Hi!
Basically the subject says it all.
I am trying to get Smina working on the GTX980 card with CUDA 7.5 toolkit.
I was able to build it successfully (mostly just had to adjust the include and library paths and #includes to the new CUDA directory layout).
However, at run time it gets stuck at the "Analyzing the binding site" (cache::populate call) stage seemingly indefinitely. What is strange is that no GPU operations are evident in the code, and the CPU version works just fine.
Do you have any ideas?
Thank you
Last edit: ZhenyOK 2016-05-22
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
The GPU code in smina isn't functional. We have a fork (https://github.com/gnina) that will be GPU accelerated, but it isn't ready yet either. Hopefully we'll have a release by the end of summer 2016.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi!
Basically the subject says it all.
I am trying to get Smina working on the GTX980 card with CUDA 7.5 toolkit.
I was able to build it successfully (mostly just had to adjust the include and library paths and #includes to the new CUDA directory layout).
However, at run time it gets stuck at the "Analyzing the binding site" (cache::populate call) stage seemingly indefinitely. What is strange is that no GPU operations are evident in the code, and the CPU version works just fine.
Do you have any ideas?
Thank you
Last edit: ZhenyOK 2016-05-22
The GPU code in smina isn't functional. We have a fork (https://github.com/gnina) that will be GPU accelerated, but it isn't ready yet either. Hopefully we'll have a release by the end of summer 2016.
Thank you for the clarification! We'll be waiting for it!
Hello:
Has there been an update re gnina?
BTW, thank you very much for a great program.
Best,
Markus
We're working on getting a release out the door.. not ETA as of yet.