Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

GPU processing not working #1

Open
500pixies opened this issue Sep 1, 2018 · 4 comments
Open

GPU processing not working #1

500pixies opened this issue Sep 1, 2018 · 4 comments

Comments

@500pixies
Copy link

I'm having trouble getting NanoJ SRRF to process my files. It doesn't seem to be able to start OpenCL and throws an error. It doesn't switch over to Java and the CPU though. It just stops. A bit frustrating because the initial (CPU-based) results were very promising before I updated everything and started trying to get the GPU to work.

Here is an example of the (not so informative) error:

sep 01, 2018 6:19:05 EM com.amd.aparapi.internal.kernel.KernelRunner executeOpenCL
WARNING: ### CL exec seems to have failed. Trying to revert to Java ###

The data is a small (cropped) 100 frame 16 bit stack from a Yokogawa spinning disk confocal. It is as close as I can make it to the sample data:
https://bitbucket.org/rhenriqueslab/nanoj-srrf/downloads/Anti-Tubulin-Alexa647%20Ultra-high%20density%20data.tif.zip
The sample data works - my data does not. I have attached a cropped but otherwise unmodified file. This particular example fades badly, but that shouldn't have anything to do with the error.

I would be most grateful for any suggestions as I have done everything I can think of.
Thanks!
AL

Sys info:
Win10 64 bit build 17134
Nvidia GeForce GTX 970 - most recent drivers 399.07, DirectX 12
Java - most recent release, version 8 update 181 (build 1.8.0_181-b13)
Fiji/ImageJ - most recent update 1.52e
NanoJ - most recent release, 2.0RC1
SRRF - most recent release 1.13stable1

Mandelbrot OpenCL test:
NVIDIA Corporation_GPU
Identity of device used: - NVIDIA Corporation_GPU
global:589824 local:(derived)1024
Execution mode=GPU
Executed at 535.0FPS
Best Device: - NVIDIA Corporation_GPU

Benchmark results:
ImageJ 512 512 610 3 3 57.908
NanoJ-JTP 512 512 610 3 3 51.583
NanoJ-GPU 512 512 610 3 3 132.019
ImageJ 512 512 610 3 3 59.008
NanoJ-JTP 512 512 610 3 3 53.319
NanoJ-GPU 512 512 610 3 3 184.720
ImageJ 512 512 615 3 3 58.598
NanoJ-JTP 512 512 615 3 3 65.116
NanoJ-GPU 512 512 615 3 3 181.586
small_stack.zip

@zach-hensel
Copy link

I had exactly the same issue today (1050 TI 4GB, x64 intel system, windows 10), found this post, and ended up resolving it by installing the latest "creator ready driver" from here - https://www.nvidia.com/Download/index.aspx?lang=en-us - I did not try the latest "game ready driver". I think the previous driver that did not work was whatever windows update installed upon detecting the card, but I am not sure; it was dated January 2019 and the driver from Nvidia is dated March 2019.

@XarlesSta
Copy link

I am having the same problem. (NVIDIA GeForce RTX 2070 super, Windows 10). I tried installing the 2 different new drivers available (Studio nad Game-ready-driver), but any of the two installation modes worked. I would appreciate any help or advice.

@zach-hensel
Copy link

zach-hensel commented May 8, 2020 via email

@XarlesSta
Copy link

Zach. thanks for the advice. I tried with both versions "Studio drivers" and the "Gamers drivers". Still not working. Version 442.92 from march-2019. There is a Creator driver-ready version 419.67 (Also march-2019), but it does not seem to support the "GeForce RTX 2070 super", only the "GeForce RTX 2070". That might be the problem.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants