Hello, Sorry, in my past message, substitute Peter for each instance of Christian. I am getting first and last names mixed up as well as people. Kris hagel@comp.tamu.edu wrote: > Hello, > In regards to Christian's message about what he did with the TOF monitor, I > would advise all of the other people responsible for monitors to write more > compact monitors as well. As people sitting on shift have noticed, the > monitors have gotten "heavy". One point is that not all histograms need to be > in the histogram lists. If you aren't going to look at it alone, it does not > need to be there. Being in the picture is enough if that is how you will look > at it. > > Another point as Christian did is that each monitor should have some "summary" > pictures that give a broad overview of the detector performance. This should > ideally be close to one, but might be more depending on what type of detector. > > Kris > > "Peter H. L. Christiansen" wrote: > > > Hi > > > > I updated the online tof monitor so it now contains much more compact > > info. > > > > Old histograms/pictures are now normally not displayed. If you compile and > > BRAT_EXPERT_MON > > is defined you get all the old stuff as well. > > There are 4 pictures now : 1) Top and Bot pedestals, 2) Hit info, 3) > > Top info 4) Bot info. > > > > I use the new BrDetectorMonitor methods to ask for sync triggers that I > > store in a TProfile with SetErrorOption( "s" ) ( error = rms, Thanks JH ) > > and when I have had 1 sync trigger I start to use pedestal vaule + 50 as a > > cut for top and bot hits. A slat with top and bottom signals ( adc > && > > tdc < 4000 ) is a hit. > > Histos for hits are : > > slat distribution, multiplicity distribution, time diff top - bot, time > > average, and this is supposed to be the most important histograms for > > online monitoring. > > if there are slats missing you want to go to the top and bottom pictures > > where you can see slat dist for top and bot tubes, the raw adc signals and > > the raw tdc signals. This should help you pinpoint the error so you can > > mail the tof guys;) > > > > Known problems are : > > **Lots of tdc signals are less than 100. I have looked before at these > > events and they go away when you require ZDC trigger, leading on to the > > point that once we start having a real good collision trigger this should > > be tested in the online monitor. > > > > **Since in the old monitor there was 2* slats number of tdc vs adc histos > > I have made 2 such plots ( bot and top ). They don't look very useful from > > the first tests, but could maybe be made useful with trigger requirements > > and maybe a small correction like adc - ped, some small tdc alignment > > using some of the other histo info. Point is I think we should let them be > > somewhere for a little while since someone made an effort to put them > > there. > > > > ** I can simply not figure out how to make some of my 2d histograms box > > plots. Can somebody tell me how to do this in the monitor enviroment. > > Please help me! > > > > ** If the monitor is given runs that have no sync triggers, not very much > > info will be given. Only top adc and tdc + bot adc and tdc !!! > > > > I hope that as people use the program they WILL write a lot of > > comments to the list so the monitor can be improved. > > > > Also I hope that someone might find some of the ideas useful and try to > > improve some of the other monitors. > > > > Cheers, > > Peter > > > > :-) --------------------------------- )-: > > |Peter H L Christiansen aka PAN @ NBI | > > |EMAIL : pchristi@nbi.dk | > > |OFFICE : Tb1 @ NBI | > > |PHONE : 353 25269 <- New!! | > > |SNAIL : Sdr. Fasanvej 14 ST 2000 F | > > |PHONE : 38 872042 | > > :-D --------------------------------- \-:
This archive was generated by hypermail 2b29 : Wed Jul 19 2000 - 16:40:48 EDT