In the latest version of
GAMESSPLUS (v2008-2) the MAX size of = the CM2=20 work array is defined by
the=20 = formula:
MXSATM*(MXSATM+1)+3*MXSBAS*(MXSBAS+1)/2+2*MXSBAS*MXSBAS+1
= where=20 MXSATM is the maximum number of atoms and MXSBAS is the maximum = number
of=20 contracted basis functions permissible for solvation calculations = using
SMx=20 models in GAMESSPLUS.
C CURRENT GAMESSPLUS LIMITATIONS FOR = SOLVATION=20 MODULE:
C ON TOTAL NUMBER OF ATOMS IS MXSATM=3D250
C = ON TOTAL=20 NUMBER OF CONTRACTED BASIS FUNCTIONS IS MXSBAS=3D2500.

Thus=20 = MAX=3D250*(250+1)+3*2500*(2500+1)/2+2*2500*2500+1=3D
=3D21,941,501>= 10,282,931

In=20 the latest version of GAMESSPLUS (v2008-2) the limits should = suffice.
http://comp.chem.umn.edu/ga= messplus/

Perhaps=20 Hui uses an older version of GAMESSPLUS with smaller
limits on = MXSATM=20 and
MXSBAS.

Chris

On Feb 7, 2009, at 3:09 PM, Jiabo Li jiaboli ~~ yahoo.com = wrote:

--

Christopher J. Cramer

University of Minnesota

Department of Chemistry

207 Pleasant St. SE

Minneapolis, MN 55455-0431

--------------------------

Phone:  (612) 624-0859 || = FAX:  (612) = 626-2006

Mobile: (952) 297-2575

cramer a = umn.edu

http://pollux.chem.umn.edu/~c= ramer

(website includes information about the textbook = "Essentials

  of Computational=20 Chemistry:  Theories and Models, = 2nd=20 Edition")

------=_NextPart_000_001F_01C9959B.3F02FE00-- From owner-chemistry@ccl.net Mon Feb 23 11:38:01 2009 From: "Richard Leo Wood woodx278]*[umn.edu" To: CCL Subject: CCL: Any molecular mechanics model good at pi-pi stacking calculation? Message-Id: <-38711-090223113629-881-bP80qyjeyaNQQXdqqsnzsg||server.ccl.net> X-Original-From: "Richard Leo Wood" Date: Mon, 23 Feb 2009 11:36:26 -0500 Sent to CCL by: "Richard Leo Wood" [woodx278=-=umn.edu] I think a lot of this depends upon the type of molecules that you have. One should do a validation study before choosing a particular functional to use for your set of molecules. Be aware that if you have sulfur containing molecules, straight DFTB is not a good method, but its "followup", DFTB+ works very well. Richard > From: owner-chemistry(!)ccl.net > To: rwoodphd(!)msn.com > Subject: CCL: Any molecular mechanics model good at pi-pi stacking calculation? > Date: Mon, 23 Feb 2009 11:51:38 +0000 > > Dear Alex > > For semi-empirical, standard methods are unlikely to work well at all. > However, there are a number of efforts to include the missing dispersion > effects - see for example: > SCC-DFTB-D: Elstner, M.; Hobza, P.; Frauenheim, T.; Suhai, S.; Kaxiras, E. > J. Chem. Phys., 2001, 114, 5149. > AM1-D and PM3-D: J.P. McNamara, I.H. Hillier, PCCP, 2007, 9, 2362. > OMx-D: T. Tuttle, W. Thiel, PCCP, 2008, 10, 2159. > > 200 atoms is probably not beyond the scope of modern DFT methods, > especially if you use the resolution of identity (RI) or density fitting > (DF) approximations which can speed things up dramatically. There are many > dispersion-corrected DFT methods now available in a number of packages. > Some DFT methods can also give good descriptions of stacking, for instance > the M05- and M06- familes from Truhlar's group, Becke's half-and-half > (BHandH) - even LDA isn't bad! > > Most molecular mechanics forcefields should include non-bonded terms that > represent pi-pi stacking, although details will depend on how close your > system is to those used to parameterize the forcefield. > > Hope this helps, > Jamie >

Hi Hui,

I assume that you compiled GAMESS with GAMESSPLUS for SMx = solvation model. The message you got indicates that the = scratch space=20 needed for your calculation is larger than the fixed scratch = space for=20 SMx calculations. The only way to solve this problem is to = modify the=20 scratch space parameter the GAMESSPLUS source code and = recompile it.=20

Here is the steps that you can try.

