GT1030 faster than Vega 11?

Message boards : Number crunching : GT1030 faster than Vega 11?
Message board moderation

To post messages, you must log in.

AuthorMessage
Ryan Munro

Send message
Joined: 5 Feb 06
Posts: 63
Credit: 18,519,866
RAC: 10
United Kingdom
Message 2011706 - Posted: 12 Sep 2019, 9:18:03 UTC

I have a machine set up with a Ryzen 2400g and a GT1030, the Vega graphics complete a unit in about 50 mins whereas the 1030 does one in about 40, looking at the specs of both the Vega should be faster? (883gflops vs 1760gflops).

I am running WU's on the CPU as well but CPU load is only about 80% when running 6 CPU units, I thought that the CPU may be pulling from the GPU's power budget but even when no CPU units are being crunched it seems to make little difference.

Is the Nvidia app just optimised more? credit wise both seem to give about the same.
ID: 2011706 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13746
Credit: 208,696,464
RAC: 304
Australia
Message 2011708 - Posted: 12 Sep 2019, 9:29:17 UTC - in response to Message 2011706.  

I am running WU's on the CPU as well but CPU load is only about 80% when running 6 CPU units, I thought that the CPU may be pulling from the GPU's power budget but even when no CPU units are being crunched it seems to make little difference.
They share power, cache, memory bandwidth, and thermal limits. All of those combined, will result in limitations on the iGPU performance, and by using the iGPU it will limit the performance of the CPU.
And considering the very poor abilities of the GT 1030, anything being outperformed by it is of concern.
Grant
Darwin NT
ID: 2011708 · Report as offensive
Ryan Munro

Send message
Joined: 5 Feb 06
Posts: 63
Credit: 18,519,866
RAC: 10
United Kingdom
Message 2011712 - Posted: 12 Sep 2019, 10:06:37 UTC - in response to Message 2011708.  

Cheers sounds like things are working as they should then, ill leave it to it :).

Is it worth not using the onboard graphics and just using the CPU and 1030 in that case? the Vega graphics still crunch a lot faster than a single CPU unit will.
ID: 2011712 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13746
Credit: 208,696,464
RAC: 304
Australia
Message 2011715 - Posted: 12 Sep 2019, 10:27:50 UTC - in response to Message 2011712.  

Is it worth not using the onboard graphics and just using the CPU and 1030 in that case? the Vega graphics still crunch a lot faster than a single CPU unit will.
Maybe.
In the case of the Intel iGPUs, it's definitely best not to use them at all, however the AMD iGPUs are much more capable, and the loss of CPU output could be offset by the greater iGPU output.
The only way to be sure would be to work out how many WUs you are doing per hour at present, then disable iGPU crunching and see if your hourly output improves, gets worse, or remains unchanged. The biggest problem in doing this is the differing runtimes of different WUs- you need to make sure you're comparing the same things.
Grant
Darwin NT
ID: 2011715 · Report as offensive
Profile Mike Special Project $75 donor
Volunteer tester
Avatar

Send message
Joined: 17 Feb 01
Posts: 34258
Credit: 79,922,639
RAC: 80
Germany
Message 2011720 - Posted: 12 Sep 2019, 12:41:59 UTC

The 2400G only has 4 physical cores, so i suggest to run at max 4 CPU tasks.
This should give the Vega 11 enough resources so your CPU and GPU times could improve.

Not to forget i would test for at least one day because tasks can vary.


With each crime and every kindness we birth our future.
ID: 2011720 · Report as offensive

Message boards : Number crunching : GT1030 faster than Vega 11?


 
©2024 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.