Could HD5XXX owners run a benchmark for me?

InvisiBill

InvisiBill

2[H]4U
Joined
Jan 2, 2003
Messages
2,608
There aren't a whole lot of posted dnetc benchmark results for the HD5000's, and the ones that are posted have some some pretty big variances. I trust you guys who know about hardware a little more than I trust the random results posted there. I'm specifically looking at the 5870, but I'd like to see other lower cards too for comparison.

  1. Go to the Pre-Release Clients page and download the [x86/Stream] client.
  2. Unzip the file somewhere.
  3. Run dnetc.exe -runoffline -l benchmark.txt -cpuinfo
  4. Run dnetc.exe -runoffline -l benchmark.txt -bench
  5. Post your results here. "CODE" tags will keep it a manageable size.
That should run dnetc, keeping it completely offline and logging to benchmark.txt for easy copying of the results. The first command will output info about the GPU, while the second one will benchmark all the different cores.

You should end up with something like this (obviously this example is from the CUDA version):
Code: 
dnetc v2.9107-516-GTR-09122714 for CUDA 2.2 on Win32 (WindowsNT 6.1).

[Mar 31 15:40:29 UTC] nvcuda.dll Version: 8.17.11.9716
Automatic processor identification tag: 8192
	name: Quadro NVS 160M (1 MPs)
Estimated processor clock speed (0 if unknown): 1450 MHz
Number of processors detected by this client: 1
Number of processors supported by this client: 128

dnetc v2.9107-516-GTR-09122714 for CUDA 2.2 on Win32 (WindowsNT 6.1).

[Mar 31 15:40:32 UTC] nvcuda.dll Version: 8.17.11.9716
[Mar 31 15:40:32 UTC] RC5-72: using core #0 (CUDA 1-pipe 64-thd).
[Mar 31 15:40:51 UTC] RC5-72: Benchmark for core #0 (CUDA 1-pipe 64-thd)
                      0.00:00:16.52 [6,410,666 keys/sec]
[Mar 31 15:40:51 UTC] RC5-72: using core #1 (CUDA 1-pipe 128-thd).
[Mar 31 15:41:10 UTC] RC5-72: Benchmark for core #1 (CUDA 1-pipe 128-thd)
                      0.00:00:16.38 [8,615,123 keys/sec]
[Mar 31 15:41:10 UTC] RC5-72: using core #2 (CUDA 1-pipe 256-thd).
[Mar 31 15:41:32 UTC] RC5-72: Benchmark for core #2 (CUDA 1-pipe 256-thd)
                      0.00:00:16.30 [9,823,934 keys/sec]
[Mar 31 15:41:32 UTC] RC5-72: using core #3 (CUDA 2-pipe 64-thd).
[Mar 31 15:41:52 UTC] RC5-72: Benchmark for core #3 (CUDA 2-pipe 64-thd)
                      0.00:00:17.75 [8,775,579 keys/sec]
[Mar 31 15:41:52 UTC] RC5-72: using core #4 (CUDA 2-pipe 128-thd).
[Mar 31 15:42:16 UTC] RC5-72: Benchmark for core #4 (CUDA 2-pipe 128-thd)
                      0.00:00:18.06 [8,881,913 keys/sec]
[Mar 31 15:42:16 UTC] RC5-72: using core #6 (CUDA 4-pipe 64-thd).
[Mar 31 15:42:40 UTC] RC5-72: Benchmark for core #6 (CUDA 4-pipe 64-thd)
                      0.00:00:19.26 [10,039,487 keys/sec]
[Mar 31 15:42:40 UTC] RC5-72: using core #7 (CUDA 4-pipe 128-thd).
[Mar 31 15:42:43 UTC] RC5 CUDA ERROR [0]: 'the launch timed out and was terminated' (cudaEventSynchronize)

[Mar 31 15:42:44 UTC] RC5-72 benchmark summary :
                      Default core : #0 (CUDA 1-pipe 64-thd)
                      Fastest core : #6 (CUDA 4-pipe 64-thd)
[Mar 31 15:42:44 UTC] Core #6 is significantly faster than the default core.
                      The GPU core selection has been made as a tradeoff between core speed
                      and responsiveness of the graphical desktop.
                      Please file a bug report along with the output of -cpuinfo
                      only if the the faster core selection does not degrade graphics performance.
 
Code: 
dnetc v2.9108-517-GTR-10021519 for ATI Stream on Win32 (WindowsNT 6.1).

