Current GPU version's requirements: any AMD (HD 5xxx or newer) or NVIDIA (4xx or newer) GPU with OpenCL support and at least 1 GB of video memory

Yes, it's coming soon (tm).

A small teaser: during the test, after the test has finished

GTX 770 completed this work unit in 24 minutes 18 seconds. Two times faster than "Intel(R) Xeon(R) CPU E5-1650 v3 @ 3.50GHz [Family 6 Model 63 Stepping 2]".

P.S. GTX 1070: 12 minutes 44 seconds.

Random thought on the GPU version; you probably do NOT want to have a "# of GPU tasks" choice like Einstein has, without doing a LOT of testing first. Much better to keep it to one task at a time and let people change it with config files if they want more. That's the opposite of the CPU case where you want them to be able to LIMIT your use easily. You probably need the whole GPU anyway for your app, since you need memory space.

I have not looked at your code, but you say there's little/no floating point, so GPU should work great whether AMD or nVidia. Will you use OpenGL only or have a Cuda version? People wanting a lot of double-precision FP go for AMD, so lots of those cards on BOINC, even though (for most cases) the nVidia cards work better on most projects. Many people have problems with drivers and OpenGL, especially now that Win10 keeps "upgrading" perfectly good drivers with stuff that doesn't work. Up to you whether it's worth having two versions or not. Regardless, be sure to test well with plenty of different cards, on multiple OS versions... or use the "run beta apps" choice in prefs and let us test it for you.

Speaking of which, where is the Mac version??? :-)

We want it ALL! And NOW! We can wait till next week for the Sun/Sparc and Mac PPC releases (maybe) I guess!

The code for CPU is here: https://github.com/SChernykh/Amicable/tree/boinc-version

GPU version is written in OpenCL, so it will work on all OpenCL-capable devices. I've tested it on NVIDIA GTX 770, GTX 1070 and Intel HD Graphics 4600 (integrated graphics in Core i7-4770K) so far. And I'll also get an AMD card for testing next week.

P.S. I don't have a Mac to build Mac version.
P.P.S. You can try to run Windows version on Mac through Wine.

Hi.
Macintosh CPU and GPU versions (ideally running on OS X 10.6.8 as well) would be much appreciated. Wine has issues on 10.6 so it's out of the question for me.
Don't know about the graphics card requirements, but I'd be glad to run any application on my oldish Nvidia cards with 256 megs, albeit quite slowly.
I'd be ready to beta-test anything you'd throw at my old machines.

---

- - - - - - - - - - - - - -
Greetings, Jens

It looks like GTX 1070 = Intel CPU with 36 physical cores (72 threads) @ 2 GHz: https://sech.me/boinc/Amicable/workunit.php?wuid=58083

If you need other test systems, I have a variety of Nvidia GPUs to run it on; 1080, 1070, 1060, 970, 960, 750Ti

---

I'll add beta version soon, maybe even today.

Just looked to make sure I had "run beta apps" checked, but it's not an option yet...

Yes, it's only visible to me right now. When I'm sure it runs on all OS + GPU combinations, I'll enable it for everyone.

Just looked to make sure I had "run beta apps" checked, but it's not an option yet...

It is now. You can already start testing GPU version.

All good here. Runtime 650 seconds.

Win 7 x64, GTX980Ti.

GPU usage looks good ~97-100%

Sorry for my bad english.

Not all good here. I don't know why but my calculation times for my tasks are all different.

On my ATI R9 2960X : runtime between 109 seconds and 1266 seconds for a task.
On my GTX 680 : runtime between 383 seconds and 2116 seconds for a task.

What should I do to always have the shortest time for the tasks ?

Win 7 x64 and ATI and NVidia in the same computer.

On my ATI R9 2960X : runtime between 109 seconds and 1266 seconds for a task.
On my GTX 680 : runtime between 383 seconds and 2116 seconds for a task.

This is normal. All WUs were normalized for CPU version, not for GPU. Some WUs get a huge speed-up when running on GPU, some don't. It depends on how well a work unit can be parallelized.

It got very warm. NVIDIA 1060 that runs SETI and Einstein at temp 70C with fan about 2000rpm
With amicable 1.07 nvidia app it got up to 74 with fan at 2600rpm and rising so i aborted.

//TRuEQ

It got very warm. NVIDIA 1060 that runs SETI and Einstein at temp 70C with fan about 2000rpm
With amicable 1.07 nvidia app it got up to 74 with fan at 2600rpm and rising so i aborted.

//TRuEQ

Well, this is because this app is optimized well and does a lot of computations in a short time. GTX 1070 runs a work unit in 4-15 minutes, compared to 40 minutes on Intel Xeon E5-1650 v3 (6 cores @ 3.5 GHz).
Temps up to 80 degrees are OK for NVIDIA GPUs. You can set fan curve to 100% at 80 degrees and you should be safe. Mine GTX 1070 runs amicable 1.07 at 75-77 degrees with fan running at 2800-2900 RPM, depending on the work unit.

Thanks for answer.

But app is a little too hot for me.

I'll pass here.

//TRuEQ

On my ATI R9 2960X : runtime between 109 seconds and 1266 seconds for a task.
On my GTX 680 : runtime between 383 seconds and 2116 seconds for a task.

This is normal. All WUs were normalized for CPU version, not for GPU. Some WUs get a huge speed-up when running on GPU, some don't. It depends on how well a work unit can be parallelized.

Thank you for your answer !
And sorry, my GPU is R9 290X and not R9 2960X.

Runs okay on my 2 Nvidia cards, the 1080 Laptop runs them in about 5-11 Min's. I stopped them running on my AMD though, the gpu fan was just screaming & it was way to hot & eventually blue screened the box so I'll stick with cpu work on that box ...

Around 4000 and 5000 seconds on a GTX750ti, around 99% GPU utilization. Very cool.
Linux 64 bit on both boxes (SuSE and Ubuntu) - waiting for verification by the wingmen...

Cheers
Holger

---

My stats
My badges

Runs okay on my 2 Nvidia cards, the 1080 Laptop runs them in about 5-11 Min's. I stopped them running on my AMD though, the gpu fan was just screaming & it was way to hot & eventually blue screened the box so I'll stick with cpu work on that box ...

I'm getting the opposite on my 7970, it's running very cool but it seems under-worked. I'm seeing 99% utilization but only 67c on the core. Voltage and clocks are constant. On the other hand I feel like it's running slower than it should compared to other cards; I'm seeing times around:

7970: ~2000sec avg
270x: ~3000sec avg
750ti: ~2400sec avg

Seems like my ATI cards are under performing.