Discrepancy between Benchmark numbers and actual numbers
#23
what makes me wondering the most is the util percentage in your benchmark-like run for -m 0:

Code:
Time.Started.....: Fri Aug 30 12:52:59 2019 (2 secs)
...
Guess.Mask.......: ?b?b?b?b?b?b?b [7]
...
Speed.#1.........: 24835.8 MH/s (17.96ms) @ Accel:64 Loops:1024 Thr:256 Vec:4
Speed.#2.........: 20023.2 MH/s (34.71ms) @ Accel:128 Loops:1024 Thr:256 Vec:4
Speed.#3.........: 16091.3 MH/s (50.37ms) @ Accel:256 Loops:1024 Thr:256 Vec:4
Speed.#4.........: 15734.6 MH/s (38.61ms) @ Accel:128 Loops:1024 Thr:256 Vec:4
Speed.#*.........: 76685.5 MH/s
Recovered........: 0/1 (0.00%) Digests, 0/1 (0.00%) Salts
...
Hardware.Mon.#1..: Temp: 33c Fan: 44% Util: 57% Core:2040MHz Mem:6800MHz Bus:1
Hardware.Mon.#2..: Temp: 25c Fan: 40% Util: 77% Core:1392MHz Mem:5005MHz Bus:1
Hardware.Mon.#3..: Temp: 37c Fan: 48% Util: 77% Core:2037MHz Mem:3802MHz Bus:1
Hardware.Mon.#4..: Temp: 38c Fan: 71% Util: 84% Core:1379MHz Mem:3304MHz Bus:1

it's probably not the best to just use 2 seconds for benchmark, but I hope/guess it's still quite accurate in your case.

The Util is always 100% on my systems if I run this same test.
Also the Bus is only 1.

I'm wondering if there is some kind of strange bottleneck with your system, e.g. PCI speed etc. It actually shouldn't matter too much to use only 1x PCI, but maybe you can explain more what kind of system this is and if you are using some risers etc. Not sure if this is really the problem, because it's strange that the benchmark with -b shows higher speed even with almost all the same settings it should show the same speed and keep showing 100% util on all cards.

You might also play around with (physically) only testing some of the cards and/or using -d to test e.g. only with 1 GPU, with 2 GPUs, 3 GPUs etc... and see if the Util drops for you.

Maybe the first thing we should clarify now is if also after for instance 5 minutes the Util is still much less than 100% .
Thanks
Reply


Messages In This Thread
RE: Discrepancy between Benchmark numbers and actual numbers - by philsmd - 08-31-2019, 11:16 AM