Automatic processor identification tag: 8
	name: RV870, RV870, RV870, RV870
Estimated processor clock speed (0 if unknown): 735 MHz
Number of processors detected by this client: 4
Number of processors supported by this client: 128

dnetc v2.9108-517-GTR-10021519 for ATI Stream on Win32 (WindowsNT 6.1).

[Mar 31 18:17:46 UTC] RC5-72: using core #0 (IL 4-pipe c).
[Mar 31 18:17:52 UTC] RC5-72: Benchmark for core #0 (IL 4-pipe c)
                      0.00:00:03.02 [1,432,037,508 keys/sec]
[Mar 31 18:17:52 UTC] RC5-72: using core #1 (IL 4-pipe c alt).
[Mar 31 18:17:59 UTC] RC5-72: Benchmark for core #1 (IL 4-pipe c alt)
                      0.00:00:04.55 [973,296,919 keys/sec]
[Mar 31 18:17:59 UTC] RC5-72: using core #2 (IL 4-pipe 2 threads).
[Mar 31 18:18:04 UTC] RC5-72: Benchmark for core #2 (IL 4-pipe 2 threads)
                      0.00:00:03.32 [1,340,001,999 keys/sec]
[Mar 31 18:18:05 UTC] RC5-72 benchmark summary :
                      Default core : #0 (IL 4-pipe c)
                      Fastest core : #0 (IL 4-pipe c)
 
Radeon 5770 at 960Mhz core:
Code: 
dnetc v2.9108-517-GTR-10021520 for ATI Stream on Win32 (WindowsNT 6.1).

Automatic processor identification tag: 9
	name: RV840
Estimated processor clock speed (0 if unknown): 960 MHz
Number of processors detected by this client: 1
Number of processors supported by this client: 128

dnetc v2.9108-517-GTR-10021520 for ATI Stream on Win32 (WindowsNT 6.1).

[Mar 31 18:37:40 UTC] RC5-72: using core #0 (IL 4-pipe c).
[Mar 31 18:37:47 UTC] RC5-72: Benchmark for core #0 (IL 4-pipe c)
                      0.00:00:04.81 [903,003,287 keys/sec]
[Mar 31 18:37:47 UTC] RC5-72: using core #1 (IL 4-pipe c alt).
[Mar 31 18:37:57 UTC] RC5-72: Benchmark for core #1 (IL 4-pipe c alt)
                      0.00:00:07.78 [577,512,966 keys/sec]
[Mar 31 18:37:57 UTC] RC5-72: using core #2 (IL 4-pipe 2 threads).
[Mar 31 18:38:06 UTC] RC5-72: Benchmark for core #2 (IL 4-pipe 2 threads)
                      0.00:00:05.79 [746,503,675 keys/sec]
[Mar 31 18:38:06 UTC] RC5-72 benchmark summary :
                      Default core : #0 (IL 4-pipe c)
                      Fastest core : #0 (IL 4-pipe c)
 
Code: 
dnetc v2.9108-517-GTR-10021520 for ATI Stream on Win32 (WindowsNT 6.1).

Automatic processor identification tag: 8
	name: RV870, RV870
Estimated processor clock speed (0 if unknown): 1000 MHz
Number of processors detected by this client: 2
Number of processors supported by this client: 128

dnetc v2.9108-517-GTR-10021520 for ATI Stream on Win32 (WindowsNT 6.1).

[Mar 31 20:58:28 UTC] RC5-72: using core #0 (IL 4-pipe c).
[Mar 31 20:58:34 UTC] RC5-72: Benchmark for core #0 (IL 4-pipe c)
                      0.00:00:03.60 [1,235,473,330 keys/sec]
[Mar 31 20:58:34 UTC] RC5-72: using core #1 (IL 4-pipe c alt).
[Mar 31 20:58:42 UTC] RC5-72: Benchmark for core #1 (IL 4-pipe c alt)
                      0.00:00:04.63 [955,356,085 keys/sec]
[Mar 31 20:58:42 UTC] RC5-72: using core #2 (IL 4-pipe 2 threads).
[Mar 31 20:58:47 UTC] RC5-72: Benchmark for core #2 (IL 4-pipe 2 threads)
                      0.00:00:03.16 [1,396,615,662 keys/sec]
[Mar 31 20:58:47 UTC] RC5-72 benchmark summary :
                      Default core : #0 (IL 4-pipe c)
                      Fastest core : #2 (IL 4-pipe 2 threads)
[Mar 31 20:58:47 UTC] Core #2 is significantly faster than the default core.
                      The GPU core selection has been made as a tradeoff between core speed
                      and responsiveness of the graphical desktop.
                      Please file a bug report along with the output of -cpuinfo
                      only if the the faster core selection does not degrade graphics performance.

dnetc5970oc.png


I was a little surprised to see the above setup was worth more than .5% of the project output.
 