1. Find the source file smx.src in the GAMESSPLUS = package.

2. Do a global change of MAXWK=3D5500000 into = MAXWK=3D110000000 in=20 smx.src

3. Recompile the smx and relink gamess to create a new = gamess=20 executable.

Please let me know if this solves your problem or = not.

Regards,

Jiabo

--- On Fri, 2/6/09, Yang, Hui=20 hyang3{=3D}UTNet.UToledo.Edu=20 <owner-chemistry=3D-=3Dccl.net> = wrote:
From:=20 Yang, Hui hyang3{=3D}UTNet.UToledo.Edu=20 <owner-chemistry=3D-=3Dccl.net>
Subject: CCL: = GAMESS - "THE=20 WORK ARRAY FOR CM2 IS NOT SUFFICIENT"
To: "Li, Jiabo "=20 <jiaboli=3D-=3Dyahoo.com>
Date: Friday, February 6, = 2009, 12:54=20 PM

Hi,

I am trying to use GAMESS to = calculate=20 the solvation free energy of a biological system which has = 584=20 electrons.

Here is the = error.

...
DIRECT SCF=20 CALCULATION, SCHWRZ=3DT   FDIFF=3DF
THE = WORK ARRAY=20 FOR CM2 IS NOT SUFFICIENT
MAX =3D 5600000 = REQUIRMENT =3D=20 10282931
EXECUTION OF GAMESS TERMINATED ABNORMALLY = AT Fri=20 Feb 6 15:08:45 2009
STEP CPU TIME=20 =3D      .95 TOTAL CPU TIME=20 =3D      674.0 (   11.2=20 MIN)
TOTAL WALL CLOCK = TIME=3D     =20 675.2 SECONDS, CPU UTILIZATION IS = 99.83%
22548143=20 WORDS OF DYNAMIC MEMORY USED
*** ERROR TERMINATION = *** IN=20 COMPUTE PROCESS     0
...

And = here is=20 the header to my input file:
$CONTRL SCFTYP=3DRHF=20 RUNTYP=3DENERGY=20 = ICHARG=3D0
        =20 MAXIT=3D500 NPRINT=3D-5 EXETYP=3DRUN $END
$SYSTEM = TIMLIM=3D6000=20 MWORDS=3D900 $END
$SCF DIRSCF=3D.TRUE. = FDIFF=3D.FALSE.=20 $END
$BASIS GBASIS=3DN31 NGAUSS=3D6=20 = NDFUNC=3D1
        =20 DIFFSP=3D.TRUE. POLAR=3DPOPLE $END
$GUESS = GUESS=3DHUCKEL=20 $END
$CM2 ISCRF=3D2=20 ICMD=3D8C1
      IAQU=3D1=20 = Dielec=3D78.0
$END
$DATA
HF/6-31+G*C1
= ;C=20 6.0 45.984800 111.010300 = 25.847100
......

Does=20 anyone know what does "THE WORK ARRAY FOR CM2 IS NOT = SUFFICIENT"=20 mean? And what is "MAX =3D 5600000"?

Thank you = for your=20 = time.
Hui

From:	"Alex Liu ylvn8_-_umkc.edu" <owner-chemistry+*+ccl.net>
To:	"Platts, Jamie " <platts+*+cardiff.ac.uk>
Date:	22/02/2009 05:03
Subject:	CCL: Any molecular mechanics model good at pi-pi stacking calculation?