GPUPI - SuperPI on the GPU

[Mr.Scott]

Yeah but who decides which of the 2 cards is the fastest in this benchmark?

Nice.

Yet another flaw.

Points for this bench were so very premature.

[rtsurfer]

I believe this flaw can be easily fixed by just disabling mixing GPUs altogether.

 

Obviously only the creator of the Benchmark can tell us how easy it is to do that.

[_mat_]

I guess mixing cards is currently on a trial run by the HWBOT staff. It's intriguing but there may be social concerns. Performance and cheating wise it should not be a problem. My suggestion for a rule is, that the fastest card decides the submission category. Therefor it doesn't matter what the other cards are, the result will always be slower or at least equal than with the same cards. The fastest card can easily be determined by the statistics below the result. The automatic submission exactly does that for you.

 

What I don't understand right now is the negative energy focused on the bench by some guys here. If you want it to fail, it's going to happen because it's not about having the bench to be unbreakable by cheaters (and haters), it's about embracing it and working things out. GPUPI adds a lot of new possibilities to the table, we can use/enjoy them, abandon others. But it only works if we do it together.

[Mr.Scott]

You confuse negative energy with being gun shy.

Many people lost a lot of points on the UCB fiasco, and your bench has much more to play with than that did.

And we won't even talk about PCM05.

I actually like the bench. I just wish it was a little easier to get going and was a little more bullet proof.

Edited April 18, 2015 by Mr.Scott

[skulstation]

As know the 3870x2 wil not work,but thad din't make me to stop trying to use more than 2 dual gpu cards

after a walk to the kiala point to pickup teemto his two old 3870x2's i start to mes whit them

thx teemto for the good deal thad wa have made.

 

five of the six gpu's ar detected , stil tring to detect all six gpu's, the three 3870x2's ar ok

i have an other 3870x2 but i think this one is not ok,

Edited April 20, 2015 by skulstation

[skulstation]

yes we did it, 3 working 3870x2

now the search after suported dual gpu's can start.

[_mat_]

As I am still waiting for HWBOT staff to test the new timing methods, I took the time to implement the promised legacy version of the benchmark. I currently have it running on Windows XP including OpenCL and CUDA support for GeForce 200 series cards.

 

Now I started experimenting with getting rid of the double precision restriction. I already coded some quadruple floating point algorithms, but it's not looking good. 1M is working, but everything higher needs much more precision as I expected. I am still trying to wrap my head around some float128 bit shifting code, that could possibly work. Too early to tell yet.

 

What would be possible is to outsource the double precision math to the CPU. It's not really a big part of the calculation, but it would alter everything that we got today. And it would be much slower for systems with double precision support.

Edited April 20, 2015 by _mat_

[Leeghoofd]

Is it allowed to alter batch size and reduction, seems many are running the benchmark at non default setting lol

[rtsurfer]

Is it allowed to alter batch size and reduction, seems many are running the benchmark at non default setting lol

 

Yes it is.

And it is a standard "tweak".

[_mat_]

Is it allowed to alter batch size and reduction, seems many are running the benchmark at non default setting lol

Both parameters adjust the workload to the GPU architecture.

[Doc.Brown]

_mat_ downloads links at overclockers.at are broken,could you fix it please I was able to download normal version by changing the adress but it doesn't work for Legacy Version.

[_mat_]

How did you guys get the download links for GPUPI 2.1? Guessed it?

It's officially not released right now, because we are currently discussing the problems with the new 64 bit version and ongoing competitions.

[ObscureParadox]

How did you guys get the download links for GPUPI 2.1? Guessed it?

It's officially not released right now, because we are currently discussing the problems with the new 64 bit version and ongoing competitions.

 

Because it's pretty damn easy to find

[newlife]

How did you guys get the download links for GPUPI 2.1? Guessed it?

It's officially not released right now, because we are currently discussing the problems with the new 64 bit version and ongoing competitions.

