On Sun, Oct 18, 2020, Kehinde Idowu wrote:
>Please, I am runing MD on Amber 18, after I have ran for some minute I
>could not run the MD further, I got this message on my md_output file
>'cudaGetDeviceCount failed unknown error' and 'Note: ig = -1. Setting
>random seed to 138448 based on wallclock time in microseconds. |
>irandom = 1, using AMBER's internal random number generator (default)' in
>my md.out file. Kindly advise on how to solve this issue.
Please look at the entire error message and output file. It's hard to
see how you would get a 'cudaGetDeviceCount failed' yet still continue
to run, but I suppose that is possible. Do a short run on a CPU to see
if any problems crop up, since error messages on the CPU are often more
helpful than those on the GPU.
To be of any more help, we would need to know some details about what
sort of MD you are running. Does the error happen right away (at step
0)? Is this the first time you have run MD on this system, or are you
continuing an earlier run? etc.
....dac
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Oct 19 2020 - 05:00:03 PDT
