CCC-0331: EGS4, Electron Photon Shower Simulation by Monte-Carlo

Program-name Package-ID Status EGS4 CCC-0331/01 Tested EGS4 CCC-0331/02 Obsolete EGS4 CCC-0331/03 Obsolete EGS4 CCC-0331/04 Obsolete EGS4 CCC-0331/05 Obsolete EGS4 CCC-0331/06 Obsolete EGS4 CCC-0331/07 Tested EGS4 CCC-0331/08 Tested

Machines used: Package-ID Orig.Computer Test Computer CCC-0331/01 IBM 3081 DEC VAX 11/780 CCC-0331/07 UNIX gen. W.S. UNIX gen. W.S. CCC-0331/08 IBM PC PC Pentium 200

3. DESCRIPTION OF PROGRAM OR FUNCTION - The EGS code system is one of a chain of three codes designed to solve the electromagnetic shower problem by Monte Carlo simulation. This chain makes possible simulation of almost any electron-photon transport problem conceivable. The structure of the system, with its global features, modular form, and structured programming, is readily adaptable to virtually any interfacing scheme that is desired on the part of the user.
EGS4 is a package of subroutines plus block data with a flexible user interface. This allows for greater flexibility without requiring the user to be overly familiar with the internal details of the code. Combining this with the macro facility capabilities of the Mortran3 language, this reduces the likelihood that user edits will introduce bugs into the code. EGS4 uses material cross section and branching ratio data created and fit by the companion code, PEGS4.
EGS4 allows for the implementation of importance sampling and other variance reduction techniques such as leading particle biasing, splitting, path length biasing, Russian roulette, etc.

4. METHOD OF SOLUTION - EGS employs the Monte Carlo method of solution. It allows all of the fundamental processes to be included and arbitrary geometries can be treated, also. Other minor processes, such as photoneutron production, can be added as a further generalization. Since showers develop randomly according to the quantum laws of probability, each shower is different. We again are led to the Monte Carlo method.

5. RESTRICTIONS ON THE COMPLEXITY OF THE PROBLEM - None noted.

6. TYPICAL RUNNING TIME - The running time is dependent upon the problem being solved.

CCC-0331/01:

NEA-DB ran most of the test cases included in this package on VAX-11/780 and IBM 3090 computers. Execution times for the test cases together with cross section preparation with PEGS4 were up to 30 minutes on VAX.

CCC-0331/07:

This version is the UNIX Version 3.0. It was installed and the sample problems have been executed by the NEA Data Bank in the following systems:
1) DEC 3000 Model 300X (Alpha architecture workstation) under DEC OSF/1 V3.2
2) Sun Solaris (UNIX System V Release 4.0 (psisun)) under SunOS 5.3 The following sample problems and benchmark programs are provided with this package:
tutor1    \
tutor2    |
tutor3    |
tutor4    \___> seven tutorial cases
tutor4p   /
tutor5    |
tutor6    |
tutor7    /
xyzdos   - timing benchmark code (running on 2 sample problems)
dosrz    - general purpose dose scoring code for cylindrical-
           planar geometry
flurz    - general purpose fluence scoring code for cylindrical-
           planar geometry
xyzp     - running the xyzdos benchmark code using the PRESTA
           algorithm (PRESTA = Parameter Reduced Electron Step
           Transport Algorithm, special algorithm for electron
           transport)
The seven tutorial codes running time range from a few seconds to a few dozens of seconds. In the DEC-Alpha workstation, the benchmark code xyzdos, takes about 90 seconds to run the problem specified in the input file 'benchf' and about 920 seconds to run the problem specified in the input file 'benche' whereas the code xyzp, implementing the PRESTA algorithm, takes about 120 seconds to run the test case of input file 'benchf' and about 1460 to run the test case of input file 'benche'.

CCC-0331/08:

This version is the PC Version 3.1. It was installed and the sample problems have been executed by the NEA Data Bank in a PC DELL Optiplex GXPRO equipped with a Pentium PRO processor at 200 MHz,
total memory 64 MByte, under Windows NT 4.00 (execution under DOS).
    The following sample problems and benchmark programs are provided
with this package:
    tutor1    \
    tutor2    |
    tutor3    |
    tutor4    \___> seven tutorial cases
    tutor4p   /
    tutor5    |
    tutor6    |
    tutor7    /
    xyzdos   - timing benchmark code (running on 2 sample problems)
    dosrz    - general purpose dose scoring code for cylindrical-
               planar geometry
    flurz    - general purpose fluence scoring code for cylindrical-
               planar geometry
    xyzp     - running the xyzdos benchmark code using the PRESTA
               algorithm (PRESTA = Parameter Reduced Electron Step
               Transport Algorithm, special algorithm for electron
               transport)
The seven tutorial codes running time range from a few tens of seconds to a few seconds. The benchmark code XYZDOS, takes about 21.2 seconds to run the problem specified in the input file BENCHF.INP and about 220.9 seconds to run the problem specified in the input file BENCHE.INP; the CPU time required to execute the benchmark code XYZP on the input files BENCHF.INP and BENCHE.INP is of about 21.2 seconds and 447.9 seconds respectively. The code DOSRZ takes about 137.5 seconds to run the test case of input file CO60SPEC.INP (100000 histories) and about 152.3 seconds to run the test case of input file CO60MONO.INP (100000 histories).