 

This is how

[_mat_]

Smart.

 

Well, that all went pretty fast now because of you guys suddenly submitting scores with the new version. That certainly was a big surprise ... in a good way of course, because you just overclocked the launch date!

 

Btw, I've updated HWBOT's internal benchmark version to 2.1 too, so from now on only submissions with 2.1 will be accepted. Automatic submissions with GPUPI 2.0 will fail, but you can still submit the scores manually.

 

As for the new data file saving and uploading: It will start working after the 3rd of May, when manual submission of scores is disabled (as it's said in the annoucement). HWBOT implements it only that way, but it won't matter in a few days anyway.

[pipes]

this is my problem: Error:

High Precision Event Timer for time measurement not found!

 

This is mandatory for Windows 8 and higher to avoid skewed timing.

You'll find a guide on how to enable the HPET timer on your system

in our FAQ. Open about dialog in the menu above for the link.

[GENiEBEN]

The error is pretty descriptive...

 

Cmd.exe as Admin:

 

Bcdedit /set {current} useplatformclock yes

 

Restart pc, voila.

[pipes]

thank you

[skulstation]

what can be wrong?

2.1 not working whit my skulltrail setup

os is win 7 tiny 32bit.

whil instal a normal 64bit os and report agen if its works or not.

[_mat_]

What's the version of the Intel OpenCL driver?

 

Btw, you should try the AMD driver included in the latest catalyst. It should be faster.

[skulstation]

its the •OpenCL™ Runtime 15.1 for Intel® Core™ and Intel® Xeon® Processors for Windows* OS (32-bit) from the intel site.

[skulstation]

whit win 7 64bit the intel opencl is not working agen.but the amd is working

 

i only need manuel enter the cpu

Edited April 28, 2015 by skulstation

[_mat_]

whit win 7 64bit the intel opencl is not working agen.but the amd is working

 

i only need manuel enter the cpu

Seems to be a driver bug, maybe the CPU is not supported by the new Intel drivers. You can try older versions like 14.x, but I recommend to benchmark with AMD's OpenCL 2.0 drivers anyway to get the best score.

 

Thanks for noting the CPU detection problem. HWBOT lists this CPU as QX9775, but the hardware device name is X9775 inside the drivers. This incompatibilities happen all the time. I will include a manual override for those CPUs with GPUPI 2.2.

[GENiEBEN]

@mat

 

//Trim and rename

result.replace("Eight-Core","",Qt::CaseSensitive);

result.replace("Quad-Core","",Qt::CaseSensitive);

result.replace("Six-Core","",Qt::CaseSensitive);

result.replace("Dual-Core","",Qt::CaseSensitive);

result.replace("AMD E","Fusion E",Qt::CaseSensitive);

result.replace("AMD C","Fusion C",Qt::CaseSensitive);

result.replace("Processor","",Qt::CaseInsensitive);

result.replace("(ES)","",Qt::CaseInsensitive);

result.replace("®","",Qt::CaseInsensitive);

result.replace("","",Qt::CaseInsensitive);

result.replace("Core2","Core 2",Qt::CaseSensitive);

result.replace("CPU","",Qt::CaseInsensitive);

result.replace("APU","",Qt::CaseInsensitive);

result.replace("with Radeon HD Graphics","",Qt::CaseInsensitive);

result.replace(" "," ",Qt::CaseInsensitive);

result.replace("Genuine","",Qt::CaseInsensitive);

result.replace("i7 X 990","i7 990X",Qt::CaseSensitive);

result.replace("i7 X 980","i7 980X",Qt::CaseSensitive);

result.replace("i7 X 995","i7 995X",Qt::CaseSensitive);

result.replace("X9750","QX9750",Qt::CaseSensitive);

result.replace("X9770","QX9770",Qt::CaseSensitive);

result.replace("X9775","QX9775",Qt::CaseSensitive);

[_mat_]

Looks very close to my implementation.

 

Thanks mate!