x38g vs x41g on an i5 750 w/3 GTX295 cards or whatever

That's a VLAR test task Victor & no surprise, as I slackened off those for usability & further work while the project doesn't send them to GPUs. wait for the 'quick timetable' that gets displayed at the end of the bench, with all V6 & V7 pg set (8 test tasks X 2 applications)

Jason
____________
"It is not the strongest of the species that survives, nor the most intelligent that survives. It is the one that is the most adaptable to change."
Charles Darwin

Ok, I thought I'd post what I've seen, as the test just confirms what I know from personal experience, It was suggested I post this here, so I did. No harm done I hope.

Victor

Quick timetable

WU : PG0009.wu
Lunatics_x38g_win32_cuda32.exe -verb -nog :
Elapsed 166.310 secs
CPU 3.822 secs
Lunatics_x41g_win32_cuda41rc2.exe -verb -nog :
Elapsed 158.840 secs, speedup: 4.49% ratio: 1.05x
CPU 2.870 secs, speedup: 24.91% ratio: 1.33x

WU : PG0009_v7.wu
Lunatics_x38g_win32_cuda32.exe -verb -nog :
Elapsed 193.548 secs
CPU 16.068 secs
Lunatics_x41g_win32_cuda41rc2.exe -verb -nog :
Elapsed 169.510 secs, speedup: 12.42% ratio: 1.14x
CPU 5.772 secs, speedup: 64.08% ratio: 2.78x

WU : PG0395.wu
Lunatics_x38g_win32_cuda32.exe -verb -nog :
Elapsed 24.650 secs
CPU 4.836 secs
Lunatics_x41g_win32_cuda41rc2.exe -verb -nog :
Elapsed 23.550 secs, speedup: 4.46% ratio: 1.05x
CPU 2.980 secs, speedup: 38.38% ratio: 1.62x

WU : PG0395_v7.wu
Lunatics_x38g_win32_cuda32.exe -verb -nog :
Elapsed 50.970 secs
CPU 15.990 secs
Lunatics_x41g_win32_cuda41rc2.exe -verb -nog :
Elapsed 34.450 secs, speedup: 32.41% ratio: 1.48x
CPU 6.146 secs, speedup: 61.56% ratio: 2.60x

WU : PG0444.wu
Lunatics_x38g_win32_cuda32.exe -verb -nog :
Elapsed 23.240 secs
CPU 4.274 secs
Lunatics_x41g_win32_cuda41rc2.exe -verb -nog :
Elapsed 22.350 secs, speedup: 3.83% ratio: 1.04x
CPU 3.136 secs, speedup: 26.63% ratio: 1.36x

WU : PG0444_v7.wu
Lunatics_x38g_win32_cuda32.exe -verb -nog :
Elapsed 49.500 secs
CPU 16.848 secs
Lunatics_x41g_win32_cuda41rc2.exe -verb -nog :
Elapsed 33.110 secs, speedup: 33.11% ratio: 1.50x
CPU 5.741 secs, speedup: 65.92% ratio: 2.93x

WU : PG1327.wu
Lunatics_x38g_win32_cuda32.exe -verb -nog :
Elapsed 16.050 secs
CPU 4.243 secs
Lunatics_x41g_win32_cuda41rc2.exe -verb -nog :
Elapsed 15.110 secs, speedup: 5.86% ratio: 1.06x
CPU 2.309 secs, speedup: 45.58% ratio: 1.84x

WU : PG1327_v7.wu
Lunatics_x38g_win32_cuda32.exe -verb -nog :
Elapsed 81.660 secs
CPU 36.021 secs
Lunatics_x41g_win32_cuda41rc2.exe -verb -nog :
Elapsed 42.191 secs, speedup: 48.33% ratio: 1.94x
CPU 9.079 secs, speedup: 74.80% ratio: 3.97x

Ok then I'm rerunning the test as I thought that was It, no more output.

Thanks Victor! and the results are interesting too.

Firstly, for V6 on main, ignoring VLAR, as not distributed by the project & currently dialled back, at the mid angle ranges you're seeing about a 1-2% slowdown, with about a 4.5% speedup for VHAR 'shorties'.

Whether or not that's a good or bad situation will depend a bit on the work mix. 'Luckily' we seem to have been getting shorty storm after shorty storm lately, so probably x41g would constitute a small net gain under those circumstances on that host. Having said that, the mid angle ranges should not have slowed down, but in fact should really have slightly sped up, like they did in Mark's case. Upping to the new 290 beta driver might have some impact there because of changes to the way CPU is used on the host, so worth a comparative run when you update those in the future.

V7 performance, by comparison, shows a performance improvement at all angle ranges (as expected), of ~3 to ~33% varying by angle range on that system.

When considering the nature of the work that has been done so far, the bench does show subtly, when you have some knowledge about how the tasks are processed, that relatively unoptimised pulsefinding has fairly unchanging, or even reduced performance on that GPU+system under the newer build, while the highly optimised VHAR & V7 code portions are pulling ahead noticeably. That means as similar refinements are done to the problem areas at fault for reduced VLAR performance and mediocre V6 midrange on that GPU, there is likely a lot more ground to be gained in coming builds.

