I got:
05-Dec-2012 12:33:10 [World Community Grid] The Help Conquer Cancer NVIDIA GPU application cannot run on your graphics card
The Ion chipset is like the Geforce 9400M graphic card and this one is listed in the https://secure.worldcommunitygrid.org/help/viewTopic.do?shortName=GPU#610 thread.
But I think that for testing and development also slower cards may be good, especially when doing tests for new platform.
Or is there another reason for not getting work?

11-Oct-2012 17:10:39 [---] NVIDIA GPU 0: ION (driver version unknown, CUDA version 4020, compute capability 1.1, 509MB, 35 GFLOPS peak)

27-Jun-2012 17:04:30 [---] NVIDIA GPU 0: GeForce GT 320 (driver version unknown, CUDA version 4020, compute capability 1.2, 1023MB, 187 GFLOPS peak)

16-Nov-2012 08:54:46 [---] NVIDIA GPU 0: GeForce GT 430 (driver version unknown, CUDA version 5000, compute capability 2.1, 2047MB, 90 GFLOPS peak)

Maybe it's the compute capability? My GT320 and the GT430 are getting beta workunits.
Stephan
PS: On my GT430 I got:
Estimated kernel execution time = 2.48244 [sec]
ERROR: Kernel execution time estimate too high, exiting.
12:55:34 (17000): called boinc_finish

PPS: Now I'm getting:
<rsc_fpops_est>25411773615583.000000</rsc_fpops_est>
<rsc_fpops_bound>508235472311660.000000</rsc_fpops_bound>
<rsc_memory_bound>78643200.000000</rsc_memory_bound>
<rsc_disk_bound>50000000.000000</rsc_disk_bound>
<computation_deadline>1354843616.000000</computation_deadline>
</app_init_data>
INFO: gpu_type not found in init_data.xml.
INFO: gpu_device_num set in init_data.xml to 0
Boinc requested NVIDIA gpu device number 0
Unzipping input images ../../projects/www.worldcommunitygrid.org/4fcb8b86dc8f6139b8f3c2a5744cf4f6.zip
Processing jobdescription
Number of Images defined in image list is 2
ERROR: VerifyGPU.cpp:65 Unknown
14:27:04 (22145): called boinc_finish

</stderr_txt>
]]>
I will switch back to CUDA 4.2 and see what will happen ...

On the GT320 it looks like this:
<rsc_fpops_est>25411773615583.000000</rsc_fpops_est>
<rsc_fpops_bound>508235472311660.000000</rsc_fpops_bound>
<rsc_memory_bound>78643200.000000</rsc_memory_bound>
<rsc_disk_bound>50000000.000000</rsc_disk_bound>
<computation_deadline>1354842277.000000</computation_deadline>
</app_init_data>
INFO: gpu_type not found in init_data.xml.
INFO: gpu_device_num set in init_data.xml to 0
Boinc requested NVIDIA gpu device number 0
Unzipping input images ../../projects/www.worldcommunitygrid.org/c2472cbd6f30acaad805e9dca3e426df.zip
Processing jobdescription
Number of Images defined in image list is 2
Found compute platform NVIDIA Corporation
Selecting this platform
CL_DEVICE_NAME: GeForce GT 320
CL_DEVICE_VENDOR: NVIDIA Corporation
CL_DEVICE_VERSION: 295.71
CL_DEVICE_MAX_COMPUTE_UNITS:
CL_DEVICE_MAX_WORK_ITEM_DIMENSIONS: 3
CL_DEVICE_MAX_WORK_ITEM_SIZES: 512 / 512 / 64
CL_DEVICE_MAX_WORK_GROUP_SIZE: 512
CL_DEVICE_MAX_CLOCK_FREQUENCY: 1302 MHz
CL_DEVICE_ADDRESS_BITS: 32
CL_DEVICE_MAX_MEM_ALLOC_SIZE: 255 MByte
CL_DEVICE_GLOBAL_MEM_SIZE: 1023 MByte
CL_DEVICE_ERROR_CORRECTION_SUPPORT: no
CL_DEVICE_LOCAL_MEM_TYPE: local
CL_DEVICE_LOCAL_MEM_SIZE: 16 KByte
CL_DEVICE_MAX_CONSTANT_BUFFER_SIZE: 64 KByte
CL_DEVICE_QUEUE_PROPERTIES: CL_QUEUE_OUT_OF_ORDER_EXEC_MODE_ENABLE
CL_DEVICE_QUEUE_PROPERTIES: CL_QUEUE_PROFILING_ENABLE
CL_DEVICE_EXTENSIONS:
cl_khr_byte_addressable_store
cl_khr_icd
cl_khr_gl_sharing
cl_nv_compiler_options
cl_nv_device_attribute_query
cl_nv_pragma_unroll
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_DEVICE_COMPUTE_CAPABILITY_NV: 1.2
CL_DEVICE_REGISTERS_PER_BLOCK_NV: 16384
CL_DEVICE_WARP_SIZE_NV: 32
CL_DEVICE_GPU_OVERLAP_NV: CL_TRUE
CL_DEVICE_KERNEL_EXEC_TIMEOUT_NV: CL_FALSE
CL_DEVICE_INTEGRATED_MEMORY_NV: CL_FALSE
Estimated kernel execution time = 1.39454 [sec]
Starting analysis of X0900075151391200610031534.jp2...
Extracting GLCM features...
Total kernel time: 905.932495 (1026 kernel executions)
Total memory transfer time: 2.928820
Average kernel time: 0.882975
Min kernel time: 0.751814 (dx=25 dy=0 sample_dist=24 )
Max kernel time: 1.073066 dx=2 dy=1 sample_dist=1
INFO: GPU calculations complete.
Total time for X0900075151391200610031534.jp2: 959 seconds
Finished Image #0, pctComplete = 0.500000
Starting analysis of X0900075151392200610031534.jp2...
Extracting GLCM features...
Total kernel time: 952.409607 (1026 kernel executions)
Total memory transfer time: 5.806847
Average kernel time: 0.928274
Min kernel time: 0.808257 (dx=25 dy=0 sample_dist=24 )
Max kernel time: 1.093689 dx=2 dy=1 sample_dist=1
INFO: GPU calculations complete.
Total time for X0900075151392200610031534.jp2: 1005 seconds
Finished Image #1, pctComplete = 1.000000
CPU time used = 1958.970000
14:37:27 (23457): called boinc_finish

