Proposal for a Simulation at PHENIX Computing Center in Japan (CC-J)
                                         Date: 11/07/00   (DD/MM/YY)

To: the Planning and Coordination Office of CC-J
Title : Caluculation of Gamma Acceptance of EMCal and Generation of pi0
        Decayed Gamma using Revised FAST Monte Carlo Code

PWG : Thermal Photon Working Group

Liason: Center for Nuclear Study, University of Tokyo

Name : Takao Sakaguchi

Tel/Fax :+81-3-5841-4620/+81-3-5841-7642

E-mail : takao@phenix.cns.s.u-tokyo.ac.jp

Person responsible for the job submission:

Name :

Tel/Fax :

E-mail :

Person responsible for the quality assurance:

  Name :

  Tel/Fax :

  E-mail :

Estimated CPU time required :      hours (with PentiumIII 450MHz)

    Generation of 10000 Events takes 30 min. 100,000 events needed for one
Condition. Totally around 100 conditions will be needed. Jobs will be run
frequently, but not continuously.

Estimated disk space usage required :    At most 100GB
Estimated HPSS space usage required :    At most 200GB
Target date for completion :  30/12/00 (DD/MM/YY)
Purpose and merit of the simulation including kinds of events being simulated :

   The goal of the thermal photon working group is to obtain the spectrum of
direct photon produced in the hot dense matter. Since the background gamma
decayed from pi0 or eta is significant, precise estimate of invariant cross
section of these mesons are essentially needed.
   So far, the analysis code for deducing invariant mass spectrum or Gamma
PID are developed and studied. However, the geometrical acceptance is not
know precisely, which can be studied using Revised Fast Monte Carlo Software
that is originally developed in FORTRAN and now being re-written in C++.
The software is also used for generation of decayed gamma from mesons, that
will be subtracted from inclusive gamma spectrum to obtain direct photon
spectrum. Generation of decayed gamma is needed to check the obtained pi0
spectrum as well. 

This simulation will need huge events to reduce statistical error
as low as possible, and will also be needed to run in several
conditions. The CPU time estimated above is somehow rough.