As a test, I used a couple of random A' signals from /work/hallb/hps/mc_production/lhe/ap/2pt2/mock[1-3]
Specifically, mock1_v1_237 (attached) and mock2_v1_12
Are you fully up to date with the latest changes or are you using a release?
I last updated yesterday (2/19/15), and I'm using hps-distribution-3.2-SNAPSHOT-bin.jar
Did you generate the SLIC (Geant4) output file by yourself?
I generated the SLIC output, using the v8 detector (same problem with v7 though), 1000 events
What simulated slcio file are you running on?
The output from the stdhep file, with the above settings
What “readout” simulation are you running? The default is to run the HPS2014ReadoutToLcio or HPS2014ReadoutNoPileup.
I used HPS2014ReadoutNoPileup. Everything works correctly at the readout stage, as far as I can tell from debug.
Can you send us the steering file for the recon (in case it’s different from the default HPS2014OfflineRecon)?
Attached is the HPS2014OfflineRecon I used (unmodified), and 'fixed' HPS2014OfflineNoPileupRecon, with the original driver I replaced commented out. TrackerDigiDriver was added. Both steering files had the same problem, which I'm suspecting has
something to do with (Data)TrackerHitDriver, or some other difference between the 2014Recons and SimpleMCRecon, which also works for me unmodified. All I can confirm is that HPS2014OfflineNoPileupRecon is now assigning tracks with the driver swap (as
seen from the debug output), but I haven't had the opportunity to analyze it any further yet.
Can you send or point us to a file that we can test?
Anything in /work/hallb/hps/mc_production/lhe/ap/2pt2/mock[1-3], but the same problem happened with other things I tried, so the input may not matter. You could also try your
own stuff that you ran before to be sure though.
Hope this helps.
Brad
From: Hansson Adrian, Per Ola <[log in to unmask]>
Sent: Friday, February 20, 2015 5:20 PM
To: Bradley T Yale
Cc: hps-software
Subject: Re: Track Problem
Hi Brad,
Sorry to hear that something is broken.
I haven’t tested running the standard tracking in a few weeks, but for my development I also use the TrackerDigiDriver and all seems fine as you say.
Others should tell us if they are able to run the standard recon?
Th TrackerDigiDriver creates sensor strip clusters from the simulated hits on the sensor directly instead of using the default hit reconstruction chain. This may or may not matter to you depending on what you are doing but the recommended way is to use
the DataTrackerHitDriver as the default recon steering files do.
I’m not sure why it doesn’t work. We should fix it.
What kind of events are you looking at?
Are you fully up to date with the latest changes or are you using a release?
What simulated slcio file are you running on? Did you generate the SLIC (Geant4) output file by yourself?
What “readout” simulation are you running? The default is to run the HPS2014ReadoutToLcio or HPS2014ReadoutNoPileup.
Can you send us the steering file for the recon (in case it’s different from the default HPS2014OfflineRecon)?
Can you send or point us to a file that we can test?
I can try running it when I get a chance (traveling to jlab right now).
Cheers,
Pelle
Sorry for this going into the weekend. I didn't really get a chance before now...
I apparently fixed the problem I was having with finding tracks, but wanted to make sure it was valid and not just ad hoc voodoo that doesn't make sense from a simulation standpoint.
Long story short:
I traced the problem to SiTrackerHits not being assigned in 'DataTrackerHitDriver', and replacing it with 'TrackerDigiDriver' fixed it.
TrackerDigiDriver is apparently supposed to be used with TrackerReconDriver anyway, according to a comment, so what is the reason for DataTrackerHitDriver being used in its place for 2014Recon?
(This would be a great learning experience for me)
Brad
Use REPLY-ALL to reply to list
