KORALZ

We have discovered a bug in KORALZ. It affects next to leading log effects at most. Most severely events with hard high p_T photons which can match radiative return to Z configurations. I will know more about significance of this bug later.

So far I do not see any distribution where it contributes significantly. That is why it was so difficult to spot. Definitely the size of the effect was far smaller than the warning message from within the program was indicating. I mean the message:

 ***************************************************************************
 *              possible loss of precision !!!!!!                          *
 *                       VVMAX too close to 1D0 ?                          *
 *              reduce VVMAX in KORALZ parameters!                         *
 *                   it will faster program also !                         *
 *              WTMAX too small ?                                          *
 *              Increase it in SUBROURTINE MODEL!                          *
 *              MANY events of negative weights?                           *
 *              Contact JADACH at CERNVM,                                  *
 *              But check folowing weights first                           *
 ***************************************************************************
....
 ***************************************************************************
 *         ratio averwt(id=70)/averwt(id=79)                               *
 *         measures severity of error                                      *
 ***************************************************************************

The bug affects however stability of the generation. Once removed, program works much better and in particular under/overweighted events as well as the above warning messages disappear.

Best greetings Zbyszek Was

Here is affected but now corrected routine:

===============================================================
      SUBROUTINE NDIST0(XX,P1,P2,Q1,Q2,ANDIS,DELI1,DELI2,DELF1,DELF2)
C     ***************************************************************
C Provides elements of beta0,             
C for transparency reasons the full reduction of momenta is done.
C     *********************************** 
      IMPLICIT DOUBLE PRECISION (A-H,O-Z) 
      COMMON / KEYYFS / KEYZET,KEYBRM,KEYFIX,KEYRED,KEYWGT   
      SAVE   / KEYYFS /
      DIMENSION XX(*),P1(*),P2(*),Q1(*),Q2(*)                
      DIMENSION PR1(4),PR2(4),QR1(4),QR2(4)                  
C   
      CALL REDUZ0(XX,P1,P2,PR1,PR2)       
      CALL REDUZ0(XX,Q1,Q2,QR1,QR2)       
      CALL GTHET0(PR1,QR1,COSTH1)         
      COSTH  = (PR1(1)*QR1(1) +PR1(2)*QR1(2) +PR1(3)*QR1(3)) 
     $            /SQRT((QR1(1)**2 +QR1(2)**2 +QR1(3)**2)    
     $                 *(PR1(1)**2 +PR1(2)**2 +PR1(3)**2))   
      SVAR1  = XX(4)**2-XX(3)**2-XX(2)**2-XX(1)**2
C ZW 03.03.97 here was bug 
C      SVAR1  = XX(4)**2-XX(3)**2-XX(2)-XX(1)**2              
      ANDIS  = BORNV (SVAR1,COSTH )       
      CALL BVIRT0(P1,P2,DELI1,DELI2)      
      CALL BVIRT0(Q1,Q2,DELF1,DELF2)      
C ...Initial/final state bremsstrahlung switches             
      KEYBIN  = MOD(KEYBRM,10)            
      KEYBFI  = MOD(KEYBRM,100)/10        
      DELI1   = DELI1*KEYBIN              
      DELI2   = DELI2*KEYBIN              
      DELF1   = DELF1*KEYBFI              
      DELF2   = DELF2*KEYBFI              
      END              
===============================================================

Source codes:

• KORALZ source code of the last version 4.02: korale-new-exp.tar.gz; 1,524kB. With respect to version 4.01 some modifications were introduced, they include: Dizet version 5.0 electroweak library was introduced. Maximum allowed center of mass energy was upgraded to 240 GeV. For more details see: koralz4-2.ps.gz;
• KORALZ source code of the version 4.01: korale-exp.tar.gz; 1,524kB. With respect to standard version 4.0 some modifications necessary for tests-comparisons with KORALB were introduced. Also maximum of center of mass energy was moved to LEP 2 range. Necessary in this case updates of electroweak libraries and tests were not yet done. After their completion new version of KORALZ will be introduced.

Write-ups and papers:

• S. Jadach, B.F.L. Ward, Z. Was,
  The Monte Carlo program KORALZ version 4.0 for the lepton or quark pair production at LEP/SLC energies Comp. Phys. Commun. 79 (1994) 503.
  (CERN-TH-7075 preprint, 1993); th-7075.ps.gz.
• S. Jadach, B.F.L. Ward, Z. Was,
  The Monte Carlo program KORALZ version 3.8 for the lepton or quark pair production at LEP/SLC energies,
  Comp. Phys. Commun. 66 (1991) 276, (CERN-TH-5994 preprint); th-5994.ps.gz.
• S. Jadach,B.F.L. Ward,
  YFS2 - The second order Monte Carlo for fermion pair production at LEP/SLC with the initial state radiation of two hard and multiple soft photons,
  Comp. Phys. Commun., 56 (1990) 351; tpju-15-88.ps.gz; (missing figures).
• S. Jadach,
  Yennie-Frautschi-Suura soft photons in Monte Carlo event generators,
  MPI-PAE/PTh 6/87, preprint of MPI-München, January 1987, unpublished; mpi-pae-pth-6-87.ps.gz; (complete).
• See also section PAPERS of our home page