I am running these on gentoo with cgminer 2.11.3 (I've tried 2.10.4 as well), and I noticed that the MH/s for each card initially goes up to 505/526 for the first minute or two. Over the next 3-5 minutes it settles into 480/490 MH/s. In the first period the gpu load is 98-100%, but as it declines and once it hits steady state the gpu load alternates between 64% and 98-100%. I have tried altering the intensity (8, 9, 10), the worksize (256, 512, 1024), the # of gpu threads (1, 2, 3), and the queue size (1,3,5), and the kernel (diakgcn, diablo, poclbm). All of these settings give the same behavior (though the starting and steady state hashing speeds differ slightly). I do not see this behavior on my boxes with 2x5830s, 2x5770s, or 1x7950 & 1x5770. Any thoughts as to what might be causing this or some other test I can try?

I am using gentoo with phoenix-1.7.3, pyopencl-2011.2, numpy-1.6.1-r2, twisted-12.0, and ati-drivers-12.3 or 12.4. When I setup the box yesterday I used 12.3 and after running eselect opencl set amd the miners started (though 20 MH/s slower than my other install, which used sdk-2.5 and fglrx-11.11). I updated a few packages (glib, mpg123, firefox, openssh, and ati-drivers-12.4), and after rebooting I get the below error. I tried re-emerging numpy and pyopencl, but the same error occurred, I reverted back to ati-drivers-12.3, and re-emerging numpy and pyopencl with the same results. A few google searches haven't shown much regarding that error. Do you have any ideas what is broken? Thanks in advance.

Code:

  ImportError: numpy.core.multiarray failed to import
terminate called after throwing an instance of 'std::runtime_error'
  what():  numpy failed to initialize

After multiple google searches and searches through this forum I haven't found a solution, so I apologize if I missed a previously posted answer.

I have 2 boxes with gentoo, kde-4, pyopencl-0.92, ati-stream-sdk-bin-2.5, ati-drivers-11.11, and phoenix-1.6.2. After updating @world the 2nd gpu won't mine on either box (they have both been mining properly since april). If I use pyopencl-0.92 then I get,

Code:

Invalid ELF header
FATAL kernel error: Failed to apply BFI_INT patch to kerne l! Is BFI_INT supported on this hardware?

If I use pyopencl-2011.2,

Code:

FATAL kernel error: Failed to load OpenCL kernel!

I have also ati-stream-sdk-2.4, the opencl included with ati-drivers, ati-drivers-11.12, and phoenix-1.73, and the above problem persists for all configurations. I call phoenix using,

Code:

python ./phoenix.py -u http://wazoo42:Enw8AqY9@bitcoinpool.com:8334/ -q 1 -k phatk2 DEVICE=1 VECTORS AGGRESSION=9 -v BFI_INT WORKSIZE=128

I have noticed that eselect only links libOpenCL.so.1 from /opt/amd-app-sdk/lib/x86_64 to /usr/lib64/libOpenCL.so.1, and I make a similar link to libOpenCL.so. Any thoughts as to what is broken and needs to be fixed? Thanks in advance.