Last edited:
Code: 
dnetc v2.9108-517-GTR-10021520 for ATI Stream on Win32 (WindowsNT 6.1).

Automatic processor identification tag: 8
	name: RV870, RV870
Estimated processor clock speed (0 if unknown): 850 MHz
Number of processors detected by this client: 2
Number of processors supported by this client: 128

dnetc v2.9108-517-GTR-10021520 for ATI Stream on Win32 (WindowsNT 6.1).

[Apr 01 04:05:13 UTC] RC5-72: using core #0 (IL 4-pipe c).
[Apr 01 04:05:18 UTC] RC5-72: Benchmark for core #0 (IL 4-pipe c)
                      0.00:00:02.32 [1,904,178,351 keys/sec]
[Apr 01 04:05:18 UTC] RC5-72: using core #1 (IL 4-pipe c alt).
[Apr 01 04:05:24 UTC] RC5-72: Benchmark for core #1 (IL 4-pipe c alt)
                      0.00:00:03.82 [1,154,547,023 keys/sec]
[Apr 01 04:05:24 UTC] RC5-72: using core #2 (IL 4-pipe 2 threads).
[Apr 01 04:05:30 UTC] RC5-72: Benchmark for core #2 (IL 4-pipe 2 threads)
                      0.00:00:02.85 [1,532,174,615 keys/sec]
[Apr 01 04:05:30 UTC] RC5-72 benchmark summary :
                      Default core : #0 (IL 4-pipe c)
                      Fastest core : #0 (IL 4-pipe c)
 
Last edited:
Thanks guys. noquarter's 5770 is a great example. It's overclocked to 960MHz and he got over 900m on the best core. In the posted benchmarks, the 860MHz 5770 only got 415m. I'm not sure if they were running an earlier version with really bad code or what.

According to the FAQ, the benchmark only tests a single CPU. I'm not sure if GPUs are handled exactly the same way or not.

Balthazor has dual 5970s at 735MHz and got 1432m. Romir's single 5970 at 1000MHz got 1396m. You'd think that if it was only a single GPU being tested, then Romir's 35% faster card would do about 35% better. If it was testing multiple cores, you'd think that Balthazor's dual cards would be a lot faster than Romir's (4 GPUs at 735MHz = 2940MHz effective, 2 at 1000MHz = 2000MHz effective). It just seems weird that the results are so close when there's a decently large difference in the cards from whatever angle you choose to look at them. *shrug*
 
I also did a stock 5970 test earlier but didn't post it because the results were erratic to say the least.

In any case, I've finished 165 packets at 3.41 billion keys a second now (see the updated image above). My water temperature has only gone up 1c in 4 hours with 600w being pulled from the wall. That's what I was curious about.

Edit: The benchmark seems to only test a single GPU.

Code: 
Automatic processor identification tag: 8
	name: RV870, RV870
Estimated processor clock speed (0 if unknown): 725 MHz
Number of processors detected by this client: 2
Number of processors supported by this client: 128

dnetc v2.9108-517-GTR-10021520 for ATI Stream on Win32 (WindowsNT 6.1).

[Mar 31 20:59:43 UTC] RC5-72: using core #0 (IL 4-pipe c).
[Mar 31 20:59:54 UTC] RC5-72: Benchmark for core #0 (IL 4-pipe c)
                      0.00:00:08.23 [538,820,923 keys/sec]
[Mar 31 20:59:54 UTC] RC5-72: using core #1 (IL 4-pipe c alt).
[Mar 31 21:00:07 UTC] RC5-72: Benchmark for core #1 (IL 4-pipe c alt)
                      0.00:00:10.23 [426,267,612 keys/sec]
[Mar 31 21:00:07 UTC] RC5-72: using core #2 (IL 4-pipe 2 threads).
[Mar 31 21:00:14 UTC] RC5-72: Benchmark for core #2 (IL 4-pipe 2 threads)
                      0.00:00:03.33 [1,331,988,424 keys/sec]
[Mar 31 21:00:14 UTC] RC5-72 benchmark summary :
                      Default core : #0 (IL 4-pipe c)
                      Fastest core : #2 (IL 4-pipe 2 threads)
[Mar 31 21:00:14 UTC] Core #2 is significantly faster than the default core.
                      The GPU core selection has been made as a tradeoff between core speed
                      and responsiveness of the graphical desktop.
                      Please file a bug report along with the output of -cpuinfo
                      only if the the faster core selection does not degrade graphics performance.
 
