I finally had been able to run some WU.
I had this results

https://www.gpugrid.net/results.php?hostid=178531

are in line with the expected performance of such system, or I have to change some settings?

BTW GPU-Z does not show me any load of the GPU, is this normal?

Any idea when new WUs will be available?

Hi, everyone:

About a year ago, I took out of service for repair an AuthenticAMD
AMD A10-5800K APU with Radeon(tm) HD Graphics 7660D [Family 21 Model 16 Stepping 1] (4 processors).

The board was repaired many months ago and is now being used to build a new research PC.

Will I be able to process the new AMD GPU work with this PC?

Thanks,

John

Any idea when new WUs will be available?

No wu for Nvidia nor AMD...
A curiosity: ss this app optimized for GCN??

Any idea when new WUs will be available?

Any news??

Or creating new WU will continue to GPU AMD?

Or GPU GRID has deleted GPU support for GPU AMD?

Any updates?

They've been experimenting with several new applications in recent months and every single one was dropped without a notice. Why would it be different this time?

They've been experimenting with several new applications in recent months and every single one was dropped without a notice. Why would it be different this time?

Very sad but very true.

7 days later and still no answer.

Even "we don't know" would be better than silence...

Please be patient. The end of a semester is always the most busy period for the staff of a university. The AMD app is only a beta, so at worst case it could be even discontinued (I hope it won't happen). The "production" side of this project is based on CUDA and it won't change in the foreseeable future. Thus the CUDA client is the most of the staff's concern. It's quite stable, so if you eager to help this project as much as you demand an AMD app, it's much more productive to adapt to this situation, than waiting for a working AMD app.

Semester has ended and I'm still patient ;)

Semester has ended and I'm still patient ;)

:-(

We expect to come back to AMD gpu soon but not too soon.

gdf

Well... First of all good to know that you are still alive ;>

lots of software changes lately which were quite absorbing.

gdf

AMD Boltzmann initiative

The solution to that problem is the Heterogeneous-compute Interface for Portability, otherwise known as HIP, which gives CUDA developers the tools they need to easily move over to AMD GPUs.

Through HIP AMD will bridge the gap between HCC and CUDA by giving developers a CUDA-like syntax – the various HIP API commands – allowing developers to program for AMD GPUs in a CUDA-like fashion. Meanwhile HIP will also including a toolset (the HIPify Tools) that further simplifies porting by automatically converting CUDA code to HIP code. And finally, once code is HIP – be it natively written that way or converted – it can then be compiled to either NVIDIA or AMD GPUs through NVCC (using a HIP header file to add HIP support) or HCC respectively

Still waiting on some AMD GPU Work Units. I might grow roots soon if you water me!

I too am waiting for ATI wu's.

Have the following setting:
ACEMD short runs (2-3 hours on fastest card): yes
ACEMD long runs (8-12 hours on fastest GPU): no
Molecular Docking (Android): no
Molecular Dynamics for CPU and AMD GPUs (Windows): yes
Quantum Chemistry (Linux, CPU): no
Test wu's: yes

Hi there

I have been using a Nvidia card sucessfully three years or so, exchanged it now as other person need it better, and put in an "old" AMD HD6850.

So I am too waiting for AMD work units.

However it seem i have other problems to solve locally first:
BOINC say it can not detect a suitable GPU.
Strange, is HD6850 not compatible?
It do not have double precision arithmetics, but should be usable for some projects)

I use the proprietary driver and have associated fglrx-opencl package installed.
OS: Linux Mageia 5, 64 bit.

Any tip?