on page 2 14 describes how to save histogram windows The Quit Button The Quit button on the main LUCID window is meant to be used to quit the lucid program itself Ideally users will start the exper iment monitor its progress for a few minutes then quit the pro gram while the experiment continues You can QUIT without affecting LUCID it will stay running without you You can SHUTDOWN the experiment You can change your mind and STAY with it Figure 2 10 The Quit Alert Sub window Selecting the Quit button brings a request for confirmation from the user Choosing Quit is the default choice the user can simply press return to select it The lucid program itself will quit but the experiment will remain whether it is currently running or not The LUCID Program 2 18 w Using LUCID e The Shutdown button will cause the READER LOOKER and WRITER to finish the current run if necessary quit and close down the experiment completely until next time e The Stay option tells LUCID to keep working as if nothing has happened If lucid is not connected
189. put LUCID s knowledge of an experiment comes from the user s de scription of it the user is required to make up description files for each of the READER and LOOKER The experimenter starts and controls his experiment with the XLUCID command which allows interaction with the READER LOOKER and WRITER and lets the user change important parameters dynamically lu cid loo cid 1 bright shining 2 transparent 3 designating an interval of sanity in a mental disorder 4 clear to the mind readily understood ucid instructions 5 clearheaded rational a ucid thinker luscid i ty lu cideness n lu cidely adv A Tutorial Introduction A Short Tutorial Session The best way to learn how to use LUCID is to sit down and try it You need a few things to get started First make sure that you have an account on the computer system You ll be given a log in name and a password which will let you access most comput ers in the Laboratory network Second find an available work station and type in your login name and password To use LUCID you ll need to start a window system called OpenWindows The system administrator might have set up your account so that windows come up automatically If not the com puter will ask what type of windows you want and you should select OpenWindows In any case your screen will look some thing like this Move the mouse pointer into this window then type demolucid and p
190. r all the LUCID events declarations for all the READER variables requested by the user defines for the event num bers and the maximum event size prefix m167 the executable version of the VME on line acquisi tion reader This program is downloaded to the VME front end computer prefix m167 0 Intermediate file created when the prefix m167 program is generated prefix c a copy of the LOOKER C routines written by the user How LUCID Works 6 5 o ems Conceptual Overview prefix f prefix_C o prefix_F o user a copy of the LOOKER FORTRAN routines written by the user an intermediate object file generated from the prefix c file an intermediate object file generated from the prefix f file a file listing the account names of the people al lowed to access the experiment on the current computer The first name in the list is the experi ment administrator If any of the other names is any body then any user with an account on the acquisition computer can connect to the experi ment and control it Programming Languages Used LUCID was written with the C programming language and it produces C programs to do your experiments A compiler writ ing program called YACC was used to understand the descrip tion files and generate the corresponding programs YACC is also written in C The LOOKER allows subroutines written in C or FORTRAN to be called while it is running Programming Languages Used
191. r holds down the right meta key Therefore right meta Home would reset the Z axis limits regardless of the cur rent program mode Typically keyboards have 2 meta keys immediately on either side of the space bar as shown in Figure 2 33 The figure shows E OOC OES A S eO l Left Meta Scale Mode EET NN 2 N Right Meta P The View Menu Button Zoom Mode Figure 2 33 Modifier Keys for Temporary Modes that temporary zoom mode is enabled by holding down the left meta key This makes sense when one remembers that scale mode and zoom mode are quite similar indeed identical for 1 di The Histogram Windows 2 43 Using LUCID The Edit Menu Button mensional histograms Temporary pan mode is enabled by hold ing down the Alt key Window Mode This mode is currently not implemented Many of the intended functions for Window mode have been moved to the View win dow Title Mode This mode is not completely implemented Currently it only al lows the mouse pointer and arrow keys to position the main title This is intended to help histogram printing Eventually the user will be able to highlight the display with textual messages and simple graphics The Edit Menu Button The Edit Data Option Currently not implemented The intention of this window is to allow the experimenter to modify data in looker variables but not in the data acquisition stream The Density Map Option This feature
192. r the user At this point it is usual that the LOOKER or READER have failed already The View Menu One of the most frequently used features of the system is the abil ity to display data The View menu has two options which bring up sub windows as shown in Figure 2 8 The first from the View Data option displays a list of variables which can be viewed These are all variables defined in the LOOKER except regions see Chapter 4 Four abbreviated menu buttons are displayed near the top of the window which can be used to limit the variables being displayed In a large experiment having hundreds of variables the user might ask that only ar rays of two dimensional histograms be listed to make the selec tion easier Regardless of what variables will be put in the list the user can select as many of the variables as is needed These vari ables will be displayed when the Display Selected Variables button at the bottom of the sub window is selected When the user is finished displaying certain variables their names can be de selected from the list and the Display Selected Variables button selected again All of the other variables will still be displayed but the de selected ones will be removed The LUCID Program 2 14 odmEEE O Using LUCID View Data Experiment Variables to List Source All Lucid Messages E Types All Sizes All Class Histograms Variable List davey davey cut spe
193. rak 6 4 Programming Languages L Sed iaasnns pie enia vic iin been 6 6 Chapter7 Utility Programs and Subroutines Utility PRG ATS esit boues aaa bei Leiva tradi 7 1 classi D 7 1 COPY CUD M X 7 1 demol cid zitierter EE EE 7 2 Daea N D aP eser roe A AEE Dr E rue EN ERN A A AEA EEN 7 2 Had Olea a tete OE aca 7 2 BENGTALG E 7 2 lodi EEEN EEEN OEN E ITAA PO EE E 7 3 E T J EA TO NEE E E E AO AE AE A E AT 7 3 jr M 7 3 NAD a a EEEE E EEE EEEE TE ETE EE E EEE E E 7 3 ure Mu e Giran MM 7 3 littere lio M 7 3 Tacidanatr ses hills eed Ra tee Ae ee ee ON gon ios INE hence 7 4 MUA CLA LEW aR E E EAEN 7 4 netcatma c su e ev eren eta A EEN AE AE ORAN 7 5 netcratelnit eese ete ttes eee ee ees Lese ee een bra Dr e NEUE Ue ruo 7 5 Tiel littere tuse teme tese teli crest cine bre ene dres reset eset erede end 7 5 nl Se M 7 7 OAS AL A cone tunt EE Mes 7 7 vine console echar ise tesi ees 7 7 Lucid User s Guide 1V ee Contents The lucid Command eee eene eren ens 7 8 LUCID Functions and Subroutines eee 7 10 Accessing Networked Directories ioo aoi seis assa lias 7 10 Reading LUCID Data Files quaque HIPS Oe iiis Rock Ml 7 10 Writing LUCID Data Records sireenin araia RH RI aa Raa 7 11 Generating LUCID Data Fil s stie opaak intimo bano 7 11 Low L
194. rbitrary resolution it is only important that the region and histogram have the same number of bins Actually bins can be defined without limits and can therefore be defined like this define validbins region 26 bins This validbins region can be used with any histogram having 26 bins Many programmers include limits in region definitions specifically the same limits as are used in their associated histo grams This makes it easier to see which will be used together Defining Two Dimensional Histograms Histograms can be very useful when analyzing data They are quite simple and essentially describe all of the changes made to a variable We often need to know more information about a variable particularly with respect to other variables For exam ple we might want to count the occurrence of different values only when some other variable has a value between say 10 and 20 In general terms we want to count the occurrence of different values in two different variables at the same time Defining Two Dimensional Histograms LOOKER Histograms 4 36 wo The Looker Example As an example let s consider a variable called energy which can contain values between 10 and 19 inclusive A histogram of this variable would be very simple to define since it would con tain only 10 counters Now assume that every time this variable changes to a new value a second variable delta e also chang es and it can contain values from 0 to 2047
195. re numbytes in event unsigned long user defined data and offsets into compressed data compressed data variable length event type unsigned short The number of bytes in the event does not include itself only the remainder of the event structure For a given event type this size will differ from one record to another only if data compression is used See struct Luciddata below written when an endrun trigger is processed It is useful to in clude this record for time information rather than depending on the following beginrun record especially for the last run of the experiment which won t have one See struct Lucidend below Written whenever a privileged user wants to write a comment to the data stream Usually included by default at the start of an ex periment and at the beginning of every run See struct Lucidcom ment below Padded with newlines to a 0 mod LUCIDALIGN_RECORD boundary Used for padding Specifically when an experiment ends or the output block must be flushed for some other reason the unused remainder of the block is filled with null bytes contained in this record See struct Lucidnull below LUCID Data Record Format Declarations LUCID Data File Format 7 17 o Utility Programs and Subroutines LREC_ERROR Used to indicate error reading data Only the Ih magic structure element is valid All other structure entries are zero define LREC FORMAT 0x553001AAL define LREC SYMBOL 0x5