InvisiBill said:
Balthazor has dual 5970s at 735MHz and got 1432m. Romir's single 5970 at 1000MHz got 1396m. You'd think that if it was only a single GPU being tested, then Romir's 35% faster card would do about 35% better. If it was testing multiple cores, you'd think that Balthazor's dual cards would be a lot faster than Romir's (4 GPUs at 735MHz = 2940MHz effective, 2 at 1000MHz = 2000MHz effective). It just seems weird that the results are so close when there's a decently large difference in the cards from whatever angle you choose to look at them. *shrug*
Click to expand...

It looks like the program might only be using 3 GPUs. That would explain the difference.

3 x 735 = 2205 MHz = 1432m
2 x 1000 = 2000 MHz = 1396m

That's just my guess. But, honestly, I haven't the slightest idea what we're talking about here.
 
Romir said:
Edit: The benchmark seems to only test a single GPU.
Click to expand...

But if that is the case, why is my 735 MHz slightly faster than your 1000 MHz?

Computational errors from the overclocking, errors that might not be significant when processing video? Or what?
 
Balthazor said:
But if that is the case, why is my 735 MHz slightly faster than your 1000 MHz?

Computational errors from the overclocking, errors that might not be significant when processing video? Or what?
Click to expand...

It's because my overclocked benchmark result is inaccurate. That's probably why there are such wide variances as noted by InvisiBill. I'll try running it a few times while logging the results.

The console shows core A and B both crunching at 1700m, and the monitor shows the combined 3400m over 5 hours.

Your results would be a staggering 4 x 1432m (5728m), enough to place you as the 13th highest daily producer.

dnetcdnetc.png

dnetc_2.png
 
Six runs @ 1000/1200. Relatively consistent here.

Code: 
Automatic processor identification tag: 8
	name: RV870, RV870
Estimated processor clock speed (0 if unknown): 1000 MHz
Number of processors detected by this client: 2
Number of processors supported by this client: 128


[Apr 01 02:42:42 UTC] RC5-72: using core #0 (IL 4-pipe c).
[Apr 01 02:42:54 UTC] RC5-72: Benchmark for core #0 (IL 4-pipe c)
                      0.00:00:10.17 [429,189,198 keys/sec]
[Apr 01 02:42:54 UTC] RC5-72: using core #1 (IL 4-pipe c alt).
[Apr 01 02:43:06 UTC] RC5-72: Benchmark for core #1 (IL 4-pipe c alt)
                      0.00:00:10.23 [426,216,728 keys/sec]
[Apr 01 02:43:07 UTC] RC5-72: using core #2 (IL 4-pipe 2 threads).
[Apr 01 02:43:12 UTC] RC5-72: Benchmark for core #2 (IL 4-pipe 2 threads)
                      0.00:00:02.54 [1,825,054,433 keys/sec]
[Apr 01 02:43:12 UTC] RC5-72 benchmark summary :
                      Default core : #0 (IL 4-pipe c)
                      Fastest core : #2 (IL 4-pipe 2 threads)

[Apr 01 02:43:43 UTC] RC5-72: using core #0 (IL 4-pipe c).
[Apr 01 02:43:54 UTC] RC5-72: Benchmark for core #0 (IL 4-pipe c)
                      0.00:00:08.23 [529,919,721 keys/sec]
[Apr 01 02:43:54 UTC] RC5-72: using core #1 (IL 4-pipe c alt).
[Apr 01 02:44:02 UTC] RC5-72: Benchmark for core #1 (IL 4-pipe c alt)
                      0.00:00:05.11 [857,141,529 keys/sec]
[Apr 01 02:44:02 UTC] RC5-72: using core #2 (IL 4-pipe 2 threads).
[Apr 01 02:44:07 UTC] RC5-72: Benchmark for core #2 (IL 4-pipe 2 threads)
                      0.00:00:02.57 [1,765,293,084 keys/sec]
[Apr 01 02:44:07 UTC] RC5-72 benchmark summary :
                      Default core : #0 (IL 4-pipe c)
                      Fastest core : #2 (IL 4-pipe 2 threads)

