Re: [gmx-users] same simulation, different results

2007-05-18 Thread Ansgar Esztermann
On Thu, May 17, 2007 at 03:01:54PM -0400, Jim Fonseca wrote: > Shutting off the fft optimization produced the same results for > 3.2.1. Then I tried running two simulations using version 3.3.1 with > the optimized fft and they differed. Your response seemed to imply > that the newer version

Re: [gmx-users] same simulation, different results

2007-05-17 Thread David van der Spoel
Jim Fonseca wrote: Shutting off the fft optimization produced the same results for 3.2.1. Then I tried running two simulations using version 3.3.1 with the optimized fft and they differed. Your response seemed to imply that the newer version should produce identical results. Could you clear

Re: [gmx-users] same simulation, different results

2007-05-17 Thread Jim Fonseca
Shutting off the fft optimization produced the same results for 3.2.1. Then I tried running two simulations using version 3.3.1 with the optimized fft and they differed. Your response seemed to imply that the newer version should produce identical results. Could you clear this up for me?

Re: [gmx-users] same simulation, different results

2007-05-16 Thread David van der Spoel
Stéphane Téletchéa wrote: Jim Fonseca a écrit : Thanks a lot--that was it. On May 15, 2007, at 12:42 PM, David van der Spoel wrote: Furthermore for a real test case, i would use a monoprocessor run since nic interfaces and smp can introduces little artifacts (roundoffs) leading to divergen

Re: [gmx-users] same simulation, different results

2007-05-16 Thread Stéphane Téletchéa
Jim Fonseca a écrit : Thanks a lot--that was it. On May 15, 2007, at 12:42 PM, David van der Spoel wrote: Furthermore for a real test case, i would use a monoprocessor run since nic interfaces and smp can introduces little artifacts (roundoffs) leading to divergence after a few ps, as repor

Re: [gmx-users] same simulation, different results

2007-05-16 Thread Jim Fonseca
Thanks a lot--that was it. On May 15, 2007, at 12:42 PM, David van der Spoel wrote: Jim Fonseca wrote: Hi, Does anyone know what would cause a simulation to change if I run it a second time? I'm starting with an EM structure and just doing a quick 10 ps simulation. I've found that (in on

Re: [gmx-users] same simulation, different results

2007-05-15 Thread David van der Spoel
Jim Fonseca wrote: Hi, Does anyone know what would cause a simulation to change if I run it a second time? I'm starting with an EM structure and just doing a quick 10 ps simulation. I've found that (in one case) the energies start to differ somewhere between step 200 and step 250--the Vir an

[gmx-users] same simulation, different results

2007-05-15 Thread Jim Fonseca
Hi, Does anyone know what would cause a simulation to change if I run it a second time? I'm starting with an EM structure and just doing a quick 10 ps simulation. I've found that (in one case) the energies start to differ somewhere between step 200 and step 250--the Vir and Pres values d