[gmx-users] why not double precision?

[Szilárd Páll]

On Mon, Oct 22, 2012 at 9:32 AM, Albert <mailmd2011 at gmail.com> wrote:

> hello:
>
> I found that the GPU version doesn't support double precison.
>
> double precision build
>
> $ cmake -DGMX_DOUBLE=ON ../gromacs-src
>
> Note that GPU acceleration is not compatible with double precision builds.
>
>
> I am just wondering why. I am trying to turn the GPU and double ON at the
> same time, but


Because a lot of the hardware out there has crippled double precision
performance. NVIDIA limits the double precision throughput of all, but some
of the most high-end Tesla GPUs to 1/24th of the single precision
throughput.


> the GPU doesn't compiled. Is it possible to solve this issue?
>

Yes, if somebody contributes double precision CUDA kernels.

Cheers,
--
Szilárd


> thank you very much
> Albert
> --
> gmx-users mailing list    gmx-users at gromacs.org
> http://lists.gromacs.org/**mailman/listinfo/gmx-users<http://lists.gromacs.org/mailman/listinfo/gmx-users>
> * Please search the archive at http://www.gromacs.org/**
> Support/Mailing_Lists/Search<http://www.gromacs.org/Support/Mailing_Lists/Search>before posting!
> * Please don't post (un)subscribe requests to the list. Use the www
> interface or send it to gmx-users-request at gromacs.org.
> * Can't post? Read http://www.gromacs.org/**Support/Mailing_Lists<http://www.gromacs.org/Support/Mailing_Lists>
>