1)
Message boards :
AP26 - AP27 Search :
AP27 Update Testing
(Message 141855)
Posted 1053 days ago by  Azmodes
C16 finished without a hitch:
366382 366387 0
25 366384 6171054912832631
000B2E610007764F
44121552 44121558 0
26 44121555 136926916457315893
05427BA60005EB17
47715106 47715111 0
26 47715109 161004359399459161
05B02649000530C8
81292136 81292143 640
27 81292139 224584605939537911
09B0D6D700051684
|
2)
Message boards :
AP26 - AP27 Search :
AP27 Update Testing
(Message 141846)
Posted 1054 days ago by Azmodes
C17 (Xeon Gold 6140, -t 3, Ubuntu 18.04):
366382 366387 0
25 366384 6171054912832631
000B2E610007764F
44121552 44121558 0
26 44121555 136926916457315893
05427BA60005EB17
47715106 47715111 0
26 47715109 161004359399459161
05B02649000530C8
81292136 81292143 640
27 81292139 224584605939537911
09B0D6D700051684
|
3)
Message boards :
AP26 - AP27 Search :
AP27 Update Testing
(Message 141845)
Posted 1054 days ago by Azmodes
I'll try and get one of those Xeon Gold CPUs on DO and do C16 and C17 too.
|
4)
Message boards :
AP26 - AP27 Search :
AP27 Update Testing
(Message 141844)
Posted 1054 days ago by Azmodes
And here we go, single-threaded CPU on Linux. All of these were done on 20.04 as well, but on a Ryzen 3700X this time (except for the first one). Everything looks good!
SSE2
366382 366387 0 (run on the 1950X)
25 366384 6171054912832631
000B2E610007764F
44121552 44121558 0
26 44121555 136926916457315893
05427BA60005EB17
47715106 47715111 0
26 47715109 161004359399459161
05B02649000530C8
81292136 81292143 640
27 81292139 224584605939537911
09B0D6D700051684
SSE41
366382 366387 0
25 366384 6171054912832631
000B2E610007764F
44121552 44121558 0
26 44121555 136926916457315893
05427BA60005EB17
47715106 47715111 0
26 47715109 161004359399459161
05B02649000530C8
81292136 81292143 640
27 81292139 224584605939537911
09B0D6D700051684
AVX2
366382 366387 0
25 366384 6171054912832631
000B2E610007764F
44121552 44121558 0
26 44121555 136926916457315893
05427BA60005EB17
47715106 47715111 0
26 47715109 161004359399459161
05B02649000530C8
81292136 81292143 640
27 81292139 224584605939537911
09B0D6D700051684
|
5)
Message boards :
AP26 - AP27 Search :
AP27 Update Testing
(Message 141838)
Posted 1054 days ago by Azmodes
And here's the multi-threaded results (1950X Threadripper, -t 32, Ubuntu 20.04)
SSE2
366382 366387 0
25 366384 6171054912832631
000B2E610007764F
44121552 44121558 0
26 44121555 136926916457315893
05427BA60005EB17
47715106 47715111 0
26 47715109 161004359399459161
05B02649000530C8
81292136 81292143 640
27 81292139 224584605939537911
09B0D6D700051684
SSE41
366382 366387 0
25 366384 6171054912832631
000B2E610007764F
44121552 44121558 0
26 44121555 136926916457315893
05427BA60005EB17
47715106 47715111 0
26 47715109 161004359399459161
05B02649000530C8
81292136 81292143 640
27 81292139 224584605939537911
09B0D6D700051684
AVX2
366382 366387 0
25 366384 6171054912832631
000B2E610007764F
44121552 44121558 0
26 44121555 136926916457315893
05427BA60005EB17
47715106 47715111 0
26 47715109 161004359399459161
05B02649000530C8
81292136 81292143 640
27 81292139 224584605939537911
09B0D6D700051684
Single-threaded coming later.
|
6)
Message boards :
AP26 - AP27 Search :
AP27 Update Testing
(Message 141830)
Posted 1054 days ago by Azmodes
C6 (RTX 2070 Super on Ubuntu 20.04, driver 440.82):
366382 366387 0
25 366384 6171054912832631
000B2E610007764F
44121552 44121558 0
26 44121555 136926916457315893
05427BA60005EB17
47715106 47715111 0
26 47715109 161004359399459161
05B02649000530C8
81292136 81292143 640
27 81292139 224584605939537911
09B0D6D700051684
|
7)
Message boards :
AP26 - AP27 Search :
AP27 Update Testing
(Message 141805)
Posted 1055 days ago by Azmodes
I can do C6, C8-11, C14 and C15.
|
8)
Message boards :
Number crunching :
Linux vs temperatures....help?
(Message 137553)
Posted 1216 days ago by Azmodes
I have had the same problem with my 3700X. Apparently temperature sensors for the 3000 processors don't work with Kernel 5.3 and earlier. I'm not sure if this has been addressed yet, but I've read online that it was going to be fixed with 5.4.
|
9)
Message boards :
Seventeen or Bust :
AMD 1950X - SOB WU's
(Message 137466)
Posted 1217 days ago by Azmodes
Oops. Thank you. And thank you Azmodes.
Can you please explain these results?
http://www.primegrid.com/forum_thread.php?id=8266&nowrap=true#123869
I understand the 1 w/u across 16 cores etc right down to 16 wu's with 1 core each but which is the most productive?
Somewhere in the middle I guess.
Well, no. Check the tasks per day column. The issue with my findings was that multithreading made things faster, but always hurt throughput, i.e. 16 single-core tasks were always the most productive.
Moi wrote: EDIT: It may be noteworthy that back then I did not set affinities to physical cores only. May be worth another try.
Although I just noticed that I tried it with SMT turned off too and got the same results, so I doubt core affinites with it on and 50% usage would make any difference.
|
10)
Message boards :
Seventeen or Bust :
AMD 1950X - SOB WU's
(Message 137395)
Posted 1217 days ago by Azmodes
http://www.primegrid.com/forum_thread.php?id=8266
EDIT: It may be noteworthy that back then I did not set affinities to physical cores only. May be worth another try.
|