7. UNUSUAL FEATURES OF THE PROGRAM -

8. RELATED AND AUXILIARY PROGRAMS - DATA LIBRARIES:
A variety of automated data sets (18 files) is provided.
EGS (Electron-Gamma Shower), based on code development in the early 1960s by H.H. Nagel, represents several years of additional development.

9. STATUS
CCC-0331/01: 15-JUN-1988 Tested at NEADB
CCC-0331/02: 31-OCT-1988 Obsolete
CCC-0331/03: 24-JUN-1998 Obsolete
CCC-0331/04: 29-DEC-1995 Obsolete
CCC-0331/05: 10-AUG-1998 Obsolete
CCC-0331/06: 15-JAN-1997 Obsolete
CCC-0331/07: 07-APR-1997 Tested at NEADB
CCC-0331/08: 29-JUL-1998 Tested at NEADB

10. REFERENCES  -
- Richard L. Ford and Walter R. Nelson
  The EGS Code System: Computer Programs for the Monte Carlo
  Simulation of Electromagnetic Cascade Showers, (Version 3)
  SLAC-210 (June 1978).
- Hideo Hirayama:
  Revision of the Sternheimer Density Effect Coefficients in PEGS4
  KEK Internal 95-17 (November 1995) R
CCC-0331/01:
- W.R. Nelson, H. Hirayama and W.O. Rogers:
  The EGS4 Code System
  SLAC-265 (December 1985).
- A. Bielajew:
  PRESTA Information
  Informal Notes (July 1987).
- W.R. Nelson and T.M. Jenkins:
  Writing SUBROUTINE HOWFAR for EGS4
  SLAC TN-87-4, SLAC, Stanford, California (August 1988).
- NEA-DB:
  Appendix to the EGS4 Report - Source: "Review of Particle
  Properties", by the Particle Data Group Physics Letters B239, 12
  April 1990
  NEA-DB (10-12-93).
- A. Bielajew et al.:
  History, Overview and Recent Improvements of EGS4
  SLAC-PUB-6499 (NRC-PIRS-0436, KEK Internal 94-4)
  (Revised June 1, 1994)
- A.F. Bielajew and D.W.O. Rogers:
PRESTA - The Parameter Reduced Electron - Step Transport Algorithm
for Electron Monte Carlo Transport
PIRS No.042 (July 1986)
- A. F. Bielajew:
README.general.3.0 (December 26, 1996)
- A. F. Bielajew:
Directions for EGS4 Installation (October 19, 1995)
CCC-0331/07:
- W.R. Nelson, H. Hirayama and W.O. Rogers:
  The EGS4 Code System
  SLAC-265 (December 1985).
- A. Bielajew:
  PRESTA Information
  Informal Notes (July 1987).
- W.R. Nelson and T.M. Jenkins:
  Writing SUBROUTINE HOWFAR for EGS4
  SLAC TN-87-4, SLAC, Stanford, California (August 1988).
- NEA-DB:
  Appendix to the EGS4 Report - Source: "Review of Particle
  Properties", by the Particle Data Group Physics Letters B239, 12
  April 1990
  NEA-DB (10-12-93).
- A. Bielajew et al.:
  History, Overview and Recent Improvements of EGS4
  SLAC-PUB-6499 (NRC-PIRS-0436, KEK Internal 94-4)
  (Revised June 1, 1994)
- A.F. Bielajew and D.W.O. Rogers:
PRESTA - The Parameter Reduced Electron - Step Transport Algorithm
for Electron Monte Carlo Transport
PIRS No.042 (July 1986)
- A. F. Bielajew:
README.general.3.0 (December 26, 1996)
- A. F. Bielajew:
Directions for EGS4 Installation (October 19, 1995)
CCC-0331/08:
- W.R. Nelson, H. Hirayama and W.O. Rogers:
  The EGS4 Code System
  SLAC-265 (December 1985).
- A. Bielajew:
  PRESTA Information
  Informal Notes (July 1987).
- W.R. Nelson and T.M. Jenkins:
  Writing SUBROUTINE HOWFAR for EGS4
  SLAC TN-87-4, SLAC, Stanford, California (August 1988).
- NEA-DB:
  Appendix to the EGS4 Report - Source: "Review of Particle
  Properties", by the Particle Data Group Physics Letters B239, 12
  April 1990
  NEA-DB (10-12-93).
- A. Bielajew et al.:
  History, Overview and Recent Improvements of EGS4
  SLAC-PUB-6499 (NRC-PIRS-0436, KEK Internal 94-4)
  (Revised June 1, 1994)
- A.F. Bielajew and D.W.O. Rogers:
PRESTA - The Parameter Reduced Electron - Step Transport Algorithm
for Electron Monte Carlo Transport
PIRS No.042 (July 1986)
- A. F. Bielajew:
README.general.3.0 (December 26, 1996)
- A. F. Bielajew:
Directions for EGS4 Installation (October 19, 1995)

11. MACHINE REQUIREMENTS - An IBM 3081 or a VAX computer is required for version CCC-0331/01. The PC version runs on a Compaq 80386/20 MHz with 4-5 Mbytes of memory (RAM).

