Jump to content
HWBOT Community Forums
_mat_

GPUPI - SuperPI on the GPU

Recommended Posts

16 hours ago, cbjaust said:

Still crashes, I've uploaded two traces from the Problem Steps Recorder, one showing it crashing with hwinfo enabled and one attempting to save the result after a run. Cheers

gpu_hwinfo_crash.zip

Just disable HWinfo, there is a drop-down menu for that in the run window of GPUpi. It's legal for HWBot submissions.

Share this post


Link to post
Share on other sites
3 hours ago, unityofsaints said:

Just disable HWinfo, there is a drop-down menu for that in the run window of GPUpi. It's legal for HWBot submissions.

It still crashes when trying to save the result to a file, even with hwinfo disabled...

Share this post


Link to post
Share on other sites

Thank you for the detailed report, much appreciated.

These seem to be two unrelated errors. HWiNFO is not compatible with this system/platform, you can try "Safe Mode" though before disabling it.
The second problem is related to the gathering of various system information (mostly WMI) and/or result encryption. It's somehow related to these old sockets, another user had the same problem with an old Athlon 64 system as well. I will look into it, hope I still have test hardware here.

  • Thanks 1

Share this post


Link to post
Share on other sites

This platform - s478 P4 Prescott-1024M is a stage in Team Cup this year, (this result) seems OK but I have no idea what they're doing different, it looks like Windows 7...

I can get a result fine just can't get the result file or upload directly, in both cases GPUPI crashes.

1586684365_SL7E5Pentium43.2GHzBX80546PG3200EGPUPI_for_CPU_100M13m32.223sat3360-168450-1160GeForce6800GT.thumb.jpg.71a767d028f534fb99a114da5a30fc49.jpg

Share this post


Link to post
Share on other sites

OS is different, that's for sure. Graphics card as well, but I doubt that's the issue. Do you have a Windows 7 install lying around?

Edited by _mat_

Share this post


Link to post
Share on other sites

I am using Windows 7 x86... I wanted to run it in Server 2003 R2 x86 (Windows XP kernal) but I couldn't find an OpenCL that would install...

Share this post


Link to post
Share on other sites

I tried my socket 478 system again for Team Cup DDR Stage 3 and this time I tried 3.1.1 but it has the same issues with crashing on the HWINFO.dll module when starting the benchmark. So I deleted HWINFO.dll and the bench ran fine but at the end it had the same crash as 3.2 on module "StackHash_0a9e".

I also tried 3.2 again but trying to save still crashed with the "StackHash_0a9e" module.

Help!

Cheers

Edit: 2.3.4 works just fine, result file attached.

cbjaust_GPUPI_for CPU_100M_12m-49.578s.result

Edited by cbjaust
add 2.3.4 result file

Share this post


Link to post
Share on other sites

Not sure if this is the place to come, however I've made a score using GPUPI3.2 using the new OpenCL2.1.  Made the score saved the datafile, now however when I try to upload the score to HWbot I get "Invalid Data file : Unable to decrypt the datafile"  any ideas what the cause is?  Or if it can be sorted so I can upload the score?

Thanks

 

Nevermind, uploaded, must have selected the wrong benchmark perhaps

Edited by Matt26LFC

Share this post


Link to post
Share on other sites

have only one tab of hwbot opened, in a fresh browser. when you have several tabs of hwbot submission opened, it will bug

if mozilla, try chrome or ie

Share this post


Link to post
Share on other sites

Hi! I'm having trouble getting 3.3 to run on my G470, I've tried a few different batch sizes as well as hwinfo detection levels and it still crashes on run. Attached is a screenshot and error logs, I can provide any other info that you need. @_mat_

error.PNG

HWiNFO64.DBG

GPUPI.cfg

Edited by yosarianilives

Share this post


Link to post
Share on other sites

Do I see this right and amd does not offer standalone open cl sdk anymore and do I also see this right that Intel excludes a big number of countries that do not kiss usa ass like sadly my country does and therefore a lot of people of the community are at least in theory excluded from using this benchmark legally? I really hope I am wrong ...

Share this post


Link to post
Share on other sites

I am talking about the fact I have to create an account and that open cl is on the random usa sanction list

This software is subject to the U.S. Export Administration Regulations and other U.S. law, and may not be exported or re-exported to certain countries ( Cuba, Iran, North Korea, Sudan, Syria, and the Crimea region of Ukraine) or to persons or entities prohibited from receiving U.S. exports (including Denied Parties, Specially Designated Nationals, and entities on the Bureau of Export Administration Entity List or involved with missile technology or nuclear, chemical or biological weapons).

Share this post


Link to post
Share on other sites

Well, I'm not a lawyer.

My personal opinion is to not give a fuck about any of that.

Share this post


Link to post
Share on other sites

Not your problem, I agree, but community comes from all countries, in the future it might be better to rely on open source stuff for benchmarks.

  • Like 3

Share this post


Link to post
Share on other sites

OpenCL support for CPUs will be deprecated in GPUPI 4 anyway, so this won't be a problem in the long run.

  • Like 1

Share this post


Link to post
Share on other sites

Anyone have any tips for getting 2080 Ti to go to 3D clocks for GPUPI?

Single card and SLI are both staying at low power clocks despite OS settings to high performance.

 

  • Like 1

Share this post


Link to post
Share on other sites

Mike did you tried a custom curve in MSI AB ? Try moving the points in the curve closer or in a straight line to where your max clocks target is. That is all i have done for my runs on gpupi using 2080Ti. Also max performance selected in the nvidia driver and in windows. 

Edited by chispy

Share this post


Link to post
Share on other sites

To anyone wondering why GPUPI is currently crashing with the W-3175X: I am working with Intel to fix the problem. No CPU sample of course, but they get me in touch with the benchmark guys. I sincerely hope that they deliver so GPUPI can help to prove the true value of this overpriced chip. 😛

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×