[Apr 01 02:44:18 UTC] RC5-72: using core #0 (IL 4-pipe c).
[Apr 01 02:44:27 UTC] RC5-72: Benchmark for core #0 (IL 4-pipe c)
                      0.00:00:06.81 [642,759,136 keys/sec]
[Apr 01 02:44:27 UTC] RC5-72: using core #1 (IL 4-pipe c alt).
[Apr 01 02:44:36 UTC] RC5-72: Benchmark for core #1 (IL 4-pipe c alt)
                      0.00:00:05.88 [756,626,535 keys/sec]
[Apr 01 02:44:36 UTC] RC5-72: using core #2 (IL 4-pipe 2 threads).
[Apr 01 02:44:41 UTC] RC5-72: Benchmark for core #2 (IL 4-pipe 2 threads)
                      0.00:00:02.55 [1,705,050,965 keys/sec]
[Apr 01 02:44:41 UTC] RC5-72 benchmark summary :
                      Default core : #0 (IL 4-pipe c)
                      Fastest core : #2 (IL 4-pipe 2 threads)

[Apr 01 02:45:01 UTC] RC5-72: using core #0 (IL 4-pipe c).
[Apr 01 02:45:10 UTC] RC5-72: Benchmark for core #0 (IL 4-pipe c)
                      0.00:00:06.83 [641,863,172 keys/sec]
[Apr 01 02:45:10 UTC] RC5-72: using core #1 (IL 4-pipe c alt).
[Apr 01 02:45:19 UTC] RC5-72: Benchmark for core #1 (IL 4-pipe c alt)
                      0.00:00:06.80 [643,713,289 keys/sec]
[Apr 01 02:45:19 UTC] RC5-72: using core #2 (IL 4-pipe 2 threads).
[Apr 01 02:45:25 UTC] RC5-72: Benchmark for core #2 (IL 4-pipe 2 threads)
                      0.00:00:02.55 [1,805,209,951 keys/sec]
[Apr 01 02:45:25 UTC] RC5-72 benchmark summary :
                      Default core : #0 (IL 4-pipe c)
                      Fastest core : #2 (IL 4-pipe 2 threads)

[Apr 01 02:45:44 UTC] RC5-72: using core #0 (IL 4-pipe c).
[Apr 01 02:45:54 UTC] RC5-72: Benchmark for core #0 (IL 4-pipe c)
                      0.00:00:08.23 [538,820,923 keys/sec]
[Apr 01 02:45:54 UTC] RC5-72: using core #1 (IL 4-pipe c alt).
[Apr 01 02:46:02 UTC] RC5-72: Benchmark for core #1 (IL 4-pipe c alt)
                      0.00:00:05.13 [854,385,791 keys/sec]
[Apr 01 02:46:02 UTC] RC5-72: using core #2 (IL 4-pipe 2 threads).
[Apr 01 02:46:07 UTC] RC5-72: Benchmark for core #2 (IL 4-pipe 2 threads)
                      0.00:00:02.55 [1,805,783,011 keys/sec]
[Apr 01 02:46:07 UTC] RC5-72 benchmark summary :
                      Default core : #0 (IL 4-pipe c)
                      Fastest core : #2 (IL 4-pipe 2 threads)

[Apr 01 02:47:07 UTC] RC5-72: using core #0 (IL 4-pipe c).
[Apr 01 02:47:16 UTC] RC5-72: Benchmark for core #0 (IL 4-pipe c)
                      0.00:00:06.80 [643,935,575 keys/sec]
[Apr 01 02:47:16 UTC] RC5-72: using core #1 (IL 4-pipe c alt).
[Apr 01 02:47:28 UTC] RC5-72: Benchmark for core #1 (IL 4-pipe c alt)
                      0.00:00:10.18 [430,427,989 keys/sec]
[Apr 01 02:47:28 UTC] RC5-72: using core #2 (IL 4-pipe 2 threads).
[Apr 01 02:47:34 UTC] RC5-72: Benchmark for core #2 (IL 4-pipe 2 threads)
                      0.00:00:02.54 [1,764,384,797 keys/sec]
[Apr 01 02:47:34 UTC] RC5-72 benchmark summary :
                      Default core : #0 (IL 4-pipe c)
                      Fastest core : #2 (IL 4-pipe 2 threads)

In one run 725/1000 resulted in:

Code: 
[Apr 01 02:49:45 UTC] RC5-72: using core #0 (IL 4-pipe c).
[Apr 01 02:49:54 UTC] RC5-72: Benchmark for core #0 (IL 4-pipe c)
                      0.00:00:06.84 [639,909,595 keys/sec]
[Apr 01 02:49:54 UTC] RC5-72: using core #1 (IL 4-pipe c alt).
[Apr 01 02:50:03 UTC] RC5-72: Benchmark for core #1 (IL 4-pipe c alt)
                      0.00:00:06.81 [645,797,908 keys/sec]
[Apr 01 02:50:03 UTC] RC5-72: using core #2 (IL 4-pipe 2 threads).
[Apr 01 02:50:10 UTC] RC5-72: Benchmark for core #2 (IL 4-pipe 2 threads)
                      0.00:00:03.43 [1,306,992,593 keys/sec]
[Apr 01 02:50:10 UTC] RC5-72 benchmark summary :
                      Default core : #0 (IL 4-pipe c)
                      Fastest core : #2 (IL 4-pipe 2 threads)