CCC-0331/01:

To run the EGS4 test cases, peak page file size was 2200 on VAX-11/780; and 500K bytes on IBM 3090.

CCC-0331/07:

This version for UNIX systems have been installed and executed by the NEA Data Bank in the following systems: DEC 3000 Model 300X (Alpha architecture workstation) and on a Sun Solaris (UNIX System V Release 4.0 (psisun)) About 26 MByte of disk space are required to hold ALL the package files. (This amount includes the diskspace occupied by intermediate files for MORTRAN processing, FORTRAN compilation, .obj files, executable files, input files, data files, etc.) The diskspace allocated by the executable files is of about 3.8 MByte. The diskspace allocated by the PEGS data files is of the order of 3.8 Mbyte. The diskspace required by the PEGS the density-correction data files (for elements and compounds) is of about 2.5 Mbyte.

CCC-0331/08:

This version for PC systems has been installed and executed by the NEA Data Bank in a PC DELL Optiplex GXPRO equipped with a Pentium PRO processor at 200 MHz, total memory 64 MByte, under Windows NT 4.00 (DOS). About 20 MByte of disk space are required to hold ALL the package files. (This amount includes the diskspace occupied by intermediate files for MORTRAN processing, FORTRAN compilation, .obj files, executable files, input files, data files, etc.)

12. PROGRAMMING LANGUAGE(S) USED -
CCC-0331/01: FORTRAN-IV
CCC-0331/07: FORTRAN
CCC-0331/08: FORTRAN-77

13. OPERATING SYSTEM UNDER WHICH PROGRAM IS EXECUTED - A FORTRAN compiler is required. MORTRAN is a structured programming language that is implemented as a set of macros which are used by a macro- processor to translate the language into ANSI-standard FORTRAN ( a
preliminary step in the job). The resulting program is then run like any other FORTRAN program. The version /03 runs using the Lahey Fortran Compiler F77L/32/EM and A.I. Architect's operating system OS-386. The NDP Fortran and Pharlap Operating System can be used, but modifications are required. Both of these 32-bit compilers run in protected mode, making use of extended memory to exceed the 640 Kbyte limit of MS-DOS necessary for full support of EGS4; and, therefore, require their respective memory manager operating systems.

CCC-0331/01:

VMS V4.6 (VAX-11/780); MVS/XA with compiler VS FORTRAN LEVEL 2.1.1 (IBM 3090).

CCC-0331/07:

The Unix Version 3.0 of EGS4 was installed, compiled and the sample problems executed in the following systems and using the following compilers: 1) DEC 3000 Model 300X (Alpha architecture workstation) under DEC OSF/1 V3.2, using the DEC Fortran compiler V3.4 2) Sun Solaris (UNIX System V Release 4.0 (psisun)) under SunOS 5.3, using the FORTRAN 3.0 compiler from SUNPro

CCC-0331/08:

The PC Version 3.1 of EGS4 was installed, compiled and the sample problems executed in a PC DELL Optiplex GXPRO equipped with a Pentium PRO processor at 200 MHz, total memory 64 MByte, under Windows NT 4.00 (execution under DOS). The Lahey Fortran-77 compiler F77L/EM-32
Version 5.20 and linker (Version 5.1) were used for the compilation and creation of the PC-executables.

14. OTHER PROGRAMMING OR OPERATING INFORMATION OR RESTRICTIONS -

15. NAME AND ESTABLISHMENT OF AUTHORS -
Contributed by:
          Radiation Shielding Information Center
          Oak Ridge National Laboratory
          Oak Ridge, Tennessee, U. S. A.
Developed by:
          Radiation Physics Group
          Stanford Linear Accelerator Center
          Stanford University, Stanford, California
          National Laboratory for High Energy Physics (KEK)
          Oho-machi, Tsu-kuba-gun, Ibaraki-ken, Japan
          National Research Council of Canada, Ottawa, Canada
          Institute for Applied Physiology and Medicine, Seattle,
          Washington

17. CATEGORIES -

- J. Gamma Heating and Shield Design

Keywords: BREMSSTRAHLUNG, CHARGED PARTICLES, CHARGED-PARTICLE TRANSPORT, ELECTRONS, HIGH-ENERGY REACTIONS, MONTE CARLO METHOD, NUCLEAR CASCADES, PHOTON TRANSPORT, SHIELDING