196. re should be ignored Do not use Creates a comment record and embeds it in the acquisition stream At the time of writing this is the only way to create a comment record Modifies the user list of people who can access the experiment This does not handle the anybody user that matches all users Controls setting the debug level for the various parts of the LUCID system Shuts down the experiment and exits the lucid program OFF LINE Opens an alternate data file for input OFF LINE Starts the playback of the current data file OFF LINE Stops the playback of the current data file OFF LINE Rewinds the input data file OFF LINE Skips forward to a specified run number OFF LINE Closes the input data file The lucid Command 7 9 o Utility Programs and Subroutines 20 Allows setting up writer devices and event lists for writer devices A separate menu is shown for this option which uses Command 20 r g s p d a e read event name list get destination names gave destination names print destination info print event names delete destination add destination edit destination info 0 9 aun Hh 1 Figure 7 2 Writer Destination Menu a character menu rather than a numeric menu C Quits the lucid program without shutting down the exper iment This can be entered at any time LUCID Functions and Subroutines Lucidopen file Accessing Networked Directories The awd and dwa r
197. read adjust 0 would cause the READER to prompt with Enter adjust 0 If the name of an array is given without a subscript the READER will prompt for each element s value in turn until each one has been read Enter adjust 0 Enter adjust 1 Enter adjust 2 e e e Although LUCID allows you to enter certain values interactively as your experiment proceeds this isn t always convenient if you re not sure when they will happen or if an inexperienced colleague is filling in for you LUCID offers two flexible options for reading values into variables The first option allows you to read values from a file The state ment read adjust from home wright adjustments The EVENT Section Reading Data Making a Reader Description File 3 15 The Reader File Format Reading from a program s output Program Options will cause the READER to get the data from the named file rather than from the user s terminal This allows large arrays to be read directly from a file while an experiment is running The named file must contain a filesystem pathname and refers to a file on the computer on which the experiment is running as specified as part of the experiment name not necessarily the computer the LUCID command was run on Naturally you must have per mission to read that particular file LUCID will use the first value in the file for the variable so the rest of the file could contain anything If you re reading an e
198. ress return a ED E HE eee eee PEHEE A inn 12 BEE HH EI al cJ z E File Manager tmp mnt users3 wright file v Vlew v Edit v Props v Home v Cote v Om rea cclib COMM Ostaspectum7 Dataspecrum 2d GNUinclude T N A Short Tutorial Session o o A Tutorial Introduction tt After typing demolucid you ll have to answer y to do you want to create a new experiment y and then wait for a few seconds until the LUCID window appears Your screen will look like this g Lucio h 4 Li 4 i i mel 2 iow v Ceropertios v Quit FEEEEEELELEETELLH CO FH EH DIENEN i HEHH In 88 mE E COCO Move the mouse pointer into each of these other windows and press the button labelled Open located on the left side of the keyboard The windows will close into icons HE HO E m HH ER HHHH HEHEHHE You would get the same result if you just typed xlucid but demolucid installs some small files which are required for this demonstration We ll find out more about these files later We see in the window that LUCID has s
199. rmine if it is a bin number the y value representing the low est edge of the bin or the y value at the midpoint of the bin The value field is always present and is the count for the current bin Histograms that are saved as binary have the same fields as ascii saves under the same circumstances with the major difference that the array index x bin number and y bin number are all saved as float values real 4 and the value field is saved accord ing to the type of histogram which is usually long int 4 Overview D 3 Overview D 4 ae INDEX A addclient 7 1 Adjust View sub window 2 27 B Building an experiment 1 7 Built in Looker variables 4 12 C Calculations on Looker Variables 4 13 on Reader Variables 3 23 CAMAC Clearing modules 3 19 Crate Operations from Reader 3 28 Crate uninhibit from Reader 3 29 Hardware Layout 3 1 Loading file data 3 27 Module polling 3 20 Modules Database 3 5 Operations from Reader 3 28 Polling modules 3 20 Read and Clear Operation 3 19 Reading variables 3 17 Saving Data 3 18 Setting inhibit from Reader 3 29 Using LAM s in Reader 3 2 Writing from Reader 3 26 Coding user routines Reader 3 29 Command Line Options 2 4 Compressed Data Reader 3 21 Constants in Reader 3 13 Control buttons Xlucid 2 19 COPYRIGHT NOTICE 1 copyrun 7 1 D Default Experiment Directory 2 16 Defining functions Looker 4 6 Defining variables Looker 4 4 Reader 3 6 demolucid 7 2 Density Map Xluci
200. rs are discussed in more detail on page 2 32 and again on page 2 35 In the Adjust View sub window the first set of choices allows the user to attach different meanings to mouse actions and key strokes By default the system is in Marker Mode as described be low Using the mouse buttons and keypad is discussed in more detail under Shortcuts for the View Menu on page 2 32 but it is important to realize that different modes will cause the system to react differently to commands Marker Mode causes the mouse pointer and buttons to affect the markers on the screen The keypad on the right side of the keyboard also becomes used to locate markers in this mode This is the most common mode for interact ing with the display Region Mode causes the user input to adjust Regions which are discussed under Histogram Regions on page 4 33 Briefly adjusting regions on the display will allow your analysis to perform different instructions At the time of writing this document the Region Mode is oper ational for 1 dimensional histograms only e Scale Mode causes the mouse pointer and other controls to change the scale of the displayed data for 1 dimension al histograms user actions will adjust the height of the bin data the Y axis being displayed Zoom Mode allows the user to use mouse buttons and keystrokes to affect the viewed limits of the display and effectively enlarge a particular area of interest Pan Mode causes the mouse an
201. s again in the LOOKER is good pro gramming practice but is not required A good compromise is to define one or two commonly used variables from the READER Defining Histograms A histogram definition might appear as define tdc hist histogram from 0 to 256 which creates a simple one dimensional histogram A two di mensional histogram is defined similarly such as define tdc hist histogram from 0 to 256 and from 1024 to 2048 By default histograms are created with each bin as type inte ger 4 This data type can be changed for instance if a histogram could have bin values beyond the range that this data type holds roughly 4 billion counts per bin then the histogram could be defined as define tdc hist histogram real 8 from 0 to 256 1 A more complete description of Histogram Definitions appears in LOOKER Histograms on page 4 26 The DEFINE Section Defining Histograms Making a LOOKER Description File 4 4 m The Looker The type of the histogram bin may be any of integer 2 inte ger 4 real 4 or real 8 The default number of bins in a one dimensional histogram is 1024 and 128 for each dimension of a two dimensional histo gram When this is inappropriate the number of bins can be specified as part of the definition define tdc hist 2D histogram from 0 to 256 1024 bins and from 1024 to 2048 16 bins The range of tdc hist 2Dspecified using from and to are inclu sive on the from value but exclusive of the
202. s and Subroutines Would produce the data record event size in bytes padded to modulus 4 v1 offsets for v2 v3 offsets for v4 v5 variable length v2 data variable length v4 data event number There may be padding between any of the entries to enforce LU CID_ALIGN alignment Information on the form of the compressed data shares space with the histogram info in the Lucid symbol table structure The information includes the bit offsets and lengths of the channel and value components of the compressed data Bit offsets are little endian after the variable has been converted to native for mat A bit offset plus a bit length can not exceed the number of bits in the underlying data type The following examples show the format in which compressed data are stored the lv type of the symbol table entry and the compression information value LVAR SHORT channel bit offset 0 channel bit length 0 value bit offset 0 value bit length 16 For a module such as a waveform digitizer or completely user defined data chan value LVAR SHORT channel bit offset 12 channel bit length 4 value bit offset 0 value bit length 12 LUCID Data Record Format Declarations LUCID Data File Format 7 20 pem oo Utility Programs and Subroutines This is similar to the old style LCMP SHORTSHORT compres sion value LVAR SHORT channel bit offset 0 channel bit length 16 value bit offset 16 value bit length 16
203. s and al ways aligned on a bin boundary Marker values need not be within the current limits for X Y axes they will be positioned on the correct bins but just not visible on the display Pressing Clear Markers simply removes the corresponding X or Y markers from the display Double clicking on a histogram in Marker mode clears both the X and the Y markers The Set To Limits button allows the user to set markers at the current limits of the viewed data That is markers will be placed on the first and last bins which are currently visible in the sub view With a 2 dimensional histogram markers appear visually as a pair of lines one marking an X axis bin and the other marking a Y axis bin The result is that only 1 bin is selected as represented by the intersection point of these two lines An area between two markers is therefore the rectangle enclosed by the four lines as shown in Figure 2 22 The data within the bins of a 2 dimension al histogram are represented on the screen as smaller rectangles which are filled to various degrees For instance if a bin con tained as much or more data as specified by the upper Z axis limit then the bin s rectangle would appear to be filled If the bin contained very little data as compared to the lower limit then the bin s rectangle would be empty A rectangle which contains a filled area of 50 means that the bin contains roughly half the data which the Z axis limits include Shortcuts for t
