R9 390x instability issues.

Jun 26, 2018
5
0
10
0
To begin sorry if I choose a wrong thread or anything its my first time here and im not from English speaking country so sorry for my gramar and spelling mistakes.

Specs:i5 4690k(no OC,Watercooled)
Sapphire R9 390X Nitro 8gb
Asus Z97-A
Kingston Hyperx DDR3 2x8gb
128gb SSD for Win7
1000gb HDD for multimedia


6 months ago I got my first crash while playing Fallout.New Vegas with mods and ENBs.
This crash is none of that fancy bluescreen crashes but blackscreen and repeating sound of what ever was playing at the moment and only way to fix it is to force shutdown my PC.
So I do some messing around and i find a 12 to 13 mV OC on GPU Core in MSI Afterburner(I use it more as a monitoring tool than what its suposed to do).
I check online and i find that it is a MSI AB bug and i should uninstall MSI AB so I do.
And with that my problems were solved.

But i wouldnt be here if they were gone for good now would I.
A week ago crashes started again and first thing I did was install TRIXX program from Sapphire because i was burned before by Afterburner,and there it was a 12-13 mV OC on GPU Core.
Cant remove the OC it just gets back to former value after doing anything on PC.
Crashes happen without warning or pattern.
It happens in games and even in desktop but ironically almost never in FurMark maybe 3 times on 20 to 25 test in last 6 months.
Fresh install with DDU.
Movies and in-game cutscenes sttuter like crazy.
I tried raising fan speeds on GPU and in case(H440) but its like it only delays the crash so thats weird.

HELP PLEASE I DONT KNOW WHAT TO DO ANYMORE!!!!!!
With the GPU mining craze i cant aford a new GPU or used.
 
Have you tried the wattman GPU tuner in the Radeon settings? Back when I had an R9 290x, I couldn't work out why my settings would keep changing, it turns out wattman over-rides the changes I was making.

From what I remember, you open wattman from the Radeon control panel in the "games" section, it's labelled something like "global wattman", from there you can set clock speeds and voltage.
 
Jun 26, 2018
5
0
10
0
Yes I tried that and it didnt help then I fresh installed my drivers and didnt agree to the Wattman Terms of service so it should be OFF but in TRIXX my 12-13 OC is still there.
I should add that VRM is at 53C at idle.
Maybe TRIXX is bugy but the instability is still here,this morning i had a crash while playing Nier Automata.
I stress tested with Prime95-no crash
Stress tested with FurMark at MAX settings and SOMETIMES it crashes not always it can run for 1 hour no crash but do it again and after 5 mins crash.
Its like when you push the OC too far but without the artifacts.
My only temp solution is ramping up my GPU fans to 100% and praying for the best,plus the perfomance seems low for this GPU.

Oh and thx for reply.
 
Jun 26, 2018
5
0
10
0
So I decided to install MSI AfterBurner to check if i missed something and I found in settings "Force constant voltage" option,turning that ON removed my 12-13mV OC on Core Voltage in TRIXX and MSI AB.
I will stress test my PC like i did before FurMark and Prime95 at the same time and seperate and if that goes good i will move onto games.

If you have any other ideas about cause fell free to share because my last "SOLUTION" lasted for 4 months.

Movies and cutscenes still sttuter but its not sttuter more like low fps but it doesnt make sense because they are videos.

UPDATE:

My OC is back which is weird because I restarted my PC more than 5 times,stress tested and played games for more than 2 hours but while i was pushing the PC to max with Witcher 3 on Ultra it crashed and when i checked my Core Voltage was on +13mV.

So the problem is back but ill leave my response beacuse it could help somebody,maybe.

 
Jun 26, 2018
5
0
10
0


**** sorry,my bad.

XFX PRO SERIES Semi-Modular 850W PSU

In the mean time ill try Driver rollback.

UPDATE:

Original Adrenaline drivers 17.12.1. are a no go.
They worked fine before.

UPDATE:

I installed newest drivers(not beta) and it is worse then ever i cant enable Vsync in any games and they sttuter like crazy plus there are more crashes than ever.
 
Jun 26, 2018
5
0
10
0
Good news!

I know what is up with my GPU,bad news its dead.

I took it to shop after not reciving any signal from my GPU to monitor,guy there said its because of heat distribution while cooling that the lid on the GPU core got seperated and that it caused the instability and its death in the end.

So take this as a warning,anybody who has similar issues should consult with IT specialist as soon as posible.
 

ASK THE COMMUNITY

TRENDING THREADS