GU: soci trudge

[David J Heisterberg <dheister@magnus.acs.ohio-state.edu>]

Dear Gamess users,

I'd like to do an MRCI geometry optimization, but I'm not having
much success, I hope I'm simply doing something wrong.  Relevant
parts of the input file are:

 $CONTRL SCFTYP=MCSCF RUNTYP=TRUDGE EXETYP=RUN ICHARG=0 MULT=4
 $DRT  GROUP=C2V FORS=.TRUE. NMCC=9 NDOC=3 NALP=3 NVAL=0 $END
 $DRT2 GROUP=C2V SOCI=.TRUE. NFZC=9 NDOC=3 NALP=3 NVAL=0 NEXT=-1 $END
 $TRUDGE OPTMIZ=GEOMETRY CIDRT=DRT2 NPAR=2 IEX(1)=21,22 $END

The MCSCF (6 CSFs) goes ok, but when it gets to the CI (46564 CSFs)
it does the integral transformation and formula tape construction,
but aborts at the diagonalization with:

 PROBLEM WITH CI VECTOR FILE
 NCSFS SHOULD MATCH NLARGE 6 46564
 NSTATE SHOULD MATCH IDHIRT 1 1
 EXECUTION OF GAMESS TERMINATED ABNORMALLY AT Wed Apr 17 11:55:59 1996

I'm using the 22 NOV 1995 version on a HAL 3500 (64 bit Sparc).

I'd appreciate any suggestions.

Dave Heisterberg