And 725/1200 resulted in:

Code: 
[Apr 01 02:53:48 UTC] RC5-72: using core #0 (IL 4-pipe c).
[Apr 01 02:54:01 UTC] RC5-72: Benchmark for core #0 (IL 4-pipe c)
                      0.00:00:10.29 [430,504,119 keys/sec]
[Apr 01 02:54:01 UTC] RC5-72: using core #1 (IL 4-pipe c alt).
[Apr 01 02:54:10 UTC] RC5-72: Benchmark for core #1 (IL 4-pipe c alt)
                      0.00:00:06.88 [636,385,574 keys/sec]
[Apr 01 02:54:10 UTC] RC5-72: using core #2 (IL 4-pipe 2 threads).
[Apr 01 02:54:16 UTC] RC5-72: Benchmark for core #2 (IL 4-pipe 2 threads)
                      0.00:00:03.43 [1,320,200,951 keys/sec]
[Apr 01 02:54:16 UTC] RC5-72 benchmark summary :
                      Default core : #0 (IL 4-pipe c)
                      Fastest core : #2 (IL 4-pipe 2 threads)
 
Well, Romir's newer overclocking results make a lot more sense, now, as does his assertion that only one GPU is being benchmarked.
 
ATI cards massacre nvidia cards on distributed computing.. only thing nvidia beats the ATI cards on is F@H.. MilkyWay@Home as another example takes 5-7 minutes a WU on a 5870 and 30-35 minutes on a nvidia card..

as for the fluctuating numbers its quite possible that the overclocks causing computation errors.. the line should stay pretty consistent and not bouncing around as much as it is.. but it could also mean that the ATI client still needs better optimization..

also another note.. the video memory clock has almost no effect on performance so if you want to try and lower the gpu temp downclock the memory to 700-800mhz..

how many crunchers are showing up on your client balthazor? also you could try looking in your settings and checking to see if its set to only run 3 crunchers.. instead of 4.. though it should be set to run as many clients as gpu's it see's.. also distributed computing which is what this program is requires a ton of shader clock speed.. so while on top of the fact that the 2 cards are running 1000mhz or 265mhz faster then your card the shaders are at 265mhz faster.. thus why it performs similar to the your 3 cards running at 765mhz.. btw its a difference of 295mhz not 205mhz.. but odds are the reason they are so close is due to the optimization because the client may not be using all 1600 shader threads..

btw its benchmarking all the cores..
[Apr 01 02:43:06 UTC] RC5-72: Benchmark for core #1 (IL 4-pipe c alt)
0.00:00:10.23 [426,216,728 keys/sec]
[Apr 01 02:43:07 UTC] RC5-72: using core #2 (IL 4-pipe 2 threads).
[Apr 01 02:43:12 UTC] RC5-72: Benchmark for core #2 (IL 4-pipe 2 threads)
0.00:00:02.54 [1,825,054,433 keys/sec]

notice the core #2 is double the score of core #1 benchmark..
 
4 crunchers show up when I run the client. I think Romir's re-run results are more consistent with his card's higher clock speeds.
 
Romir said:
In any case, I've finished 165 packets at 3.41 billion keys a second now (see the updated image above).
Click to expand...

FYI, that's just shy of 10x what my GTX285 + 9800GT put out. Wow. That should come out to 75,000+ blocks per day.


If you're going to keep running this, make sure you enter your email address in the config and join the [H] team. =)
 
Here's a stock XFX 5770 at 850/1200.
Code: 
dnetc v2.9108-517-GTR-10021520 for ATI Stream on Win32 (WindowsNT 5.2).

Automatic processor identification tag: 9
	name: RV840
Estimated processor clock speed (0 if unknown): 850 MHz
Number of processors detected by this client: 1
Number of processors supported by this client: 128

dnetc v2.9108-517-GTR-10021520 for ATI Stream on Win32 (WindowsNT 5.2).

