Posts: 8
Threads: 1
Joined: Dec 2020
(12-28-2020, 07:38 PM)dismal Wrote: (12-28-2020, 06:21 PM)CursedGPU10000 Wrote: (12-27-2020, 09:21 PM)dismal Wrote: why don't you run it when it's the only one connected?
Because I want to run a multiple GPU rig.
That is very nice, but I proposed to run it solo for a test.
Seems to perform normal when it's the only card plugged in. I really don't issue what the issue could be.
I'm trying to think maybe it's Mobo which is the issue?
MSI Z270 GAMING PRO CARBON.
Although I'm fairly confident this is a premium Mobo (which i've had for years without issues).
Posts: 13
Threads: 0
Joined: Sep 2020
well now you can try to run it in a different slot / with another (one, then two) card installed.
try different configurations, use logic, that way you exclude some stuff and come to conclusions.
it's the only way to debug any problem
Posts: 381
Threads: 1
Joined: Aug 2014
(12-27-2020, 07:12 PM)CursedGPU10000 Wrote: Speed.#1.........: 6315.8 MH/s (283.58ms) @ Accel:64 Loops:1024 Thr:1024 Vec:1
Speed.#2.........: 2801.4 MH/s (288.74ms) @ Accel:64 Loops:512 Thr:1024 Vec:1
Speed.#3.........: 3707.7 MH/s (346.00ms) @ Accel:64 Loops:1024 Thr:1024 Vec:1
Speed.#*.........: 12824.9 MH/s
In this running the GTX 1080 also seems to perform not very good. They are all at the same bus lane? How many lanes does each card get? Hashcat is usually not affected by the number of lanes, but maybe there are latency issues.
Does the slow performance stay on the specific card when you switch the two Tis? Any fancy BIOS on the cards? Which brand/type are you using anyway?
Posts: 8
Threads: 1
Joined: Dec 2020
To update;
To my knowledge apparently the issue was due to bottlenecking because of the CPU temperature... which my braindead idiocy never thought to check. I know thermal throttling was mentioned on this thread, but I for some reason never thought it'd be my CPU.... I opened Speccy when hashcat was running and it was hitting 90-95c. I removed CPU cooler + CPU, removed the abundance of dust, and reapplied new thermal paste. Now I'm sitting at a healthy 58-65c under full GPU utilization and speeds seem to be normal... for now... so I'm concluding that the CPU temp was the issue!
Thanks
Posts: 8
Threads: 1
Joined: Dec 2020
A further update, this did not entirely fix my problem. It seems this card may have some sort of power delivery issues or something. It can be fine, like it was today for hours, now in the past hour the speeds have dropped insanely now. To about 1/3 of what a 1080ti should be receiving. However when restarting PC the speeds go back to normal.
I've tried different PCI slots, different PSU, different PSU cables, different riser.
I guess I've just got a shit card, and not sure how to do more advanced tests on it to try and diagnose the problem.
Gg.
Posts: 2,301
Threads: 11
Joined: Jul 2010
Thanks for the detailed reply and great to hear you were able to identify the problem (to some extend at least).
I don't know how old your card is but maybe try to RMA it?