204. s example of increment espect region rspect using adc could be change to a shorter form incr rspect espect using adc Arrays of regions can be used with histograms or arrays of his Using Arrays of Regions tograms in a variety of ways Consider the following definitions define adc 32 previous data int 2 define x hist from 0 to 100 100 bins define y 32 hist from 0 to 100 100 bins define rx region from 0 to 100 100 bins define ry 32 region from 0 to 100 100 bins event newdata incr rx x using adc 0 This seems quite straightforward and increments bins in x using the first value in the adc array via the region rx Similarly we could increment 32 histograms using the corre sponding values from the adc array using an array of 32 regions incr ry y using adc If we wanted to increment all 32 histograms from the 32 adc val ues but use the same region in each case we could use incr ry 0 y using adc or incr rx y using adc In the first example any subscript could be used to specify a sin gle region Using an array of regions is allowed only in conjunc tion with an array of histograms The following example is incorrect and would cause an error message to be printed incr ry x using adc The EVENT Section Using Regions With Histo Making a LOOKER Description File 4 19 The Looker Remember that the histogram and not the region is being incre mented and the region can only work in conjunction
205. s not automatically find and copy the list of users who can access the experiment When reading multiple run number ranges pressing stop then rewind can hang XLUCID C 1 Problems When Running On Line Stopping LUCID from a paused state does not give the user the option of Just Stop or Resume then Stop e SunOS does not always add an EOT when a tape device is closed LUCID on line data tapes will report an error after reading all the data although the data and the gen erated histograms are fine Requested Enhancements As you read this section remember that these are only the re quested enhancements that are most likely to be implemented Although many of these changes will be done and some are al ready being scheduled there is no guarantee that all will be add ed Changes to READER e Add Loops statements for the experiment reader file Changes to WRITER No known requests outstanding Changes to LOOKER Allow on line data acquisition to continue while a looker is being rebuilt Off line lookers could optionally close the data stream at the end of the last requested run increment statements should have the option to incre ment by a value other than 1 or 1 Processing of some or all events should be switchable out side the looker description file The print statement should allow some type of formatting option e Add two dimensional arrays Allow non fatal read errors
206. s quickly The value is typically 3 or 4 but 10 would allow a 1 second delay between subse quent clicks This delay is used in conjunction with the histogram windows discussed in the next section LUCID Debugging Level can be adjusted for 7 different parts of the system It is used mostly by system adminis trators or programmers to determine problems within the system Changing these numbers cause messages to be displayed which describe what the system is doing The numbers can be thought of as thresholds so that a val ue of 10 or more allows no extra messages to be printed A value of 9 will cause a few extra status messages to come up 8 will bring up even more and so forth BE WARNED that setting any of these values to a low num ber will cause a lot of output to be generated For exam ple if the debugging level of the manager was set to 0 the simple action of building an experiment could gener ate a dozen pages of output Text Editor to Use instructs the system which text editor you want to use when making or changing description files It can be set to LUCID emacs edt or vi If edt or vi is chosen the system will start a terminal sub window in which to run the editor Automatic Build can be set to true or false this allows the system to build the experiment automatically when it connects to an experiment which hasn t be built yet Online or Offline selects the default state to be used in the Build sub window and hence
207. s recorded in the header LUCID Data Record Format Declarations LUCID Data File Format 7 23 o Lue Utility Programs and Subroutines struct Lucidvolume struct Lucidheader lv head standard record header u char lv version LUCID version number i e X 11 u char lv release LUCID release number i e 1 XX char lv pad 2 Pad to 0 mod 4 char lv volname LUCIDNAMESIZE volume name 14 Definition of new session record struct Lucidsession struct Lucidheader 1s head standard record header char ls expname LUCIDNAMESIZE name of experiment char ls dscname LUCIDNAMESIZE description file base name char ls name LUCIDNAMESIZE name of person beginning session hi Definition of new run record struct Lucidbegin struct Lucidheader lb head standard record header u long lb runno run number char lb name LUCIDNAMESIZE name of person beginning run E Definition of actual event record struct Luciddata struct Lucidheader 1d head standard record header unsigned long ld _data 1 The remainder of this record type is dynamic and depends upon the user s specification as found in the symbol table record LUCID Data Record Format Declarations LUCID Data File Format 7 24 o w Utility Programs and Subroutines Definition of comment record struct Lucidcomment struct Lucidheader lc_head standard record header char lc name LUCIDNAMESIZE name of commenter char lc text 4 body of
208. s with Q format data files Usage is qview cehd qfile The flag arguments and their effects are C Display comments in begin run and comment records e Display event counters in end run records h Display event headers in data records d Display event data in data records implies h If no files are specified on the command line Q data is expected from the standard input readsara Readsara communicates with the accelerator control system to retrieve and display values of named signals The common use is to use readsara in a READER startup event or a timed status event to obtain values of accelerator settings and save them Any number of signal names can be given on the command line and readsara prints out values in the same order Listing the approximately 2000 different sara signals is beyond the scope of this manual vme console The vme console command is used to connect to the console port of the acquisition computer This is useful for debugging purposes and for resetting the acquisition computer Usage is simply vme console machine name where machine name is the name of the acquisition computer If a connection cannot be made from the current machine vme_con sole gives a diagnostic message with the name of the computer to connect from Utility Programs 7 7 Lee Utility Programs and Subroutines The lucid Command The xlucid command only allows an experiment to be contro
209. same as above Lucidclose Closes the LUCID data stream If the data comes from a pipe which was initiated by Lucidfdopen Lucidclose must be called before pclose Writing LUCID Data Records This group of routines provide a low level method of writing LUCID data files Lucidwopen file The named file is opened for output If the file already exists out put will be appended to the end of the file Lucidwrite data size The LUCID record referred to by data containing size bytes is added to the current LUCID data block If this fills the data buff er the buffer is written Lucidpad The current buffer is filled with an LREC NOTHING record and written out Lucidweot Two EOF marks are written to the output tape device and the write head is then positioned between these marks Lucidwclose The output file is closed Unfortunately the LF_ routines below do not make use of the Lucidwrite routines Until this changes Lucidwrite and LF event cannot be combined in the same program for pro ducing a single output data stream Generating LUCID Data Files The LF_ Lucid Format set of subroutines also write out LUCID data files but provide a higher level of access to the user Unlike Lucidwrite which expects all the proper LUCID records in the correct order here records can be built up an element at a time LF_rsymtab filename The filename can either be NULL indicating that there is no rsymtab file and t
210. see User written Code on page 3 29 The second type of inclusion is with the load keyword e g load home wright Lucid readercommon o load home wright Lucid readerlib a Files named with the load statement may have been generated from FORTRAN ASSEMBLER or C or may be a library or ob ject files so generated The only restriction on the compiler used to create the object file is the functions must be C callable or the global variables must be C accessible Object files or libraries built in this way do not have direct access to the variables or def initions created from the DEFINE section instead calling rou tines must pass the variables or their addresses to the routines in the object files 1 The source files for loaded object files and libraries could include frontend buildname h file in the experiment directory to gain access to the DEFINES and structure definitions The user must find the path to this file see The LUCID Experiment Directory on page 6 4 The user s source files are not automatically recompiled when a change in the reader file causes a change in the h file The USER CODE Section Software Triggers Making a Reader Description File 3 11 The Reader The EVENT Section A word of caution is necessary at this point LUCID will record the reader description file to the output device but does not at tempt to record any files referenced by a program or load state ment This means it is no longer possibl
