I got a new (old) GPU, AMD R9 290X, which is running fine on Milkyway, PG, and even FGRPB1. But when running O3AS, I get only Validate Error, or marked as invalid.
Here is a Link to all of my invalid WUs.
Maybe somebody has an Idea.
Greets
Felix
Copyright © 2024 Einstein@Home. All rights reserved.
[SG wrote:Felix]... when
)
I suspect this may be because the hardware (GCN 2nd generation -- Hawaii) is rather old.
I know for certain that GCN 1st gen cards (eg. Tahiti) do not work because I tried one quite a while ago. I don't have any 2nd gen cards to test so can't say for sure. However 1st and 2nd gen were reasonably similar so I wouldn't be surprised. I have 4th gen cards (Polaris) and they work fine.
The staff don't have a lot of different cards (particularly older ones like yours) so I don't think they will know for sure either. Perhaps someone else may have a 2nd gen card and can advise if it can work.
I use Linux only and would run clinfo to check the OpenCL version and capabilities of the card. I imagine there might be a similar utility for Windows that you could run?
Cheers,
Gary.
From
)
From coproc_info.xml:
<coprocs>
<ati_opencl>
<name>AMD Radeon(TM) R9 290X</name>
<vendor>Advanced Micro Devices, Inc.</vendor>
<vendor_id>4098</vendor_id>
<available>1</available>
<half_fp_config>0</half_fp_config>
<single_fp_config>190</single_fp_config>
<double_fp_config>63</double_fp_config>
<endian_little>1</endian_little>
<execution_capabilities>1</execution_capabilities>
<extensions>cl_khr_fp64 cl_amd_fp64 cl_khr_global_int32_base_atomics cl_khr_global_int32_extended_atomics cl_khr_local_int32_base_atomics cl_khr_local_int32_extended_atomics cl_khr_int64_base_atomics cl_khr_int64_extended_atomics cl_khr_3d_image_writes cl_khr_byte_addressable_store cl_khr_gl_sharing cl_khr_gl_depth_images cl_amd_device_attribute_query cl_amd_vec3 cl_amd_printf cl_amd_media_ops cl_amd_media_ops2 cl_amd_popcnt cl_khr_d3d10_sharing cl_khr_d3d11_sharing cl_khr_dx9_media_sharing cl_khr_image2d_from_buffer cl_khr_spir cl_khr_subgroups cl_khr_gl_event cl_khr_depth_images cl_khr_mipmap_image cl_khr_mipmap_image_writes cl_amd_liquid_flash cl_amd_planar_yuv</extensions>
<global_mem_size>4294967296</global_mem_size>
<local_mem_size>32768</local_mem_size>
<max_clock_frequency>1000</max_clock_frequency>
<max_compute_units>44</max_compute_units>
<nv_compute_capability_major>0</nv_compute_capability_major>
<nv_compute_capability_minor>0</nv_compute_capability_minor>
<amd_simd_per_compute_unit>4</amd_simd_per_compute_unit>
<amd_simd_width>16</amd_simd_width>
<amd_simd_instruction_width>1</amd_simd_instruction_width>
<opencl_platform_version>OpenCL 2.1 AMD-APP (3240.6)</opencl_platform_version>
<opencl_device_version>OpenCL 2.0 AMD-APP (3240.6)</opencl_device_version>
<opencl_driver_version>3240.6</opencl_driver_version>
<device_num>0</device_num>
<peak_flops>5632000000000.000000</peak_flops>
<opencl_available_ram>4294967296.000000</opencl_available_ram>
<opencl_device_index>0</opencl_device_index>
<warn_bad_cuda>0</warn_bad_cuda>
</ati_opencl>
<warning>No NVIDIA library found</warning>
<warning>No ATI library found.</warning>
</coprocs>
So I would say OpenCL 2.0 is supported