Perfect! Thanks
On Fri, Sep 19, 2014 at 7:37 AM, Marc van der Kamp <marcvanderkamp.gmail.com
> wrote:
> PS a few weeks ago - thread was called: "cuda performance advice"
>
> See here relevant answers from Ross Walker & Scott Le Grand:
>
> To add some info.
>
> CUDA 6.5 is supported as of last week for AMBER 14 - see update.6
>
> The 5% perf regression over cuda 5.0 still exists in 6.5 though so 5.0 is
> still the recommended version. I believe a fix for the perf regression is
> planned for cuda 7.0.
>
> All the best
>
> Ross
>
>
>
> On 9/2/14, 1:26 PM, "Scott Le Grand" <varelse2005.gmail.com> wrote:
>
> >The first rule of PMEMD Club is CUDA 5.0
> >
> >The second rule of PMEMD Club is CUDA 5.0.
> >
> >Were you thinking of building with 5.5, 6.0, or 6.5? OK. sure. We'll
> >throw in a 5-10% unfixable perf regression (compiler bug introduced in
> >5.5)
> >for no additional charge.
> >
> >Scott
> >
> >
>
> On 19 September 2014 13:34, Marc van der Kamp <marcvanderkamp.gmail.com>
> wrote:
>
> > You can search the amber mailing list - this was discussed about a week
> > ago.
> > --Marc
> >
> > On 19 September 2014 13:31, Carlos Sosa <sosa0006.r.umn.edu> wrote:
> >
> >> Hello,
> >>
> >> AMBER 14 benchmarks were done with CUDA 5.0. Does anybody know the
> impact
> >> in performance using CUDA 5.5 or 6.0 versus 5.0?
> >>
> >> Thanks
> >>
> >> carlos
> >> _______________________________________________
> >> AMBER mailing list
> >> AMBER.ambermd.org
> >> http://lists.ambermd.org/mailman/listinfo/amber
> >>
> >
> >
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Sep 19 2014 - 06:00:09 PDT
