how do i run the stock GPU app and Lunatic CPU apps

Forum > GPU crunching

<< < (2/2)

Jason G:

--- Quote from: perryjay on 28 Jan 2011, 11:05:48 am ---Here it is...
--- End quote ---

Thanks! It spins me out a little when I see the ~3+ times 'Min' figures. It means that the slowest kernels liked the attention I gave. 1.42x average processing rate should translate to a decent percentage improvement on whole tasks (probably around 20% or so without further refinement). The next few weeks levering in those to the pipeline is going to be interesting.

Jason

Josef W. Segur:

--- Quote from: RottenMutt on 27 Jan 2011, 10:06:58 pm ---i tried installing using lunatic unified installer and not checking the gpu app and ended dumping my cache :P
...
--- End quote ---

Richard Haselgrove wrote a very good FAQ, Run Seti Enhanced on Fermi class GPUs (4xx) (Advanced users). Although it's specifically for the 6.10 cuda_fermi stock application, the pattern is the same for any stock application.

Because the project has resend_lost_results on, when you get app_info.xml fixed up the "dumped" tasks may be sent again.
Joe

Claggy:
Since i've also got a new GPU, a GTX 460 GLH, here's my Test #10 results (Win 7 x64, 266.58):

--- Code: ---Device: GeForce GTX 460, 1600 MHz clock, 993 MB memory.
Compute capability 2.1
Compiled with CUDA 3020.
PowerSpectrum+summax Unit test #10 (FFT pipeline throughput)
Stock:
Processing... Done!
Compute Thoughput GFlops Avg( 69.92) Peak( 114.81) Min( 8.91) [OK]
Memory thoughput GB/s Avg( 37.67) Peak( 54.34) Min( 15.05)

Opt1 (worst case): 256 thrds/block, 2 x 524288 element streams
revert to single stream from size 512
Processing... Done!
Compute thoughput [GFlops] -
Avg( 102.27, 1.46x) Peak( 151.62, 1.32x) Min( 21.28, 2.39x) [OK]
Memory thoughput [GB/s] -
Avg( 60.11, 1.60x) Peak( 77.00, 1.42x) Min( 37.53, 2.49x)

--- End code ---

Claggy

RottenMutt:
all 266.58 drivers and x64 vista or 7
8800gts: interesting, loaded the video engine alittle...

--- Code: ---Device: GeForce 8800 GTS 512, 1674 MHz clock, 492 MB memory.
Compute capability 1.1
Compiled with CUDA 3020.
PowerSpectrum+summax Unit test #10 (FFT pipeline throughput)
Stock:
Processing... Done!
Compute Thoughput GFlops Avg( 41.89) Peak( 58.05) Min( 11.00) [OK]
Memory thoughput GB/s Avg( 25.23) Peak( 36.89) Min( 18.43)

Opt1 (worst case): 64 thrds/block, 2 x 524288 element streams
revert to single stream from size 128
Processing... Done!
Compute thoughput [GFlops] -
Avg( 48.37, 1.15x) Peak( 68.83, 1.19x) Min( 17.61, 1.60x) [OK]
Memory thoughput [GB/s] -
Avg( 30.32, 1.20x) Peak( 40.81, 1.11x) Min( 20.93, 1.14x)

--- End code ---

GTX475 (460 flashed to make it think it is a 475; i think i can, i think i can, I can)

--- Code: ---Device: GeForce GTX 470, 1250 MHz clock, 1248 MB memory.
Compute capability 2.0
Compiled with CUDA 3020.
PowerSpectrum+summax Unit test #10 (FFT pipeline throughput)
Stock:
Processing... Done!
Compute Thoughput GFlops Avg( 75.32) Peak( 119.15) Min( 8.83) [OK]
Memory thoughput GB/s Avg( 41.63) Peak( 60.16) Min( 15.57)

Opt1 (worst case): 256 thrds/block, 2 x 524288 element streams
revert to single stream from size 512
Processing... Done!
Compute thoughput [GFlops] -
Avg( 115.05, 1.53x) Peak( 164.29, 1.38x) Min( 38.70, 4.38x) [OK]
Memory thoughput [GB/s] -
Avg( 69.71, 1.67x) Peak( 88.85, 1.48x) Min( 49.02, 3.15x)
--- End code ---

GTX295:

--- Code: ---Device: GeForce GTX 295, 1369 MHz clock, 896 MB memory.
Compute capability 1.3
Compiled with CUDA 3020.
PowerSpectrum+summax Unit test #10 (FFT pipeline throughput)
Stock:
Processing... Done!
Compute Thoughput GFlops Avg( 81.66) Peak( 113.97) Min( 16.45) [OK]
Memory thoughput GB/s Avg( 47.72) Peak( 67.99) Min( 29.01)

Opt1 (worst case): 128 thrds/block, 2 x 524288 element streams
revert to single stream from size 256
Processing... Done!
Compute thoughput [GFlops] -
Avg( 78.32, 0.96x) Peak( 114.21, 1.00x) Min( 23.06, 1.40x) [OK]
Memory thoughput [GB/s] -
Avg( 47.65, 1.00x) Peak( 65.35, 0.96x) Min( 21.84, 0.75x)
--- End code ---

GTX480:

--- Code: ---Device: GeForce GTX 480, 1440 MHz clock, 1504 MB memory.
Compute capability 2.0
Compiled with CUDA 3020.
PowerSpectrum+summax Unit test #10 (FFT pipeline throughput)
Stock:
Processing... Done!
Compute Thoughput GFlops Avg( 80.18) Peak( 121.88) Min( 9.81) [OK]
Memory thoughput GB/s Avg( 44.31) Peak( 64.58) Min( 17.29)

Opt1 (worst case): 256 thrds/block, 2 x 524288 element streams
revert to single stream from size 512
Processing... Done!
Compute thoughput [GFlops] -
Avg( 145.60, 1.82x) Peak( 208.34, 1.71x) Min( 34.86, 3.55x) [OK]
Memory thoughput [GB/s] -
Avg( 86.97, 1.96x) Peak( 113.33, 1.75x) Min( 61.46, 3.55x)
--- End code ---

another gtx480

--- Code: ---Device: GeForce GTX 480, 1440 MHz clock, 1503 MB memory.
Compute capability 2.0
Compiled with CUDA 3020.
PowerSpectrum+summax Unit test #10 (FFT pipeline throughput)
Stock:
Processing... Done!
Compute Thoughput GFlops Avg( 84.40) Peak( 130.24) Min( 9.68) [OK]
Memory thoughput GB/s Avg( 46.41) Peak( 68.03) Min( 17.06)

Opt1 (worst case): 256 thrds/block, 2 x 524288 element streams
revert to single stream from size 512
Processing... Done!
Compute thoughput [GFlops] -
Avg( 144.95, 1.72x) Peak( 208.23, 1.60x) Min( 34.33, 3.55x) [OK]
Memory thoughput [GB/s] -
Avg( 86.54, 1.86x) Peak( 113.26, 1.66x) Min( 60.53, 3.55x)
--- End code ---

perryjay:
Okay, you guys are just showing off now!! ;D

Navigation

[0] Message Index

[*] Previous page

Go to full version