Hi Jonathan,
This is fine - most of the differences are very small. The bigger
differences are due to differences in the random number generator between
GPU models. I am working on making the test cases more robust in this
respect.
If you want to check that the GPUs themselves are working properly
download and run the following validation suite:
https://dl.dropboxusercontent.com/u/708185/GPU_Validation_Test.tar.gz
All the best
Ross
On 7/11/14, 2:53 PM, "Jonathan Gough" <jonathan.d.gough.gmail.com> wrote:
>I just installed amber14 and ambertools14 on a new installation of centOS
>6.4
>
>Using
> NVIDIA-SMI 340.24 Driver Version: 340.24
>
>and Nvidia Toolkit 5.0
>
>2 GPU's
>[root.mdbox amber14]# nvidia-smi
>Fri Jul 11 17:47:48 2014
>+------------------------------------------------------+
>
>| NVIDIA-SMI 340.24 Driver Version: 340.24
>|
>|-------------------------------+----------------------+------------------
>----+
>| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr.
>ECC |
>| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute
>M. |
>|===============================+======================+==================
>====|
>| 0 GeForce GTX TITAN On | 0000:01:00.0 N/A |
>N/A |
>| 30% 36C P8 N/A / N/A | 97MiB / 6143MiB | N/A E.
>Process |
>+-------------------------------+----------------------+------------------
>----+
>| 1 GeForce GTX 780 On | 0000:04:00.0 N/A |
>N/A |
>| 26% 28C P8 N/A / N/A | 11MiB / 3071MiB | N/A E.
>Process |
>+-------------------------------+----------------------+------------------
>----+
>
>
>+-------------------------------------------------------------------------
>----+
>| Compute processes: GPU
>Memory |
>| GPU PID Process name
>Usage |
>|=========================================================================
>====|
>| 0 Not
>Supported |
>| 1 Not
>Supported |
>+-------------------------------------------------------------------------
>----+
>[root.mdbox amber14]#
>
>Persistence and Compute Exclusive Modes were set.
>
>configure was done using gnu
>
>however the cuda.test gave the following failures (see attached files)
>
>99 file comparisons passed
>26 file comparisons failed
>
>the errors are mostly small, but obviously things aren't exactly correct.
>
>Is it safe to proceed? Is there something I maybe missing?
>
>Thanks,
>Jonathan
>_______________________________________________
>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 Jul 11 2014 - 15:30:02 PDT
