Dear Colleagues,
Last week, we held this winter's second LCD simulation meeting at
Fermilab. At this moment, most of our projects are still in their
preliminary stages, but a large number of projects have been started
which we hope will produce results for the Sitges meeting. A list of
the talks we are preparing for Sitges is given in the accompanying
e-mail message. Please notify me of any omissions.
The scanned tranparencies from the Fermilab meeting are now available
on the Web at:
http://fnphyx-www.fnal.gov/conferences/nlc/9902/agenda.html
The rest of this message gives a status report on the simulation and
analysis tools. As of now, these are only partially complete. I will
give my understanding of the priorities toward producing results for
Sitges. Discussion of these priorities, requests for new tools, and
announcements of new tools, will take place in the lcd-sim email conference.
(To subscribe, send email to [log in to unmask] with the message
`subscribe lcd-sim'; to post a message, send email to
[log in to unmask])
1.) Full MC detector simulations
------------------------------
The first large-scale full MC runs turned up a number of glitches, but
these have now been corrected. Event samples for the model small and
large detectors are now being produced at SLAC (e+e- -> fermion pairs),
Penn (e+e- -> WW, ZZ), Michigan (e+e- -> ttbar), and Colorado
(e+e- -> se+se-). The one important event sample that is missing
is the gamma gamma background, which was to be run at Vanderbilt. Gary
Bower volunteered to work with Bob Panvini to begin this run. Our plan is
to put all of the full simulation output on a server at Penn so that it is
open to all for analysis. Don Benton plans at the end of February to set
up protocols for transmission of the data, and for formatting the data
in a manner compatible with the ROOT and JAS analysis tools.
At Fermilab, Richard presented single-particle studies of the calorimetric
resolution of the large and small detector models. Gary Bower has promised
to post calibration constants from this study on the simulation group Web
pages.
At Fermilab, we agreed not to tweak the two model detector designs further
before Sitges. However, if additional bugs are discovered in the full
simulation software, Richard may post a `version 1.0' at the time of the
Michigan meeting and ask that the data samples be rerun. We will discuss
at Michigan whether this will be necessary.
2.) Fast MC simulations
--------------------
The fast MC written by Richard Dubois' group is available but has not
yet been put to the test in any analysis. I strongly encourage groups
planning to use this software for Sitges to get busy experimenting with
it. Report any problems to the lcd-sim mailing list.
There are two important deficiencies with the fast MC program. At the
moment, the model small and large detector designs are wired in, although
the virtue of this program was supposed to be that one could use it to
study variations in the detector design. We are promised that it is straight-
forward to modify the detector geometry by actually going into the code.
People interested in doing this should contact Joanne Bogart
([log in to unmask]). I hope that some general instructions for this
modification can be posted to lcd-sim.
Second, at the moment, the fast MC returns smeared tracks but not yet track
error matrices. Bruce Schumm has promised to add this information to the
code.
For our attempt to prioritize work for Sitges, it is very important for us
to understand how many groups are planning to use the fast MC and what
capabilities they will require. If this facility is important to you,
please contact Joanne Bogart as soon as possible at the address above.
3.) General analysis tools
-----------------------
At Fermilab, we agreed to try to make tools available either for the
ROOT or the JAS analysis framework, whichever can be done most conveniently.
I suspect that people will be developing special purpose tools on an ad hoc
basis, and that the task of building a general-purpose library of analysis
tools will have to wait until after Sitges. However, it would be good to
alert the collaboration to new tools when they become available, by a
posting to the lcd-sim mailing list.
At the moment, we have an understanding of the tracking resolution of the
two detector models, based on Bruce Schumm's simulation results, but no
detailed track finding and fitting. At Fermilab, Rob Kutschke reported on a
Kalman filter for track fitting that could be extracted from the Fermilab
MCfast package and adapted to our analyses. Keith Riles offered to work with
Rob to make this available; however, this may not happen soon enough to be
relevant to studies for Sitges.
Many useful analysis tools, including cluster finders and a track-finding
package ported by Mike Ronan from the BABAR software, are included in the JAS
analysis package. Tony Johnson has promised to submit an update on the
tools available in JAS at the end of February.
At Fermilab, Howie Baer reported on the new release of ISAJET, which now
includes beamstrahlung and initial state radiation as well as initial-state
polarization. Michael Peskin reported on a new event generator, pandora,
which in principle provides a general framework for linear collider processes.
People who need an event generator with specific properties not available
from ISAJET or PYTHIA should contact Michael ([log in to unmask]).
4.) For the future
----------------
In our current system, it takes about a week to generate a full MC sample
for one model detector, distributing five classes of physics events to the
five sites now involved. This means that, once the system moves from its
current initial phase into routine production, we can explore variations
of the large and small detector models. We already agree that it would be
useful to simulate:
a.) A version of the L design with forward silicon disks and possibly
other improvements of forward tracking
b.) A very small S design, set up to reveal the limits of this strategy
Please think about other design options that should be studied, perhaps for
the next round of LCD studies beginning in the summer.
I look forward to seeing you all at the University of Michigan, March 26-28.
Good luck,
Michael