211. so that variable names can be asso ciated with them in your description file The following example shows the definition for a Kinetic Systems model 3615 scaler define doubles 3615 B 0 C 1 N 10 A 0 2 The DEFINE Section CAMAC Variables Making a Reader Description File 3 4 eee The Reader CAMAC Database TheB and C keywords CAMAC subaddress codes This particular statement tells LUCID that you have such a scaler in station 10 of CAMAC crate 1 on branch 0 It also tells LUCID that you are interested in only the first three inputs subaddress es numbered 0 through 2 The module can be referred to with the variable name doubles and we will see that it is actually an ar ray containing three values LUCID knows details of how to access this module because it has access to a CAMAC database see Appendix A The CAMAC Module Database All common modules have entries in this system database so that each experimenter need only specify a minimum amount of information to have the module accessed properly In this example the identifier 3615 was recognized because it is listed in the database along with a description of the module being a 100 MHz scaler Another name for the same module could be used such as KS3615 or Kinetic Sys tems 3615 both of which appear as alternate names in the system file This system module database is simply a text file that has been set up and is maintained by the system administrator
212. spaces If the name doesn t con tain the computer part the experiment will start on the com puter running the lucid program itself The user who first creates an experiment is considered the exper iment administrator and is allowed to name other users who can access the experiment Regardless of this fact the first person to gain access to a particular session of an experiment assuming he The LUCID Program 2 6 m Using LUCID HW Experiment Start Connect 2 Experiment Search Experiment Experiment Name antares arcturus Hosts z __ _ _ j altair Start Connect Cancel aries bootes canopus carina Start Connect i cetus i crux Search and then Connect E 9 cygnus co Change Directory Search Select Search All Cancel ow Experiment Disconnect oH Experiment Change Directory Experiment demo_doug skatter Directory Name home murray After Disconnecting work in Background Shutdown DEG Cre Change Directory fhome murray Figure 2 3 Sub windows from the Experiments Menu has permission from the administrator may use all functions de scribed in this manual other colleagues who subsequently con nect to the same experiment at the same time are visitors in that they are not allowed to make use of the control buttons on the main window They are allowed to view dat
213. stogram that value is silently ignored no diagnostic mes sages will be issued and no bins will be incremented For more information refer to page 4 28 The LOOKER also understands the decrement keyword which simply decrements the appropriate histogram bins in the same way that the increment statement increments them It might be used after incrementing a histogram and discovering a mistake The same restrictions apply to decrement as apply to increment The EVENT Section Incrementing Histograms Making a LOOKER Description File 4 17 The Looker Assigning Histograms To and From Arrays A histogram can be converted into an array by simply assigning it For example the following segment transfers the values from the bins of the spect histogram into elements within the junk ar ray define junk 1024 int 4 define spect hist from 100 to 200 endrun junk spect The rules for assigning an array to another one of different size apply to histogram assignments to arrays see page 4 15 for de tails Similarly the values within array elements may be assigned to histogram bins The rules mentioned above still apply Using Regions With Histograms The LOOKER allows certain bins within histograms to be grouped together and treated as a sub histogram Regions have a multitude of applications such as disabling counts from broken wires in a wire chamber or integrating bins of a peak in an energy spectrum A region is simply a m
214. stream by se lecting the Forward button This button is only usable when the system is off line The Eject Button The Eject button is only usable when the system is off line and the input file represents a tape drive unit If the Stop button is se lected the Eject button can be used to rewind the tape and take it off line from the computer If the tape is a cartridge type it will be physically ejected from the drive The Update Visible Button The user can update all displayed histograms at once by press ing this button It is the same as selecting Update in each visible histogram window Status Buttons Status buttons appear on the right side of the LUCID window beneath the Run Number display They describe the current sta tus of the READER LOOKER and WRITER Run 8 Reader Looker Writer Figure 2 13 Status Buttons In Figure 2 13 we see an example in which the READER and LOOKER exist but the WRITER does not Further the READER and LOOKER are both stopped Each of the active status buttons may be selected to reveal one of four different messages each subsequent selection will cycle through the subsequent messages until all four have been dis played In sequence they represent 1 The actual state of the program such as Stopped Running Starting or Stopped at End of File 2 The number of bytes processed so far in the current run The LUCID Program 2 22 opdmEER O Using LUCID
215. t bit which has an offset of 0 A length of 0 indicates the field isn t used so a module with a shortword word size and a compression definition of channel length 0 value length 16 offset 0 would indicate no channel data and a single shortword for the data value The READER does not actually use the compression informa tion provided Rather this information is made available to the LOOKER to decode the compression A more complete discussion of different compression formats can be found in the section LUCID Data File Format on page 7 15 Defining the Modules A 7 Sample CAMAC Module Database The following definitions originate from a CAMAC module da tabase on an active LUCID system The modules are in no specif ic order but are presented to show how the statements should appear in the file You should consult the user s manuals for each of these modules to see how the CAMAC commands are repre sented here Remember that modules may be duplicated if they have different names or slightly different functions If LUCID finds duplicate module names it will use the first name in the file module Accelerator lab homemade 16 channel ECL D Flip Flop Coincidence Module SAL Dffcoin SAL Dff shortword read is f0 a0 module Accelerator lab homemade 8 channel ECL Delay Module SAL Ecldelay readsize 8 shortword write is f16 a0 7 read is f0 a0 7 module Accelerator lab homemade super tagger unit SAL sup
216. tabase as described in Appendix A Saving Data Probably the second most important feature of LUCID after read ing the data is saving it someplace We ve mentioned that the WRITER will actually record the data but the READER must de termine which variables should be saved The save statement tells the READER to save a variable for the rest of the system to use Remember that reading a variable does not mean that it will be saved The save statement is usually combined with the read state ment read and save adc This method is the most common way of acquiring data but it places a restriction on the rest of your event Specifically if LU CID finds a statement of this form where the variable is being read and saved on the same line it will move the data directly from CAMAC to the data stream This is the fastest way of acquir ing data but it means that the data won t be available to the rest of your event statements or even within this event statement Itis quite acceptable to use two separate statements read adc Save adc This is less efficient than the previous example but it tells LUCID to save the data for future computations by the READER As one might expect using the save statement on an array with no subscript will save the entire array You should also be aware of a restriction placed on the save statement when the if state ment is used Specifically you aren t allowed to save data within an if statement
217. tatement defines a single precision variable named ener gy Several variables of the same type may be defined in the same statement define el e2 e3 integer 2 The DEFINE Section Defining Variables Making a Reader Description File 3 6 OdEmER The Reader Variable Data Types Defining Arrays Variable Names Memory restrictions The following table shows the allowable variable types Definition Range of Values 12x10 9 to 3 4x10 8 appr 7 decimal digits of precision integer 2 32768 to 32767 integer 4 2147483648 to 2147483647 character N Only used to store text N may range from 1 to 255 boolean Oor1 One dimensional arrays of any type can be used and are defined by adding a number in brackets to the name For example define calib 100 integer 2 This allocates an array of 100 integers each containing two bytes When the array is actually used a subscript is often used to iden tify a single element in the array The subscript always starts at zero so that elements in this array can be accessed with sub scripts from 0 to 99 This is important for CAMAC operations since sub address 0 from a module is commonly the first element in an array Arrays are further discussed under Using Variables on page 3 12 When thinking of names for your variables try descriptive ones that will be obvious to your colleagues Remember that these names will be associated with your data when analyzed off line p
