this is fine and perfectly normal. different processors will produce
different
results after long periods.
===================================================================
Carlos L. Simmerling, Ph.D.
Assistant Professor Phone: (631) 632-1336
Center for Structural Biology Fax: (631) 632-1555
Stony Brook University Web: http://comp.chem.sunysb.edu/carlos
Stony Brook, NY 11794-5115 E-mail: carlos.simmerling.stonybrook.edu
===================================================================
Grace Li wrote:
>Dear Amber Users:
>
>I recently compiled AMBER 6 on my lab's cluster, I believe the cluster is
>running in a unix based operating system with a intel pentium 4 processor
>and g77 was used to compile the package.
>I did a test Molecular Dynamics simulation of 1500 NSTEPs and compared it
>to the output of the previous simulation that I did on a SGI
>supercomputer. I noticed that after 900 NSTEPs the results starts to
>differ as follows:
>
>from my lab cluster>>>>>>>>>>
> NSTEP = 2350 TIME(PS) = 44.700 TEMP(K) = 313.16 PRESS = 0.00
> Etot = -3229.0084 EKtot = 787.8359 EPtot = -4016.8443
> BOND = 2.8060 ANGLE = 2.9679 DIHED = 3.1331
> 1-4 NB = -0.1879 1-4 EEL = -261.5035 VDWAALS = 692.9282
> EELEC = -4456.9881 EHBOND = 0.0000 CONSTRAINT = 0.0000
> Ewald error estimate: 0.1298E-03
>
>from the SGI machine >>>>>>>>>
>
>NSTEP = 2350 TIME(PS) = 44.700 TEMP(K) = 306.94 PRESS = 0.00
> Etot = -3221.2988 EKtot = 772.1892 EPtot = -3993.4880
> BOND = 1.6549 ANGLE = 2.9067 DIHED = 3.3709
> 1-4 NB = -0.1205 1-4 EEL = -261.1509 VDWAALS = 693.3475
> EELEC = -4433.4965 EHBOND = 0.0000 CONSTRAINT = 0.0000
> Ewald error estimate: 0.1634E-04
>
>
>I was wondering if the differences in temperature here and and total
>energy (Etot) is too big and if I have done something wrong with the
>compilation?
>
>
>Thanks for your help,
>
>Grace
>-----------------------------------------------------------------------
>The AMBER Mail Reflector
>To post, send mail to amber.scripps.edu
>To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu
>
>
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber.scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu
Received on Fri Jul 30 2004 - 18:53:00 PDT
