[gmx-developers] NVIDIA misrepresenting simulation package performance

David van der Spoel spoel at xray.bmc.uu.se
Tue May 24 08:47:44 CEST 2016


my attention was brought to this page by a colleague:
Here, NVIDIA gives the relative performance of different codes on their 
GPUs. As we know, GROMACS is very efficient on CPUs and therefore the 
scale-up is not as impressive as other codes.

Should we try to have NVIDIA post a fairer picture (ns/day)?

Or debunk it on our own website?

David van der Spoel, Ph.D., Professor of Biology
Dept. of Cell & Molec. Biol., Uppsala University.
Box 596, 75124 Uppsala, Sweden. Phone:	+46184714205.
spoel at xray.bmc.uu.se    http://folding.bmc.uu.se

More information about the gromacs.org_gmx-developers mailing list