R9 290 - Strange artifacts appearing during both idle and gameplay

Amanatsu

Reputable
Mar 1, 2014
3
0
4,520
Before I get started, specs:

CPU: Intel Core i5-4670K @ 3.4GHz
GPU: AMD Radeon R9 290
Mobo: ASUS Z87 PRO V Edition
PSU: CORSAIR HX Series HX750 750W
(If you need anything else, let me know)

During the past week or so my computer has been displaying strange artifact-type things and it's really bothering me. I can't pinpoint whether or not it's my PCI-E slot or my graphics card itself. I'm leaning more towards the card itself right now, but idk, there are arguments both ways I guess.

http://imgur.com/WSW7N42,udJggUr,TlKXclb#0

The images above are from various points in the week. The first one (TV static-looking one) happened on more than one occasion before I decided to change the PCI-E slot my graphics card was connected to. The following two were from when I was in-game on FF14:ARR, after I had swapped the PCI-E slot it was connected to.

I am currently running this custom fan spread on my card using MSI Afterburner: http://i.imgur.com/MuocsVi.png (This does not appear to be a problem)

- None of the parts in my computer are currently over- or under-clocked.
- My 290 is stock.
- This appears to happen whether or not I'm in game. EDIT: Just kidding. It appears to only be happening when I close or open FF14:ARR.

If anyone has any idea what could be causing this/how I can fix it, please let me know. I've already contacted AMD about this but I don't foresee getting a response until Monday at the earliest.

EDIT: I forgot to mention, my graphics card isn't running particularly hot. I sit around 45C idle, 50C when I'm watching a video or something similar, and around 60-67C when running FF14:ARR at maximum settings.
EDIT2: Also worth noting, I sometimes get the "a display driver has crashed and recovered" message when I get the green dots. Could this be a TDR issue?
 
Solution

pyr0_m4n

Honorable
Feb 4, 2013
950
0
11,360
When the display driver crashes, this can be caused by a number of things.
1. Overheating
2. Unstable overclock
Since you said you didn't overclock and the heat is within standard operating temps, those options can be eliminated.
3. Try installing a different driver version. If there is a newer driver available for your card, use it. If you are currently using the most recent version, try rolling back to a previous driver.
4. Memory issue. This is a very rare issue where the memory controller on the graphics card is failing. It's an unfixable problem. Artifacts are normally caused by a unstable memory speed on the graphics card. Using a program such as MSI Afterburner, I would suggest trying to lower the memory clock by 50MHz and seeing if that solves the problem.
 

Amanatsu

Reputable
Mar 1, 2014
3
0
4,520


I've been talking to my friend about it and I've reseated the card, uninstalled the drivers using DDU, reinstalled the newest stable version of CCC, and increased the "CoreClockTarget" and "MemoryClockTarget" values based on a post on MSI's website about bluescreening, which has happened twice today. It seems stable so far, but I have yet to try out anything graphics intensive. The link to the post is here. My only concern is that in my Profiles.xml file, nothing had a value. Most notably, there were no voltage target values. Also, every time I open CCC the values all reset to 0.

EDIT: Started running afterburner after I had redone everything listed above, it ended up resetting my values to 0 in the profiles.xml and the driver failed and recovered twice in a row, almost immediately after I started it. I've turned it off for now, but my gfx card will surely run super hot until I'm able to find a way to fix the issue.
 
Solution