Another observation is that x41g benched to be using about 7 to ~13% less CPU under V6 conditions, so whether freeing up that CPU gives you extra CPU AKv8b2 processing that's worthwhile enough to make up for that ~1-2% mid angle range loss would depend on the rest of the system & how you run etc.

I should probably add at this point that refinement is also directed at reducing inconclusives longer term, so I'd suggest the increased reliability and stability of newer builds should offset minor performance variations somewhat.

In summary, I think your immediate choice with V6 crunching, on main only, is between nice fast 'shorties', lower CPU usage & easier validation with x41g, or ~1-2% quicker mid Angle range tasks with x38g. That's the way it looks from that bench, but also consider that changing driver might well change the picture again, and the lower system contention is designed specifically to lighten the load on the host, so feeding more GPUs becomes easier while at full crunch.

Hope that helps, It certainly illustrates the code areas I've been looking into could do with the attention.

Jason

Your Data:

Quick timetable

WU : PG0009.wu
Lunatics_x38g_win32_cuda32.exe -verb -nog :
Elapsed 337.613 secs
CPU 14.258 secs
Lunatics_x41g_win32_cuda32.exe -verb -nog :
Elapsed 351.522 secs, speedup: -4.12% ratio: 0.96x
CPU 14.258 secs, speedup: 0.00% ratio: 1.00x

WU : PG0009_v7.wu
Lunatics_x38g_win32_cuda32.exe -verb -nog :
Elapsed 391.040 secs
CPU 25.132 secs
Lunatics_x41g_win32_cuda32.exe -verb -nog :
Elapsed 378.398 secs, speedup: 3.23% ratio: 1.03x
CPU 19.219 secs, speedup: 23.53% ratio: 1.31x

WU : PG0395.wu
Lunatics_x38g_win32_cuda32.exe -verb -nog :
Elapsed 54.788 secs
CPU 15.194 secs
Lunatics_x41g_win32_cuda32.exe -verb -nog :
Elapsed 55.835 secs, speedup: -1.91% ratio: 0.98x
CPU 14.134 secs, speedup: 6.98% ratio: 1.07x

WU : PG0395_v7.wu
Lunatics_x38g_win32_cuda32.exe -verb -nog :
Elapsed 108.480 secs
CPU 26.146 secs
Lunatics_x41g_win32_cuda32.exe -verb -nog :
Elapsed 84.954 secs, speedup: 21.69% ratio: 1.28x
CPU 18.549 secs, speedup: 29.06% ratio: 1.41x

WU : PG0444.wu
Lunatics_x38g_win32_cuda32.exe -verb -nog :
Elapsed 49.942 secs
CPU 14.726 secs
Lunatics_x41g_win32_cuda32.exe -verb -nog :
Elapsed 50.344 secs, speedup: -0.80% ratio: 0.99x
CPU 13.478 secs, speedup: 8.47% ratio: 1.09x

WU : PG0444_v7.wu
Lunatics_x38g_win32_cuda32.exe -verb -nog :
Elapsed 100.818 secs
CPU 25.537 secs
Lunatics_x41g_win32_cuda32.exe -verb -nog :
Elapsed 76.799 secs, speedup: 23.82% ratio: 1.31x
CPU 18.034 secs, speedup: 29.38% ratio: 1.42x

WU : PG1327.wu
Lunatics_x38g_win32_cuda32.exe -verb -nog :
Elapsed 37.101 secs
CPU 14.446 secs
Lunatics_x41g_win32_cuda32.exe -verb -nog :
Elapsed 35.415 secs, speedup: 4.54% ratio: 1.05x
CPU 12.464 secs, speedup: 13.72% ratio: 1.16x

WU : PG1327_v7.wu
Lunatics_x38g_win32_cuda32.exe -verb -nog :
Elapsed 163.242 secs
CPU 40.529 secs
Lunatics_x41g_win32_cuda32.exe -verb -nog :
Elapsed 109.846 secs, speedup: 32.71% ratio: 1.49x
CPU 24.430 secs, speedup: 39.72% ratio: 1.66x

You're welcome Jason, I just crunch on the gpu, I don't use any of the AKv8b or AKv8b2 software at all anymore. I'm downloading the 290.36 beta driver and I'll retest once I get everything in order here, Hey with 6 gpus, It had better be in order, imagine what 12 would be like, twice as long maybe to just install.

OK, that's a spanner in the works. When you initially ran the installer did you alter the folder to point to a writeable location ? Not an issue I've come across, Running that here comes up with a box for me to tell it to extract 'wherever' I tell it to, so I would have presumed telling it somewhere else would work. Not having a C drive is unusual, but shouldn't prevent the initial extraction.

I wouldn't necessarily panic until next WHQL comes out & it does the same thing. Maybe someone else has come across something like this & will chime in, in the meantime.

Jason
____________
"It is not the strongest of the species that survives, nor the most intelligent that survives. It is the one that is the most adaptable to change."
Charles Darwin