testBeam.html

hstDisplay

I- Overall commands

The DAQ program fills histograms which are stored in a shared memory file called HstGeneric.map which exists in the online directory. The hstDisplay program accesses this file and so can display the histograms while the run is progressing. The histograms are all identified using their ROOT labels. The classes used by hstDisplay are all daquser/inc/hst/*.C.

To display the online (immediate) histograms, use
```
        calice00> hstDisplay [options] [displayname]
```
Here, [displayname] is the name of a particular set of plots which you want to look at.
To get a list of the possible values for displayname, do
```
        calice00> hstDisplay -h
```
which should give some output like this.
The valid options depend on the displayname. To see the possible options, do
```
        calice00> hstDisplay -h [displayname]
```
All displaynames (except HstList) have an option -m
```
        calice00> hstDisplay -m 1000 [other options] [displayname]
```
which specifies the time to wait between histogram updates in msec.
To make a postscript file (or other type of file) of the canvas, use the usual ROOT "File" pulldown menu to "Save...".
To stop the display, either Ctrl-C the terminal window running hstDiplay, Ctrl-C the ROOT canvas window or use the ROOT pulldown menu to "Quit ROOT".

The sections below explain more detail on the different displaynames.

II- HstList

HstList is used to print a list of the ROOT labels of all existing histograms in the shared memory file to the screen.

This displayname has no options. Doing
```
        calice00> hstDisplay HstList
```
will give output something like this. The first field in each row is the ROOT label, the second the histogram type and the third the amount of memory taken by the histogram.

III- HstTH1F and HstTH2F

These are basic routines to simply display a single histogram using its ROOT label.

The only specific option is -n to give the ROOT label
```
        calice00> hstDisplay -n HstRecordRunRSize HstTH1F
```
will display the HstRecordRunRSize histogram (which shows the size of the records in kBytes for this run) in a canvas looking like this (ps, pdf).
HstTH2F is almost identical but for 2D histograms. There is one extra option which is -o to give display options, e.g.
```
        calice00> hstDisplay -n HstCrcSignalCrate0Slot09 -o lego HstTH2F
```
The ROOT interface to shared memory does not distinguish between one or two dimensional histograms so be sure to check which displayname to use.

IV- HstRecord

This display shows records sizes, record rates and data rates.

This display opens up to four canvases, numbered 0 to 3. The four canvases correspond to; the time since the DAQ program ("job") was started (i.e. the last startUp), since the last run was started, since the last configuration started and since the last acqusition started, respectively. To select which canvases to display, use the -b option with an integer value which uses bits 0 to 3 to specify the canvases
```
        calice00> hstDisplay -b 5 HstRecord
```
would display the job and configuration canvases. The default is all four canvases.
Each canvas should look something like this (ps, pdf). The upper plot shows the size of records, with event records in red, trigger records in blue and other records in white. The middle plot shows the rate of records vs. time in secs. The colour coding is the same as for the upper plot. The lower plot shows the total data rate of the records vs. time in secs.

V- HstCrcNoise

This displays mean and rms values for each CRC channel.

There are two display modes; for CRCs with no VFE electronics inputs and for CRCs with inputs. The selection between these two modes is with the -w option, where 0 gives the no input-display and any non-zero value gives the input display
```
	calice00> hstDisplay -w 1 [other options] HstCrcNoise
```
For the no-input display, the multiple samples taken by each ADC on the CRC are averaged; with no inputs, these should all be the same level. Hence, the number of channels per CRC considered is 8FE x 12ADC = 96. For the input display, each ADC sample corresponds to a different physical channel, to the ADC samples are averaged separately. The number of channels per CRC is then 8FE x 12ADC x 18channels = 1728.
The other options select which CRCs to display. This can be done using -b to give a bitted word for multiple CRCs, or a single CRC can be selected with the -c and -s options to specify a crate and slot. The convention is that the ECAL crate can be numbered as 0 or 0xec and the AHCAL crate as 1 or 0xac.
```
	calice00> hstDisplay -c 0xac -s 9 HstCrcNoise
```
The default is all CRCs will be displayed.
One canvas will be opened for each CRC selected. The display will look like this, (ps, pdf), where the input display has been selected in this example. The upper plot shows the number of times the channel was read out during the last acquisition vs. the channel number. The middle plot shows the mean of the ADC value, averaged over the acquisition, vs. then channel number. The lower plot shows the rms of the ADC value around that mean vs. the channel number.
If using the no-input display when real inputs are present, then putting all 18 ADC samples in one average means several physical channels are being averaged. The spread of the means between channels means the rms will appear to be substantially higher.

VI- HstCheck

This displays results of comparisons of configuration data written to the hardware and read back from the hardware.

VII- HstLc1176

This displays plots of the LeCroy 1176 TDC data.

The TDC has 16 channels, numbered 0-15, of which only 8 are used for the drift chambers. To select which channels to display, use the -b option, which takes a bitted integer with bits 0 to 15 corresponding to the 16 channels. Doing
```
        calice00> hstDisplay -b 0x1a3 HstLc1176
```
selects channels 0, 1, 4, 8 and 9. The default is channels 0 to 7.
A canvas is opened for each channel selected. The canvas will look like this (ps, pdf). The upper plot shows the number of times (for leading and falling edges) recorded for the channel. The middle plot shows the time distribution of the leading edge hits in ns. The lower plot shows the same for the falling edges.

VIII- Offline viewing of online histograms

It is possible to run the online monitoring code on a previous run. This is useful if you want to compare the current run histograms with an earlier run.

This should NOT be done in the online directory as it will disturb the histogramming of the current run. Go to the daquser area
```
        calice00> cd
        calice00> cd daquser
```
Make sure the run you want to look at is listed in the data/run directory
```
        calice00> ls data/run/
```
If it is not, then check where the data directory link is pointing to
```
        calice00> ls -l data
```
and change the link if necessary.
To run the histogramming on a run binary file, then do
```
        calice00> hstGeneric [runnumber]
```
where [runnumber] is the six digit run number. This will fill the memory mapped file in HstGeneric.map in the daquser area. The histograms can now be displayed from a second terminal window using exactly the same commands as above.
If the command is not recognised, you may need to build the executable using
```
        calice00> build hstGeneric
```
The hstGeneric program will stop when it reaches the end of the run. You can continue to display the histograms after this time. However, if you finish your work before hstGeneric stops, then it is good idea to kill it using Ctrl-C to save CPU time.