From owner-chemistry@ccl.net Fri Jan 5 06:08:01 2007 From: "Cory Pye cpye[]crux.smu.ca" To: CCL Subject: CCL:G: G03:freq calculations Message-Id: <-33331-070104191957-17120-SgnnlJ1vuYmegTKSZCjMEw###server.ccl.net> X-Original-From: Cory Pye Content-Type: TEXT/PLAIN; charset=US-ASCII Date: Thu, 4 Jan 2007 19:38:39 -0400 (AST) MIME-Version: 1.0 Sent to CCL by: Cory Pye [cpye*o*crux.smu.ca] Hello, I seem to recall that there is a utility in Gaussian called freqmem that you can use to calculate the optimum amount of memory for your particular system. I think it works for HF (and possibly B3LYP as well). If this value is less than the amount of physical memory on your computer, than you can use it. I routinely use %mem=512Mb, which on a shared-use 10 processor SMP with 5Gb memory running in serial mode seems to work fine. -Cory On Wed, 3 Jan 2007, errol lewars elewars:trentu.ca wrote: > > You did not say what your molecule is or what your program is (it looks > like Gaussian). If applicable, you could try running the molecule within > its symmetry (rather than ignoring symmetry). You could also try using > more (or less!) memory and see what happens. > > Very long times for frequencies are "reasonable". > > soumya s soumya_samineni_._rediffmail.com wrote: > > >Dear people, > >Naive question ... sry. > >I try to run a freq calculation on porphyrin > > > >%nproc=8 > >%mem=8MW > >opt freq b3lyp/6-31g* nosymm > > > >0 1 > > > >and the frequency calculation took almost 10hrs (is this resonable!!).. how can i make the calculations run better (other than increasing the no. of procs..) > > ************* ! Dr. Cory C. Pye ***************** ! Associate Professor *** ** ** ** ! Theoretical and Computational Chemistry ** * **** ! Department of Chemistry, Saint Mary's University ** * * ! 923 Robie Street, Halifax, NS B3H 3C3 ** * * ! cpye^_^crux.stmarys.ca http://apwww.stmarys.ca/~cpye *** * * ** ! Ph: (902)-420-5654 FAX:(902)-496-8104 ***************** ! ************* ! Les Hartree-Focks (Apologies to Montreal Canadien Fans)