[Apr 02 03:02:19 UTC] RC5-72: using core #0 (IL 4-pipe c).
[Apr 02 03:02:26 UTC] RC5-72: Benchmark for core #0 (IL 4-pipe c)
                      0.00:00:04.84 [[b][i]912,249,188 keys/sec[/i][/b]]
[Apr 02 03:02:26 UTC] RC5-72: using core #1 (IL 4-pipe c alt).
[Apr 02 03:02:37 UTC] RC5-72: Benchmark for core #1 (IL 4-pipe c alt)
                      0.00:00:07.67 [573,077,310 keys/sec]
[Apr 02 03:02:37 UTC] RC5-72: using core #2 (IL 4-pipe 2 threads).
[Apr 02 03:02:45 UTC] RC5-72: Benchmark for core #2 (IL 4-pipe 2 threads)
                      0.00:00:06.43 [696,234,446 keys/sec]
[Apr 02 03:02:45 UTC] RC5-72 benchmark summary :
                      Default core : #0 (IL 4-pipe c)
                      Fastest core : #0 (IL 4-pipe c)

That's close to 3x what the GTX285 does.
 
My Sapphire 5750 @ 800/1200:
Code: 
dnetc v2.9108-517-GTR-10021520 for ATI Stream on Win32 (WindowsNT 6.1).

Automatic processor identification tag: 9
	name: RV840
Estimated processor clock speed (0 if unknown): 800 MHz
Number of processors detected by this client: 1
Number of processors supported by this client: 128

dnetc v2.9108-517-GTR-10021520 for ATI Stream on Win32 (WindowsNT 6.1).

[Apr 02 04:04:38 UTC] RC5-72: using core #0 (IL 4-pipe c).
[Apr 02 04:04:46 UTC] RC5-72: Benchmark for core #0 (IL 4-pipe c)
                      0.00:00:05.22 [842,329,268 keys/sec]
[Apr 02 04:04:46 UTC] RC5-72: using core #1 (IL 4-pipe c alt).
[Apr 02 04:04:57 UTC] RC5-72: Benchmark for core #1 (IL 4-pipe c alt)
                      0.00:00:09.12 [476,767,851 keys/sec]
[Apr 02 04:04:57 UTC] RC5-72: using core #2 (IL 4-pipe 2 threads).
[Apr 02 04:05:06 UTC] RC5-72: Benchmark for core #2 (IL 4-pipe 2 threads)
                      0.00:00:06.49 [674,149,688 keys/sec]
[Apr 02 04:05:06 UTC] RC5-72 benchmark summary :
                      Default core : #0 (IL 4-pipe c)
                      Fastest core : #0 (IL 4-pipe c)

For comparison, the same card @ stock speeds:

Code: 
dnetc v2.9108-517-GTR-10021520 for ATI Stream on Win32 (WindowsNT 6.1).

Automatic processor identification tag: 9
	name: RV840
Estimated processor clock speed (0 if unknown): 700 MHz
Number of processors detected by this client: 1
Number of processors supported by this client: 128

dnetc v2.9108-517-GTR-10021520 for ATI Stream on Win32 (WindowsNT 6.1).

[Apr 02 04:12:16 UTC] RC5-72: using core #0 (IL 4-pipe c).
[Apr 02 04:12:30 UTC] RC5-72: Benchmark for core #0 (IL 4-pipe c)
                      0.00:00:12.57 [349,475,041 keys/sec]
[Apr 02 04:12:31 UTC] RC5-72: using core #1 (IL 4-pipe c alt).
[Apr 02 04:12:45 UTC] RC5-72: Benchmark for core #1 (IL 4-pipe c alt)
                      0.00:00:11.17 [386,291,532 keys/sec]
[Apr 02 04:12:45 UTC] RC5-72: using core #2 (IL 4-pipe 2 threads).
[Apr 02 04:12:55 UTC] RC5-72: Benchmark for core #2 (IL 4-pipe 2 threads)
                      0.00:00:07.41 [585,581,589 keys/sec]
[Apr 02 04:12:55 UTC] RC5-72 benchmark summary :
                      Default core : #0 (IL 4-pipe c)
                      Fastest core : #2 (IL 4-pipe 2 threads)
[Apr 02 04:12:55 UTC] Core #2 is significantly faster than the default core.
                      The GPU core selection has been made as a tradeoff between core speed
                      and responsiveness of the graphical desktop.
                      Please file a bug report along with the output of -cpuinfo
                      only if the the faster core selection does not degrade graphics performance.
 
Last edited:
Here's the stock XFX 5870 at 850/1200.

Code: 
dnetc v2.9108-517-GTR-10021520 for ATI Stream on Win32 (WindowsNT 6.1).

