 |
 |
I had reloaded JAS and lcd.jar when I tried your code. The
example that Tony sent me also works now. I do not
know what dumb thing I had done before, but it seems to have been
corrected. Thanks to both of you.
I do have a technical problem with the smearing routine. It crashes
my eventloop after 195 WW events or 995 ttbar events using
the S2 detector. For L2 it runs through 1000 of each without problems.
(FYI: On my 650 MHz laptop - for 1000 ttbar events (quite high multiplicity),
it takes about 104 seconds with no analysis call;
44 secs for 1000 WW events).
hep.lcd.util.error.LCDException: Error matrix not positive definite!
at hep.lcd.mc.fast.SmearTrack.smearTrack(SmearTrack.java:22)
at hep.lcd.mc.fast.ReconTrack.<init>(ReconTrack.java:54)
at hep.lcd.mc.fast.MCFast.process(MCFast.java:62)
Both stdhep files were generated with the latest Pandora/Pythia releases.
Is there a way to trap this exception so the
job can move on to the next event? Is this wise?
Bob
Wolfgang Walkowiak wrote:
Hi Bob,I am currently running Jas 2.2.1 with lcd.jar 1.2 on my linux laptop
without any problems, at least not those you experience.Since lcd.jar 1.2 is is necessary that you explicitly select either the S2
or the L2 detector design in your driver's constructor with:
Detector.setCurrentDetector(new Detector("S2"));
in your driver's constructor(!).
The appropriate (old) ini files are supplied with the same version of
lcd.jar which contains the new MCFast with full cov. track smearing.The appropriate .ini-files are contained in lcd.jar.
Here an example from one of my programs:
import hep.analysis.*;
import hep.physics.*;
import hep.lcd.event.*;
import hep.lcd.mc.fast.*;
import hep.lcd.util.driver.*;
import hep.lcd.geometry.*;
import java.util.*;
import java.io.*;final public class TrackVtxTest extends Driver
{
public TrackVtxTest() throws IOException
{
// set the detector -- necessary for stdhep input
Detector.setCurrentDetector(new Detector("S2"));// initialize MCFAst, select beamSpotConstraint
add(new MCFast(MCFast.noBeamSpotConstraint));// add analysis
add(new TrackVtxCheck()); // analysis routine
}
}class TrackVtxCheck extends AbstractProcessor
{
....
}I have run that program about 5 min ago on a setup similar to yours. Pay
attention that the detector is selected before calling the constructor of
MCFast -- the constructor already reads the resolution tables. :-(
The argument to the MCFast constructor
(MCFast.noBeamSpotConstraint/MCFast.withBeamSpotConstraint) is optional.We should probably make "S2" or "L2" default ...
As with lcd.jar 1.3 the resolution tables will be updated to account for
Bruce Schumm's bug fix, which he doesn't expect to show a large effect.I hope this helps. Let me know, if there are more questions. ( I have
recently worked with this code and included the full covariance matrix for
track smearing into the existing structure.)w.w.
On Tue, 10 Oct 2000, Robert J. Wilson wrote:
> Hello,
>
> I am unable to run FastMC in JAS 2.2.1 with lcd.jar 1.2 on a linux
> machine (despite help from Tony). The latest complaint is when I try to
> load
> (it compiles okay) -- see below. It cannot find a ResolutionTableFile.
> Perhaps all I need are the correct resolution and/or parameters files in
> the correct path. Can anybody tell me what additional files I need and
> where I should put them?
> Thanks,
> Bob
>
>
> Error occured during instantiation of CheckFast
> hep.lcd.util.error.LCDException: ResolutionTableFile not found! Check
> for detector settings. Only L2 and S2 work in this version!
> at
> hep.lcd.mc.fast.TrackResolutionTables.<init>(TrackResolutionTables.java:29)
>
> at hep.lcd.mc.fast.MCFast.init(MCFast.java:34)
> at
> hep.lcd.util.driver.AbstractProcessor.setContext(AbstractProcessor.java:44)
>
> at hep.lcd.util.driver.Driver.add(Driver.java:39)
> at CheckFast.<init>(CheckFast.java:13)
> at jas.job.AbstractJob.loadObject(AbstractJob.java:184)
>
>
> --
> Robert J. Wilson
> Professor of Physics
> Department of Physics
> Colorado State University
> Fort Collins, CO 80523
> Phone: (970) 491 5033
> Fax: (970) 491 7947
> Email: [log in to unmask]
>
> August 2000 - July 2001:
> ========================
> Institut de Fisica d'Altes Energies
> Universitat Autonoma de Barcelona
> E-08193 Bellaterra (Barcelona)
> SPAIN
> Phone: (011 34) 93 581.28.32
> Email: [log in to unmask]
> ========================
>
>
>=============================================================================
Wolfgang Walkowiak Phone +1-(650)-926-3432 (@ SLAC)
Stanford Linear Accelerator Center
Mail Stop 94
P.O.Box 20450
Stanford , CA 94309 ! --> [log in to unmask]
USA [log in to unmask]=============================================================================
-- Robert J. Wilson Professor of Physics Department of Physics Colorado State University Fort Collins, CO 80523 Phone: (970) 491 5033 Fax: (970) 491 7947 Email: [log in to unmask] August 2000 - July 2001: ======================== Institut de Fisica d'Altes Energies Universitat Autonoma de Barcelona E-08193 Bellaterra (Barcelona) SPAIN Phone: (011 34) 93 581.28.32 Email: [log in to unmask] ========================