218. tation the program was based on may not be current Command syntax is lucid to q ooutputfile inputfile Where outputfile is the destination O file and the rest of the argu ments are LUCID data files If the output file is not specified data is written to the standard output There is no current documentation for this command lucidlog The command syntax is lucidlog N experiment machine_name Utility Programs 7 3 o Utility Programs and Subroutines Lucidlog displays the last N lines of the logfile for the named ex periment If machine_name is not specified the log file on the cur rent computer is used lucidman The lucidman command prints out the LUCID manual pages on systems lacking the MANPATH environment variable e g ear ly ULTRIX systems It accepts the same arguments as the man command on the current machine lucidview Lucidview displays the contents of the named files in human readable form Lucidview reads from the standard input if no file arguments are specified The flag arguments control the amount and style of output produced By default lucidview displays ev erything The flag arguments and their effects are C Display headers and contents of comment records d Display event data in data records implies e e Display data record headers and event headers within the data records h Display LUCID record headers l Display headers and contents of ne
219. te int 4 beginrun spectrum 0 increases 0 number 0 gate 0 event pulse some increment spectrum using demo adc gate gates event count increases scaler number total Figure 7 2 Demodata LOOKER File ec o Appendix C Known Problems Like many other things today LUCID is good but not perfect The first concern of any user of any software package is What doesn t work quite as documented The second concern is Will you add this feature The following sections try and ad dress these concerns by documenting known problems as com pletely as possible and by listing the requested enhancements that are most likely to added Known Bugs ayy The C unary operator al operator Use of XLUCID properties causes difficulties when mov ing from one experiment to the next The luciddraw utility program when used at the same time as the XLUCID program can lock up the experi ment XLUCID will allow you to change directory for working with different reader looker files but does not change the directory for the LUCID program Changes made to the program files are not seen until XLUCID is restarted in the new directory e Shutting down an experiment completely from XLUCID does not always kill the manager process The XLUCID message window can overflow is not supported as a condition Problems When Running Off Line Starting an existing off line experiment on a new machine doe
220. the READER description file was changed halfway through an experiment so that several new variables were saved in some event then LUCID will recognize that change whenever the new data is encountered It will rebuild the The LUCID Program 2 13 Using LUCID Menu Buttons The View Menu software to accommodate the changes and tell you about it In put can also come from a program directly to the READER when off line This might happen when comparing acquired data to data from a simulation program If a description file was set up to ask the user for input at some point then LUCID will stop the data flow and bring up a win dow asking for the appropriate value This happens automatical ly and the experiment won t continue until the input is supplied LUCID allows you to bring up the input window again in case it was accidentally dismissed The user can select the Supply In put option to redisplay the window but the option itself will be inactive if there is not input expected at that time The LUCID data stream allows comment records to be inserted Selecting Add Comments brings up a text window not shown into which the user can enter a comment and then have the com ment record entered into the data stream This only works when on line The Clear Outstanding Commands tells LUCID to forget about ever getting a response to a command sent to the READER or LOOKER This should only be used when LUCID refuses to run new commands fo
221. the operating mode for the experiment New Histogram Windows can be set to Open or Close to determine the status of newly displayed histogram win dows Zoom by Percentages affects the ZOOM buttons on the histogram window See The Histogram Windows on page 2 23 If set to Yes the windows zoom by a stepsize percentage otherwise the windows zoom by a stepsize amount The LUCID Program 2 17 Using LUCID User Permissions Menu Buttons The Quit Button The user must select the Store button to ensure that the changed properties will stay in effect next time the system is used The User Permissions sub window allows the user to check on the names of colleagues which are allowed access to the experi ment The window lists all valid users on the computer systems as well as the users of the experiment If the current user is the experiment s administrator that is the one who created the ex periment then the list can be changed The setup allows moving names from the computer s user list over to the experiment s user list Names can also be deleted from the experiment s user list completely The creator of the experiment can also name any other user as the administrator by typing the new name into the appropriate field at the top of the sub window as seen in Figure 2 9 The anybody special user name allows any user to access the experiment This is helpful when there is more than one person involved in running an experiment The o
222. these buttons tells us that LUCID will use descrip tion files which start with the word demo The LOOKER de scription file will therefore be named demo l Accordingly the READER description file would be called demo r if we were on line This brings up an important point about description files A READER description file is only required when data is being ac quired online the description file itself is saved along with the data This means that when you eventually analyze the data LU CID can just look at the data file and write the software automat ically For this tutorial the description files should all start with demo A Short Tutorial Session 1 8 OodmHERE A Tutorial Introduction Now lets see what a description file looks like At the right side of the window are the edit buttons which allow you to create or change a description file Choose the box labelled Edit demo l and a sub window will appear containing a simple LOOKER de scription file Edit define demo adc 12 previous data int 2 define spectrum 12 histogram from 0 to 2048 event pul se command reset increment spectrum using demo adc spectrum 0 ca h LOOKER Description de mo l The window that appears is actually a text editor By positioning the pointer and selecting a position in the file you ll be able to easily change the contents Chapter Four will discuss in detail what the description file c
223. tion File 3 5 The Reader Module Timeouts For some experiments it may be efficient for certain modules not to interrupt the computer Rather than waiting for a LAM LUCID can continuously poll a module to determine when data is ready This is important when you have a module with a slow conver sion time that is to be read out after a LAM is received from some other module The details of how you wait for the data will be discussed under the heading Polling Camac Modules on page 3 20 but at this point the concern is timing It usually isn t a good idea to just wait forever for a module to be come ready so LUCID allows you to specify a timeout period This tells the READER that if it must wait for the module then it must not take longer than a certain amount of time If it does take longer the data is considered bad A CAMAC module s timeout value must be given with the definition as shown in the example below define energy LeCroy 2259 N 10 timeout 10 microseconds Times up to several minutes may be specified although only a few microseconds should ever be necessary for most applica tions You may actually use the following keywords to specify times microseconds milliseconds seconds min utes A default timeout is included in the CAMAC module da tabase for most slow modules How to DEFINE Variables LUCID variables are defined in a way similar to FORTRAN as shown in this example define energy real 4 This s
224. tion in the event stream that a rejected event was encountered The reject statement is very useful in conjunction with the qresponse variable within an if then else statement You can then reject an event if a module doesn t come ready in time After a reject any conditionally triggered events are triggered Note however that events that are to be software triggered i e every N other event see page 3 9 willnever occur after a reject Writing to Camac Modules LUCID allows writing to CAMAC modules This might not seem acommon thing for a data acquisition system to do but it proves very useful in a variety of situations For example you could load an entire programmable logic unit when the experiment is started or when a power supply overcurrent trigger arrives you could write a zero value into its controller You can use the write statement as shown in the following ex ample write 100 into dac The EVENT Section Rejecting an Event Making a Reader Description File 3 26 OdEmER The Reader This simply puts the named value into the module named dac using the default subaddress and function codes for that CAMAC module as defined in the CAMAC database To write larger amounts of data you could use an array name For instance to load an entire file into a programmable logic unit you could use the following event statements read pludata from home norum plu datafile write pludata into plu This is ineffic
225. tive and must divide the entire histogram into an exact number of bins In other words the difference be tween the last and first bins must be a multiple of the binsize LU CID doesn t care if you have a histogram with only 1 bin but it will complain if the bins don t divide the total histogram into equal pieces For example define pl hist first 0 last 99 binsize 8 1 To have the values rounded to the nearest bin boundary higher or lower you can add one half of the bin width half the resolution to the variable before histogramming Histogram Definition Defining Histograms LOOKER Histograms 4 32 m The Looker This implies that 12 bins have a width of 8 and the last one has a width of 4 Since all bins must be the same width LUCID would print an error message for this definition Probably the best fix would be to use define pl hist first 0 last 99 13 bins This histogram would have 13 bins between 0 and 99 inclusive Since the value 99 names the last bin there are actually only 12 bins which are less than 99 Each bin is therefore 8 25 units wide The last bin will accept values from 99 up to but not including 107 25 This example also shows some of the shortcomings of using the first and last keywords most of the time they aren t intuitive and don t fit the concept that you may have of the histogram Us ing the rom and to keywords are recommended most of the time One dimensional histograms default to
