Re: [gmx-users] pseudo-random algorithm

2014-11-04 Thread Johnny Lu
Thanks for answering my question. On Wed, Oct 29, 2014 at 6:36 PM, Roland Schulz wrote: > Hi, > > Gromacs < 5.0 uses a Mersenne twister. See gmx_random.h. Starting with 5.0 > it uses Random123 threefry2x64. Why do you think your results are affected > by the PRNG? > > Roland > > On Wed, Oct 29,

Re: [gmx-users] pseudo-random algorithm

2014-10-29 Thread Roland Schulz
Hi, Gromacs < 5.0 uses a Mersenne twister. See gmx_random.h. Starting with 5.0 it uses Random123 threefry2x64. Why do you think your results are affected by the PRNG? Roland On Wed, Oct 29, 2014 at 6:23 PM, Johnny Lu wrote: > Hi. > > Before I start to feel optimistic about my alanine tripeptid

[gmx-users] pseudo-random algorithm

2014-10-29 Thread Johnny Lu
Hi. Before I start to feel optimistic about my alanine tripeptide simulation, can I know which pseudo-random number algorithm does my installation of gromacs 4.6.6 use? It was installed on linux centos 5 with kernel 2.6.32-220.7.1.el6.x86_64, and the machine has 12 intel cpu and no gpu. I was u