</stderr_txt>
]]>

Each GPU task starts with a mini-test to see if there is reasonable performance on the system, which of course is a combination of CPU+GPU+OS+MomentaryLoad. When the mini-test does not give a reasonable kernel duration test, the task is aborted. My GT220 had a KT of 2.6 seconds and got rejected, so I've hooked it to a profile which has GPU work deselected for WCG.

Personally, to get a test optimal [for medium/lower end cards], the system should be idle before a task starts, so setting e.g. a few minutes delay after end of user input, would allow the system to come to quite and then do the best test possible **. What I'm saying is, the system may do the test when idle such that it wont fail.

** Similarly the CPU benchmark is performed when BOINC computing is temporarily suspended [30 seconds].

I received 228 WUs so far on 3 cards (nVidia GTX 460, GTX 550 Ti and AMD Radeon HD 6950). 204 are valid, 11 pending and 12 in progress. 1 unit was invalid (BETA_X0900075151083200610031539--201212042007340). Apart from the one invalid unit everything seems to be working very smoothly.

Ummm ... With the Dashboard showing only 53 days of crunching left in the HCC project, why start a GPU version under Linux now?
Or is there some information about the duration of HCC that hasn't been released to us mortals yet?
In my case, such info would determine whether I'll buy some suitable graphics hardware and join the fray.
For others who are already running the project, such info would enable informed decisions to be made about graphics hardware upgrades.

Hello kateiacy,

Thanks for your [Dec 5, 2012 1:57:51 PM] post. I'll follow the steps there now, see how that goes, and post some feedback.
;
; andzgridPost#716
;
edit1_2012.12.05We.1538utc:
Currently downloading 'fglrx'. Funny, there is indicated 'Applying changes' when it is more like 'Downloading files'... Ok, the download and the apply seems to be done as I now see the radio-button on 'Using Video driver for the AMD graphics accelerators from fglrx (proprietary)' where the radio-button used to be at the 'Using Video driver for the AMD graphics accelerators from fglrx-updates (proprietary)'... Rebooting now...
;
edit2_2012.12.05We.1626utc:
I had a flawless AMD/ATI driver install . I now see GPU-use-mode controls under the Activity menu where there was once none. The event log now has a description of the AMD/ATI GPU card as follows:
-- ATI GPU 0: Juniper (CAL version 1.4.1741, 1024MB, 804MB available, 2720 GFLOPS peak)
-- OpenCL: ATI GPU 0: Juniper (driver version 1016.4, device version OpenCL 1.2 AMD-APP (1016.4), 1024MB, 804MB available).

The Catalyst UI, however, is something I'm not sure of if it's the latest for Ubuntu-Linux. But that'll be for another day. For now, thanks for the guidance, kateiacy!
;