226. to an experiment a simple yes no confir mation is asked for Control Buttons The square buttons located on the left of the main LUCID win dow allow the user to control the data flow after the experiment software has been built Playback Sequence Figure 2 11 The Control Buttons Before connecting to an experiment or building the software the control buttons will be inactive Once built certain buttons may remain inactive For instance the Record button in Figure 2 11 is inactive because that example represented an off line session having no WRITER destinations Once a control button is selected a thick edge will be drawn to highlight its border For instance the currently selected button in Figure 2 11 is the Stop button On a colour display the button will have the 3 dimensional appearance of being pressed in If a user connects to an experiment in progress the control but tons will be displayed to reflect the current operation in progress Also if the user is just visiting the control buttons will change as the controlling user makes changes 1 Technically the OpenLook specification refers to these buttons as choices They are called buttons here for simplicity Control Buttons The LUCID Program 2 19 Using LUCID The Record Button When the user requests that a WRITER be present in the experi ment the Record button will be made active In such a case the Rewind button will not b
227. to enable and disable LAMs and how to read and clear modules in a single operation LUCID silently reads this CA MAC database every time you start it up so you normally never need to know of the file s existence Defining the Modules Definitions are listed one after another in the database file Each definition starts with a line containing the word module fol lowed by one or more names for that module For example the definition for a Kinetic Systems model 3615 scaler starts off like this module Kinetic Systems 3615 6 Input 100 MHz Scaler Scaler3615 ks3615 3615 You can have several aliases for a CAMAC module but each of them must be surrounded by double quotes Recall that when you specify a type of CAMAC variable in your READER descrip tion file the type of variable must match one of the names given in the database For instance our first example under The DE FINE Section on page 3 4 was define doubles 3615 crate 1 slot 10 addr 0 2 LUCID recognizes the module 3615 because it appears as one of the aliases in the CAMAC database file The first name for a module should be a long descriptive one That s the name LUCID will display if it thinks you re using a module incorrectly in your READER description file The other names should be written in decreasing order of length as shown in the first example There is no reasonable limit on the number of aliases you can have for a module As with all
228. to the nearest bin boundary The button labelled Set From Markers may be selected to allow zooming in to the area enclosed between markers on the display For the Y axis limits of a 1 D histogram the button labelled Set From Markers performs a slightly different function the extents of the Y axis are set to match the extent of the data values in the area between the markers This essentially sets the Y axis limits so that data between the markers will fill the display When a 2 dimensional histogram is being viewed the Y axis lim its button labelled Set From Markers will perform exactly the same function as the button does for the X axis Further the but ton labelled Set From Markers for the Z axis is usable and will set the Z axis limits to the limits of the data contained within the marked area If the user adjusts the Y limits for a 1 D histogram or the Z limits for a 2 D histogram the Autoscale option is turned off automat ically Values below the from limit are drawn as being at the from limit and values above the to limit are drawn as being at the to limit Set From Markers for these axes will copy the high value from within the area delimited by the markers The Reset buttons allow the user to easily reset the limits to their default full scale values Rebinning The Rebin Factor provides a simple dynamic integer rebinning facility for XLUCID The bin width indicates the number of X axis bins that are summed to obtain th
229. to value In the above example the first dimension has a bin size of 0 25 the first bin ranges from 0 00 to 0 24 the second bin ranges from 0 25 to 0 49 and the last bin ranges from 255 75 to 255 99 Using range speci fiers for first and last includes the high value of the range caus ing the bins to be that much wider Defining Regions Although regions are used with histograms they are defined in dependently in the DEFINE section of the looker The important point to remember is that a region must have the same number of bins dimensions and array elements as the histogram that will be using the region For the sake of clarity regions should also have the same ranges defined as their corresponding histo grams A simple two dimensional region definition is almost identical to a histogram definition define tdc region region from 0 to 256 1024 bins and from 1024 to 2048 16 bins contains from 1024 to 1536 from 0 to 128 from 1536 to 2048 from 128 to 256 The tdc region could now be used with the tdc hist 2D in the earlier example The contains entries define a checkerboard as shown in Figure 4 1 the low half of the Y range in parenthesis with the low half of the X range and the high half of the Y range with the high half of the X range The contains entry is optional and if excluded the region defaults to the entire range 1 A more complete discussion of using Regions can be found in Region Definition on pag
230. ty over this type of trigger except for other ones being performed constantly 1 The XLUCID program does not recognize control characters it only allows commands to be run by button selection For this reason it makes more sense to have commands with meaningful names rath er than single key presses The TRIGGER Section Software Triggers Making a Reader Description File 3 10 ED IM The Reader Trigger Examples Some examples of trigger statements are listed here trigger good hit every tdc lam trigger counting every 2 seconds trigger delta every test command or every coincidence lam or every 10 overflow event or every 60 minutes trigger calibrate every minutes The USER CODE Section It is common for a non standard module in an experiment to re quire some special handling which the READER language can not easily provide The user code section has two statement types one of which allows direct inclusion of C code and the second which allows specifying object modules or libraries to be used C code is included directly with the program keyword e g program home wright Lucid bitswap c This statement causes the named file to be included in the gener ated frontend program The C code can declare global variables and functions and have access to the variables and definitions created from the DEFINE section Variables and functions de fined here can only be accessed by user code entries in the EVENT section
231. u should see 10 messages endrepeat The repeat statement must contain a variable optionally fol lowed by a beginning value and ending value and a step size All of the statements up to the endrepeat will be repeated so long as the variable is within the specified range This particular ex ample goes from 0 to 100 in steps of 10 changing the value of x at every iteration The following example shows an infinite loop repeat x print Another iteration endrepeat The beginning end and step portions of the statement are op tional allowing many variations repeat x from 1000 by 100 repeat x to 10000 repeat x by 2 repeat x from 1 The variable which does the counting may be changed within the loop to affect the control as you would expect For instance you might explicitly set it to the upper limit so the loop will ter minate Repeat statements are allowed to be placed within other repeat statements The variable s initial value should be less than the loop s final value and the increment must be positive This is because the looker always tests that the variable be less than or equal to the final value The EVENT Section The Repeat Statement Making a LOOKER Description File 4 23 The Looker Printing Values from the LOOKER The print statement is quite general and allows the printing of messages to the user s terminal screen The messages may con tain values from variables of any type The following example sh
232. ubroutines define LVAR PADNAME lt lt pad gt gt name for all pad bytes define LVAR HISTFLAG 0100 histogram bit or d in define LHIST MAXDIMS 2 maximum histogram dimensions define MAXHISTBINS 4096 maximum 1 dimensional hist bins define MAXSCATBINS 256 maximum 2 dimensional hist bins Data compression variations define LCMP SHORTSHORT 1 Shor define LCMP LONGSHORT Long define LCMP LONGLONG Long t channel short data 2 channel short data short pad 3 channel long float data define LCMP LONGDOUBLE 4 Long define LCMP HISTOGRAM Shor data define LCMP DOUBLEHISTOGRAM Shor pad double channel long pad double data 5 t X channel short Y channel long float 6 t X channel short Y channel long data define LCMP VARIABLE 7 New style compression A variable stored using LCMP VARIABLE data compression appears in the data stream as a variable length and fixed length component The variable length components appears after all fixed length data Indices into the variable length data are stored among the fixed length variables The offsets are the offsets from the beginning of the event of the first and just past the last byte of the associated variable For example event el read and save v1 read compress and save v2 read and save v3 read compress and save v4 read and save v5 LUCID Data Record Format Declarations LUCID Data File Format 7 19 o Lue Utility Program
