I have Cuda crunching at the moment but that’s all that’s crunching. I’m running the standard latest client with following spec PC.
Q9300 oc’d to 3.05
8GB Ram
8800GT 256mb
Any help or tricks? I know I should trawl the internet for solutions but I’m too impatient.
Oh, it says Running 0.10 CPUs, 1 Cuda.
How many cores does a G92 8800GT have?
So BM has downloaded CUDA tasks and no CPU tasks? Are your preferences set correctly at the SETI web site? Astropulse v5 (really 5.03) is turned off by default. Astropulse (really 5.00) WUs are few and far between these days because they are only available as reissues.
Everyone should check out the charts available from http://www.techarp.com/showarticle.aspx?artno=88&pgno=0 for specs of graphics cards. It’s not limited to NVIDIA so Folders will find it useful as well.
[QUOTE=Andy Williams;435087]So BM has downloaded CUDA tasks and no CPU tasks? Are your preferences set correctly at the SETI web site? Astropulse v5 (really 5.03) is turned off by default. Astropulse (really 5.00) WUs are few and far between these days because they are only available as reissues.
Everyone should check out the charts available from http://www.techarp.com/showarticle.aspx?artno=88&pgno=0 for specs of graphics cards. It’s not limited to NVIDIA so Folders will find it useful as well.[/QUOTE]
I only have Cuda tasks even though my settings are correct
That is bizarre because usually the exact opposite is the problem. Are any of the CUDA tasks running at High Priority?
Only 1 Cuda task is running and it doesn’t say high priority.
I’m out of ideas. High Priority will prevent the download of any new WUs.
Also, it doesn’t indicate which GPU I have which I thought it was supposed to.
That is a “feature” of the new BM introduced with 6.4.6 or 6.4.7, don’t remember which.
Running 6.4.7
My return log.
<core_client_version>6.4.7</core_client_version>
<![CDATA[
<stderr_txt>
setiathome_CUDA: Found 1 CUDA device(s):
Device 1 : GeForce 8800 GT
totalGlobalMem = 268435456
sharedMemPerBlock = 16384
regsPerBlock = 8192
warpSize = 32
memPitch = 262144
maxThreadsPerBlock = 512
clockRate = 1600000
totalConstMem = 65536
major = 1
minor = 1
textureAlignment = 256
deviceOverlap = 0
multiProcessorCount = 14
setiathome_CUDA: CUDA Device 1 specified, checking…
Device 1: GeForce 8800 GT is okay
SETI@home using CUDA accelerated device GeForce 8800 GT
setiathome_enhanced 6.03 Visual Studio/Microsoft C++
libboinc: 6.3.22
v_BaseLineSmooth (no other)
v_GetPowerSpectrum 0.00018 0.00000
v_ChirpData 0.00985 0.00000
v_Transpose4 0.00720 0.00000
FPU opt folding 0.00368 0.00000
Cuda error ‘cudaMalloc((void**) &dev_GaussFitResults’ in file ‘c:/sw/gpgpu/seti/seti_boinc/client/cuda/cudaAcceleration.cu’ in line 314 : out of memory.
setiathome_CUDA: CUDA runtime ERROR in device memory allocation (Step 1 of 3). Falling back to HOST CPU processing…
setiathome_CUDA: Found 1 CUDA device(s):
Device 1 : GeForce 8800 GT
totalGlobalMem = 268435456
sharedMemPerBlock = 16384
regsPerBlock = 8192
warpSize = 32
memPitch = 262144
maxThreadsPerBlock = 512
clockRate = 1600000
totalConstMem = 65536
major = 1
minor = 1
textureAlignment = 256
deviceOverlap = 0
multiProcessorCount = 14
setiathome_CUDA: CUDA Device 1 specified, checking…
Device 1: GeForce 8800 GT is okay
SETI@home using CUDA accelerated device GeForce 8800 GT
Restarted at 37.39 percent.
SETI@Home Informational message -9 result_overflow
NOTE: The number of results detected exceeds the storage space allocated.
Flopcounter: 9856123271651.503900
Spike count: 29
Pulse count: 1
Triplet count: 0
Gaussian count: 0
called boinc_finish
</stderr_txt>
]]>
[QUOTE=JUGGY;435114]My return log.
…
SETI@Home Informational message -9 result_overflow
NOTE: The number of results detected exceeds the storage space allocated.
Spike count: 29
Pulse count: 1
Triplet count: 0
Gaussian count: 0[/QUOTE]
The -9 error is quite common. It can result from simply a noisy WU, but many people have seen the error recur time after time until the host is rebooted. Note the high spike count.
http://setiathome.berkeley.edu/cuda_faq.php
I think your problem is that standard BOINC will not run GPU & CPU SETI at the same time, you would need to run another project or one of the custom clients to make your CPU work at the same time. My new 8600GT is doing exactly the same thing but is crunching LHC on the CPU so I don’t mind. If I have the wrong end of the stick my apologies.
You can also use the optimised app from lunatics to crunch both CUDA and AP. If no AP units are available, it will then crunch the MB units via GPU and the CPU.
[QUOTE=Mojo;435244]http://setiathome.berkeley.edu/cuda_faq.php
I think your problem is that standard BOINC will not run GPU & CPU SETI at the same time, you would need to run another project or one of the custom clients to make your CPU work at the same time. My new 8600GT is doing exactly the same thing but is crunching LHC on the CPU so I don’t mind. If I have the wrong end of the stick my apologies.[/QUOTE]
FAQ or not that’s nonsense. Just noticed I have a machine here running 6.08 CUDA and 6.03 CPU at the same time. BOINC client 6.4.7. Maybe they changed the logic?
Copied this from the link you had Mojo:
Q) Does SETI@home run GPU and CPU versions simultaneously?
No. If BOINC determines your CPU is capable of running the CUDA version, only the CUDA version of SETI@home will run. One copy will run on each GPU you have installed. If you want to keep your CPUs occupied at the same time, you can join another BOINC project.
The BOINC developers expect the capability to run CPU and GPU jobs simultaneously for the same application will be added in the next couple months.
Currently running 2 AP units and 1 CUDA. Sort of just happened by itself:shrug:
Thanks for the help though gents.
Juggy, get the optimised client from here:
You may have to sign up to be able to download it, but it will get that quad cruching on all cpu’s plus the graphics card. Also, the AP app is optimised so the V5.03 units will only take slightly longer than the AP 5.00 optimised units did.