HI David, I don't know the answer for Joanne's cluster finder, but I did
mean to tell you that in the new release of hep.lcd (v 0.97) which was
released yesterday, the setThreshold method in SimpleClusterFinder does
now work (it now takes a double arg).
I looked at single 5GeV pions going in to the calorimeter, and with the
default parameters both SimpleClusterFinder and JRBClusterFinder produce
about 10 clusters/event. You can reduce the number of clusters found by
increasing the threshold in SimpleClusterFinder, but on first look that
seemed to have a fairly disastrous effect on the total energy of the
clusters, ie the small clusters that were thrown away by the energy cut
seemed to contribute a significant fraction of the energy of the pion.
This is obviously something that needs to be looked into further.
For the JAS problem, if it causes an error box to appear when you look
at the histogram can you tell me what the error message was, and if you
can push the traceback button and send me the traceback (which appears
in the console window from which JAS was run) that would be helpful too.
Also when data set did you run the tutorial example on.
Tony
> -----Original Message-----
> From: David L. Wagner, University of Colorado
> [mailto:[log in to unmask]]
> Sent: Monday, April 19, 1999 11:34 AM
> To: Joanne Bogart
> Cc: Tony Johnson
> Subject: JrbCluster threshold
>
>
> Hi Joanne,
>
> I was just wondering if there was a way of adjusting the threshold for
> JrbCluster's. I run on a sample of selectron pairs (you can
> find some of
> these files at the Penn server), so there are always two
> electrons in the
> final state and no other visible particles. Therefore one
> would expect
> exactly two EM clusters per event. When I run JrbCluster, I get an
> average of thirty, so it seems like the threshold is too low.
>
> Also, have you tried using the ClusterAnalyzer processor that's in the
> tutorial on the web? When I try using that and look at the histogram
> filled with 'histogram("R").fill(c.getEnergyRadius());' it
> seems to cause
> some sort of unrecoverable exception inside JAS (maybe this is more a
> question for Tony...).
>
> -David
>