233. uch data can be written to the module at one time These read and write sizes are used to check and possibly limit the amount of data transferred to or from a module in a single operation With respect to defin ing a CAMAC variable in a READER description file this size de termines the array size of the variable All modules should have a readsize or writesize defined An option for hold mode transfers can be specified with these keywords this will be dis cussed under HOLD Mode Transfers on page A 4 Shortword and Longword Most CAMAC modules transfer 16 bit values or less However LUCID must be able to communicate with 24 bit modules as well such as high density ECL coincidence registers The key words shortword and longword tell LUCID whether the module will transfer 16 or 24 bits respectively If neither of these key words is present LUCID assumes the module will transfer 16 bits Only one of these keywords may be included in one module definition Checking Module Identity Some CAMAC modules have an internal status value which might contain a serial number or I D number If a module s def inition includes a check statement then LUCID s program will check that value before it starts the experiment This is useful when setting up an experiment because LUCID can help remind you when modules have been moved around The general for mat of this statement is check is f1 al data amp 0140 0100 This tells LUCID to
234. until both the LOOKER or WRITER are ready to accept it Fortunately the LOOKER allows certain events to be processed only when there is enough time to do so Consider the following example define spec hist event hadron some increment spec using adc When on line the LOOKER knows that it needs to process only some of the hadron events and will not hold up the READER while it works If other hadron events go through the data stream while its processing a previous event the LOOKER will ignore them By default the LOOKER processes all events with the given name but the keyword some will make sure that the READER won t be held up while processing goes on Similarly if no event name is given and the statements are per formed for all events the some keyword may be used For ex ample event some print An event has been seen The print statement will be performed when any event is en countered but doing so will not keep the data stream from flow ing When off line the some keyword is ignored as it is most impor tant that the looker sees all events The EVENT Section Event Types Making a LOOKER Description File 4 10 w The Looker Actions at the Beginning or End of a Run User Specified Commands Statements may also be performed whenever the beginning of a new run is encountered in the data stream Important values and messages could be printed when the analysis is ready to start for example define nume
235. uote marks LOOKER Histograms Probably the most common practice followed when analyzing data is to count the number of times that specific values are found in a variable For instance it might be important to know how many times a certain variable contained the value 103 or how many times the fourth bit position in a variable was set to a 1 The EVENT Section Using Your Own Subrou LOOKER Histograms 4 26 OodmgEEER The Looker Histogram Definition Normally we have a separate counter for every possible value that a variable might contain every time the variable changes the counter which corresponds to its value is incremented by one Keep in mind that a variable which contains N bits will need 2N counters For example an 8 bit variable would require 256 counters Similarly a common integer 4 variable would re quire more than 4 billion counters Unfortunately the computer can t manage 4 billion counters so a compromise us usually made where several different values share the same counter Regardless of the number of counters we can visualize this pro cess by thinking of a bar graph with one bar representing each counter the bars will increase in height whenever the variable is found to contain the corresponding values For example we can visualize a set of twenty six counters like this 55 26 counters booed H As previously suggested it is sometimes important to count the times
236. use the reader to continuously send a CAMAC command which tests for data being ready Most modules respond to the CAMAC F 8 command by issuing an appropriate Q response A Q response of true a one indicates data is ready The CAMAC function code that will be used is looked up in the system CA MAC database for the module and LUCID will warn you if a par ticular module has no polling mechanism Because it is possible the wait command will timeout the qre sponse variable should be tested after a wait statement wait adc if qresponse 0 then Adc not ready due to timeout Reject this event reject endif The EVENT Section Other CAMAC Operations Making a Reader Description File 3 20 ED IM The Reader As in previous examples the wait keyword can be used with certain other keywords The most common example is wait read clear and save adc Each such keyword can be separated from the next by a comma or the and keyword for readability The wait command has no effect on non CAMAC variables and LUCID will warn about such usage CAMAC Q Response The wait command can be very useful for certain modules but could give incorrect data in other cases the timeout period that was specified for a module may have expired and the data that was acquired could be meaningless To let you check for this condition LUCID provides a boolean variable called qre sponse It does not have to be defined like other variables since
237. ust have the form read is 0 a0 5 This tells LUCID to use F 0 with a sub address code A 0 through A 5 Also the keyword is is optional If the module only understands one A code then the definition would be what one would expect read is f0 aO LUCID will warn you if it finds a function code for reading which is not between 0 and 7 It also complains if the A codes are not between 0 and 15 HOLD Mode Transfers Certain types of CAMAC modules allow reading or writing sev eral channels without changing the A code every time They keep track internally of what has been accessed and automati cally proceed to the next channel after every transfer These are called hold mode transfers because the computer holds the same A code on every transfer to or from the module In some cases hold mode transfers can be performed slightly faster than ordi nary transfers More importantly CAMAC only allows sixteen different A codes to be used and some modules need to transfer more data than can be done in sixteen transactions Defining the Modules A 4 ae With hold mode transfers a large amount of data can be trans ferred in a single read statement from LUCID For instance a waveform digitizer can transfer an array of 1024 bytes with a sin gle read statement from the READER To tell LUCID that a mod ule uses hold mode reads or writes you must include the word hold after the readsize or writesize statements in the database For
238. ut columns in the file as FOR TRAN does but experience recommends that each line be kept short so as to be easy to change with a text editor Related state ments should be indented with a TAB character for instance by the same amount Blank lines are acceptable and encouraged to allow easy reading by other people An octothorpe f is recog nized as the start of a comment LUCID ignores this symbol and all other characters up to the end of that line The only exception to this is when that symbol appears between quotes as part of a string constant The DEFINE Section LUCID allows you to do computations with your data as it ar rives so you can check its validity Results of these computations can be saved as part of an event along with the raw data LUCID allows you to define variables very much like variables in most programming languages Furthermore LUCID treats CAMAC modules as variables allowing easy access and simple manipu lation The first section of the READER description file tells LUCID what variables you want to use in your experiment All variables must be defined so that LUCID can build correct programs and to make sure there are no typing mistakes in your description LU CID can also make other checks about the consistency of data from CAMAC modules For example it will warn you if you at tempt to assign a twenty four bit CAMAC module variable into a sixteen bit integer variable CAMAC modules are defined
239. ve Region ODUOf uuo iioi teeb rsads 2 45 The Print Menu BUON au et ctetdtttktniet reputant MR ARRA MEAM eni 2 46 Te Prnt ODIO did du siete cura Re mi Fs boveubUb En OR QUU 2 46 Th Layout ODDO oisi rer ran baci doces ei euo an pro orden D edi 2 46 Chapter3 The Reader Camac Hardware LayOUt ssieiss euo Aiteksto Lon edi edili iid 3 1 ISIN Ein UR TM 3 2 Making a Reader Description PIe ooi cie i ha brc eiit 3 2 Four 5ectionsof a Description File esce RR ER BRBRH MEM R ER A ids 3 3 The DEFINE Sech ON uud den ix HU PONI eaim prb Err liner pe Pu 3 4 How t DEFINE Vatablesi seite d dubi cr e dais uet 3 6 The TRIGGER OCCU edicti iid ie b vem tn teu be UP Du MR Udo 3 8 The USER X CODE Secho ereinen uhi MINIMI n ME 3 11 The EVENT SOOO Seksist eiii ak iE A eaa aaaeeeaa 3 12 Using VasaB es s nesuaconttse Nia E a a 3 12 IET oiii zc PN E E 3 14 Savine p 3 18 Other CAMAC Operations asisite srin dass tube cidem 3 19 Polling Camac Modules uieusuacotstitepnicetphei eer tcrra r 3 20 CAMAC CI RESP OSG uiiciuadcerca ppt n nien tech bn S Da Ricci cen 3 21 Compressing Afray JUGE oae tehenid ei ceteni tit id REF CU b enl 3 21 Calculations and Variables uai esee EB Libre teh tue 3 23 Printing Messages During Event Processing 3 24 Making Decisions about the Data scars bie irit coheret e eth 3 25 Triggering Another BVent ueecaena te tetra irc tani etr edt 3 26 Rejecting an EVIL assis acid vase bam ari Hr
240. vents int 4 event numevents numevents 1 beginrun print numevents seen so far This will produce a diagnostic at the beginning of each run spec ifying how many events have been seen so far in the data stream You could also use a beginrun section to automatically reset his tograms or clear software scalers that are being kept Similarly it is possible to perform some actions when the end of a Run is encountered in the data stream Simply use endrun instead of beginrun in the first line of the specification Keep in mind that variables saved in the READER are not accessible within either of these specifications To this point the instructions within a LOOKER file can only be performed after recognition of something in the data stream such as event types or a Beginning of Run record The LOOKER also recognizes arbitrary commands from the User when the data stream is being processed so that internal status informa tion may be printed or variables may be reset The following example shows a common use for this option define spect hist event hadron incr spec using adc command reset spec 0 A histogram is incremented every time a hadron event is en countered and that same histogram is cleared every time the user issues a reset command Again variables saved in the READER are not accessible from such statements The com mands may contain any type of text and follow the same restric tions as w
