PDA

Archiv verlassen und diese Seite im Standarddesign anzeigen : Asus strix 2080ti Nvlink OC issue

Erwan
2019-03-11, 17:24:29
Hello,

I build a dual 2080ti computer a few weeks ago. After a few weeks one of my gpu died probably because of faulty memory chips. I got a replacement card from the store , exact same model. Everything seems to be running fine again but yesterday I went into afterburner to try some overclocking. First card seems to be overclocking nicely and stable , but the new card seems to crash the system as soon as I Overclock. I ran OC scanning on both and it show some really strange numbers , first card seems to be allowed to go up to +170 and second card seems to not go higher than +90. For both OC scanner seems to say that my dominant limiter is the power. I have no idea what is going on. I thought maybe I didnt have luck again with the second card, but without overclocking everything is stable and smooth. Any idea on what is going on ?

Intel Core i9-9900K
ASUS ROG Maximus XI Code
G.Skill Trident Z RGB Series 32GB
EVGA PowerSupply (PSU) Gold 1000W Modular
Samsung SSD 960 EVO 1TB
Samsung SSD 960 EVO 250GB
2x ASUS ROG Strix GeForce RTX 2080 For 11GB Advanced
Mortalvision
2019-03-11, 18:20:25
can you give us the full system specs, please? Especially mainboard, cpu, ram, storage and above all the PSU!
cR@b
2019-03-11, 18:32:49
I think you've got bad luck with the new card or it has a contact issue between cooler and GPU DIE. Are you using the original cooler or warter cooling on your cards?
Erwan
2019-03-11, 19:01:32
can you give us the full system specs, please? Especially mainboard, cpu, ram, storage and above all the PSU!


Sure , sorry for not sharing this in OP

Intel Core i9-9900K
ASUS ROG Maximus XI Code
G.Skill Trident Z RGB Series 32GB
EVGA PowerSupply (PSU) Gold 1000W Modular
Samsung SSD 960 EVO 1TB
Samsung SSD 960 EVO 250GB
2x ASUS ROG Strix GeForce RTX 2080 For 11GB Advanced


Thanks for the help
Erwan
2019-03-11, 19:03:17
I think you've got bad luck with the new card or it has a contact issue between cooler and GPU DIE. Are you using the original cooler or warter cooling on your cards?

Yes I was thinking the same thing, but when I do not OC it runs perfectly. That makes me think something else is going wrong.
I use custom cooler on first card and stock on second . The faulty one is the stock one.
Mortalvision
2019-03-11, 19:20:21
I suppose you have already tried switching slots before using sli?

Else: check what string you get the pcie power from in case your psu has multirail.

then, there is always the silicon lottery on top of all
Erwan
2019-03-11, 19:38:50
I suppose you have already tried switching slots before using sli?

Else: check what string you get the pcie power from in case your psu has multirail.

then, there is always the silicon loztery on top of all

Actually I have try every PSU connections already , I havent tried to switch card yet as custom gpu cooler is a pain in the ass to take off. Also didnt really wanna check that since first 2 gpu were running great overclocked at plus 170. I will try that tho as I am running out of options.
Just tried Time spy extreme again and the process goes through the end at stock setting, but as soon as I add plus 5 on core clock it crashes before the end. Also I noticed during OC Scan first gpu takes around 25 minutes to go
through the process. Second gpu is done in less than 13 minutes.
Mortalvision
2019-03-11, 20:51:55
If oc does not work, give undervolting a try. I dunno about 2080Ti UV ideas though. Maybe someone else can help?
Erwan
2019-03-11, 21:25:22
Ok, so everything seems to be almost back to normal now. Well at least untill plus 170. What I really dont get tho are the numbers given by OC Scanner . One card show up as safe up to 170 and the other one up to 90. Thats seems like a very big difference to me knowing that none of the cards are limited by temp but only power. Thanks for your help mate, very appreciated.