On Wed, Feb 21, 2018, Baker D.J. wrote:
>
> Is it still true that it isn't really worth building the CUDA
> enabled version with the Intel compilers?
That is still correct. The reason is that the entire calculation
(almost!) is done on the GPU, so there is no advantage in trying to
speed up the CPU part. In fact, Amber developers rarely test CUDA with
Intel.
Not sure if you got any answer through another channel. If not,
apologies for the delay.
...dac
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Mar 05 2018 - 05:30:04 PST