Automatic processor identification tag: 8
	name: RV870
Estimated processor clock speed (0 if unknown): 850 MHz
Number of processors detected by this client: 1
Number of processors supported by this client: 128

dnetc v2.9108-517-GTR-10021520 for ATI Stream on Win32 (WindowsNT 6.1).

[Apr 19 19:46:19 UTC] RC5-72: using core #0 (IL 4-pipe c).
[Apr 19 19:46:24 UTC] RC5-72: Benchmark for core #0 (IL 4-pipe c)
                      0.00:00:02.51 [1,803,493,580 keys/sec]
[Apr 19 19:46:24 UTC] RC5-72: using core #1 (IL 4-pipe c alt).
[Apr 19 19:46:30 UTC] RC5-72: Benchmark for core #1 (IL 4-pipe c alt)
                      0.00:00:04.10 [1,119,688,064 keys/sec]
[Apr 19 19:46:30 UTC] RC5-72: using core #2 (IL 4-pipe 2 threads).
[Apr 19 19:46:37 UTC] RC5-72: Benchmark for core #2 (IL 4-pipe 2 threads)
                      0.00:00:04.32 [1,087,249,010 keys/sec]
[Apr 19 19:46:37 UTC] RC5-72 benchmark summary :
                      Default core : #0 (IL 4-pipe c)
                      Fastest core : #0 (IL 4-pipe c)

1800mkeys/sec, nice! That's around 6x what the GTX285 does.
 
963/1300 5970

Code: 
dnetc v2.9108-517-GTR-10021520 for ATI Stream on Win32 (WindowsNT 6.1).

Automatic processor identification tag: 8
	name: RV870, RV870
Estimated processor clock speed (0 if unknown): 963 MHz
Number of processors detected by this client: 2
Number of processors supported by this client: 128

dnetc v2.9108-517-GTR-10021520 for ATI Stream on Win32 (WindowsNT 6.1).

[Apr 19 21:31:53 UTC] RC5-72: using core #0 (IL 4-pipe c).
[Apr 19 21:31:58 UTC] RC5-72: Benchmark for core #0 (IL 4-pipe c)
                      0.00:00:02.07 [2,146,822,074 keys/sec]
[Apr 19 21:31:58 UTC] RC5-72: using core #1 (IL 4-pipe c alt).
[Apr 19 21:32:03 UTC] RC5-72: Benchmark for core #1 (IL 4-pipe c alt)
                      0.00:00:03.41 [1,294,481,957 keys/sec]
[Apr 19 21:32:03 UTC] RC5-72: using core #2 (IL 4-pipe 2 threads).
[Apr 19 21:32:09 UTC] RC5-72: Benchmark for core #2 (IL 4-pipe 2 threads)
                      0.00:00:03.60 [1,209,927,902 keys/sec]
[Apr 19 21:32:09 UTC] RC5-72 benchmark summary :
                      Default core : #0 (IL 4-pipe c)
                      Fastest core : #0 (IL 4-pipe c)

725/1100

Code: 
dnetc v2.9108-517-GTR-10021520 for ATI Stream on Win32 (WindowsNT 6.1).

Automatic processor identification tag: 8
	name: RV870, RV870
Estimated processor clock speed (0 if unknown): 725 MHz
Number of processors detected by this client: 2
Number of processors supported by this client: 128


dnetc v2.9108-517-GTR-10021520 for ATI Stream on Win32 (WindowsNT 6.1).

[Apr 19 21:37:34 UTC] RC5-72: using core #0 (IL 4-pipe c).
[Apr 19 21:37:40 UTC] RC5-72: Benchmark for core #0 (IL 4-pipe c)
                      0.00:00:02.73 [1,604,583,190 keys/sec]
[Apr 19 21:37:40 UTC] RC5-72: using core #1 (IL 4-pipe c alt).
[Apr 19 21:37:46 UTC] RC5-72: Benchmark for core #1 (IL 4-pipe c alt)
                      0.00:00:04.55 [961,878,961 keys/sec]
[Apr 19 21:37:46 UTC] RC5-72: using core #2 (IL 4-pipe 2 threads).
[Apr 19 21:37:53 UTC] RC5-72: Benchmark for core #2 (IL 4-pipe 2 threads)
                      0.00:00:04.08 [1,058,310,104 keys/sec]
[Apr 19 21:37:53 UTC] RC5-72 benchmark summary :
                      Default core : #0 (IL 4-pipe c)
                      Fastest core : #0 (IL 4-pipe c)
 
Last edited:
You must log in or register to reply here.
Back
Top