[gmx-users] Simulation runs on iMac but explodes on cluster

[Luke Goodsell]

>
> I'd still be skeptical. 15000 steps is a long time for the trajectories to
> diverge. If you can run to 14550 with SSE and stop, and then show that
> continuing with SSE explodes and unoptimized kernels do not, then you have
> serious evidence of a bug. Diagnosing the cause of such a bug will need such
> a test case, anyway.
>

I will try to do this, but I will have to check that my collaborator is
happy to submit specific system information before submitting it as a test
case.

Luke
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20110713/2915f195/attachment.html>