241. ving data points which have zero counts The default option is to save non zero data only The user may restrict the amount of data to be saved by using the final option labelled Amount to Save By default data from ev ery bin in the entire histogram will be saved assuming the pre vious option Non Zero Data Only is selected The Histogram Windows 2 25 Using LUCID Read From File The File Menu Button If the user opts to save Selected Bins Only then only data asso ciated with bins between the histogram s markers will be saved Markers will be discussed momentarily but if only one marker is present then that single data value will be saved according to the other criteria selected on this sub window Once all the desired options have been chosen the user may se lect the Save button and data will be stored The second option within the File menu is labelled Read from File and allows the user to re read histograms previously saved by the Save Data To File sub window as shown in Figure Ei Read Data from File Variable Name spectrum 7 Directory Name users3 wright Lucid xlucid Prog File Name Dataspectrum fie Fermat Data Source Data File Program Read Cancel Figure 2 17 The Read Data from File Sub window 2 17 The Directory Name field defaults to the current working directory This field is included as it is quite possible that previ ously saved histograms are stored in a directory other th
242. w LUCID ses sion records n At every end run record and at the end of each file display the number of each type of event T Display headers and contents of reader description records S Display headers and contents of event symbol table records t Table of contents showing run numbers and num ber of megabytes in each run V Display headers and contents of volume label records wwidth Sets the maximum width of displayed event data to width characters default 80 lucidman Utility Programs 7 4 OdEHMB Utility Programs and Subroutines netcamac netcamac Netcamac sends CAMAC commands to a remote acquisition computer for running on an attached CAMAC highway The command synopsis is netcamac d level D level frontend branch crate module address function data 1 1 The d and D flags set the debugging level of the front end host and the netcamac process respectively The frontend is the name of an acquisition computer The branch crate module ad dress and function specify the operation to be performed Oper ations which require data input need a data argument If any of the arguments are missing netcamac prompts for input When all arguments are supplied the program exits after the command is run Otherwise netcamac repeats the prompts to al low running another command netcrateinit Netcrateinit takes an acquisition CPU name branch and crate number as arguments The
243. w the same guidelines as one di mensional regions as described on page 4 18 Arrays of regions are allowed but the restrictions on array size apply here as they do with one dimensional regions 1 All histograms in an array have the same limits binsize and number of bins by definition The EVENT Section Incrementing Two Dimen Making a LOOKER Description File 4 20 w The Looker Simple Comparisons Histogram Acceptance Tests The If then else Statement The If then else statement allows the flow of execution to be al tered within the LOOKER description file There are two general forms of the statement First an if statement contains comparisons between variables and constants For instance the following statements check that a scaler value is below 30 000 define scaler 6 previous data int 4 event checkdata if scaler 0 lt 30000 then print Overflow else print Scaler O K endif The then keyword is optional but the endif keyword is re quired The else keyword and the statements associated with it are optional too An unlimited number of statements may be in cluded within the then or else part of the statement including other if then else statements Note that the comparison allows FORTRAN style keywords such as eq ge and so forth Standard comparison operators are also allowed as found in C such as lt gt and so forth More complex comparisons are allowed such as if x eq 10
244. wed parentheses can be used for grouping and lines may be broken into pieces when they become too long The list of sym bols which the LOOKER recognizes as valid operators is shown in Figure 4 2 Histograms cannot be used in expressions requiring a single re sult For instance assume the assignment statement in the previ ous example was changed to new e 0 10 spect 130 2 2 2 LUCID would print an error message because this expression is expected to yield a single result and the spect histogram con tains several values 1 The UNIX timestamp is the number of seconds since 12 00 AM GMT January 1 1970 There are a number of standard library routines that convert timestamps into usable values See especially the UNIX manual entry for ctime and localtime The EVENT Section Calculations and Assign Making a LOOKER Description File 4 13 The Looker Symbol Function exponentiation multiplication division addition subtraction bitwise left shift bitwise right shift integer modulus remainder from division bitwise and bitwise or bitwise exclusive or summation integration one s complement negation boolean not Figure 4 2 Valid Expression Operators The integration operator for histograms yields a single result and may be used to sum all channels in a histogram or all valid channels in a region of a histogram The following example sets the variable sum to the total count
245. wing multiple histograms to be saved in the same file Header The histogram header consists of one or more lines of free for mat information The different header fields consist of a key word possibly followed by a list of parameters The number of parameters is fixed for each keyword Variable Has a single parameter which is the name of the variable that was saved saved Followed by two parameters which are the date and time the histogram was saved run The single parameter is the run number when the variable was saved compress No parameters this indicates that fields that are zero have not been saved to the file array This indicates that the histogram is an array and has the lower and upper bounds of the array as its two parameters entry entry indicates that this is just one element from a histogram array the single parameter tells which el ement Histogram Verifies that the variable saved was a histogram type The variable type saved the single parameter is short long float double or string X Indicates that the following from to and bins key words apply to the X axis Y Indicates that the following from to and bins key words apply to tye Y axis from A single parameter which is the lower limit of val ues saved to this histogram to A single parameter which is the upper limit of val ues saved to this histogram bins A single parameter which is the number of bins for the current a
246. xis byvalue No parameter byvalue indicates that data is saved as an X or Y value representative of the bin informa tion being written out midpoint No parameters this indicates that the X and Y val ues written out are at the midpoint of the bin The default is that values are printed as of the lowest edge of the bin bybin No parameter opposite of byvalue binary No parameter indicates data is in binary instead of ascii This allows smaller data files but is not always portable to another computer and is diffucult to ma nipulate with anything but a dedicated program ascii No parameter opposite of binary Body The body of the histogram file varies greatly depending upon the way the histogram was saved However the header informa tion is sufficient to allow determining which format the data is in Ascii files have the data from one bin on each line There can be up to four fields in each line which are array index x bin number y bin number value The array index only appears if the array keyword is in the head er The x bin number is present for all histograms although it could be a bin number the x value representing the lowest edge of the bin or the x value at the midpoint of the bin depending again on keywords bybin byvalue midpoint The y bin number Overview D 2 w is only present for scatter plots 2 dimensional histograms and as the x bin number the keywords bybin byvalue and midpoint dete
247. y needed The Proper Menu Buttons The Properties Menu The LUCID Program 2 15 memes Using LUCID ties menu contains three sub windows and two other options as shown in Figure 2 9 e Properties Adjust Maximum Number of Messages to Keep Tenths of seconds for double clicking LUCID Debugging Level Defaults Permissions Save Window Positions Restore Window Positions 3 65 25 Jul 1995 Properties User Permissions Debugging Level for READER 10 _ Debugging Level for LOOKER Debugging Level for WRITER Debugging Level for Experiment Finder Debugging Level for Experiment Manager bbbbbbbpi Debugging Level for PreProcessor Text Editor to use emacs Auto Automatic Build Online or Offline Online Close No New Histogram Windows Zoom by Percentages Co co Experiment demo_doug skatter Administrator doug mputer Users kelvin i i Insert gt ken i kevin i Remove lt kim Update Permissions Experiment Users Property Defaults Experiment Name demo glen amp virgg lr4299 modules myme167 hubble home wright Lucid xlucid Database File CAMAC Processor Device Default Experiment Directory Base Name for Description Files demo File Containing Offline Data 4 lucid lib demo demodata Store CANCEL CHANGES TO THESE ITEMS AFFECT ONLY XLUCID STARTUP Figure 2 9 Su

LUCID User's Guide - Experimental Subatomic Physics

Contents

Download Pdf Manuals

Related Search

Related Contents