Broken R9 290? Strange artifacting

Discussion in 'Videocards - AMD Radeon' started by Sukovsky, Apr 16, 2014.

  1. Sukovsky

    Sukovsky Guest

    Messages:
    967
    Likes Received:
    76
    GPU:
    GTX 1080
    Hi guys,

    I have a problem with one of my 290's from Gigabyte that would seem quite obvious, but is not. One of the cards show artifacing and the other one does not. I thought the one 290 was defective, and so I sent it for RMA. However, the company as well as Gigabyte to which they sent it afterwards, it came out clean after severe testing. The weird thing is, if I put the card in another system, it SEEMS that it works without artifacing at all.
    Now the situation itself:

    Sometimes it occurs after gaming, sometimes it occurs when quitting mining. Sometimes it even occurs when starting up windows. If it happens when starting windows, the artifacts are quite severe and it usually ends in a BSOD and a restart. Sometimes but not always I have the bad luck that it gets into some kind of of loop where the whole thing repeats. Starting up, artifacting and BSOD. The artifacts are random flashing stripes that seem to me as typical bad memory artifacts. When the driver crashes, which it does alot, the artifacts are gone. It even sometimes occurred that when Windows started up with artifacts, I turned on MINING, and the artifacts would go away until I stopped it, and then the artifacts sometimes returned EVEN MORE SEVERELY.
    If I started a game while the system started with artifacts, it usually ended in a BSOD sooner or later.

    I have tried a ton of things and I'm totally confused here, as I cannot reproduce the problem in my dad's pc, which almost makes me think this could be another part of the system that's causing the malfunction.

    A few of the things I tried:
    -Swapping drivers from 13.12 to the 14.1 beta's (ofcourse with driversweep in safe mode)
    -Swapping the cards positions as well as taking the good card out
    -Putting them in different slots
    -Downclocking to reference speeds (temperature has not been an issue at all btw)
    -Put the 'good' card in every same scenario as the bad one with different outcomes.
    -Updated bios (both the same)

    One would say that the other card is obvious faulty, as they are perfect comparable as they are the same. Ran 2 reference 290's from Sapphire before without issues.

    I'm at a complete loss here, but what could another factor be? Memory? HDD? No clue, and I usually know my way around when **** hits the fan.
     
  2. AcidSnow

    AcidSnow Master Guru

    Messages:
    438
    Likes Received:
    24
    GPU:
    Sapphire 5700XT
    What BSOD code are you receiving? I'm going to assume you're using Win7, so make sure Windows is set to save BSOD reports, and then go download bluescreenview-x64 from NirSoft. Then after you get your next BSOD, run the program and you'll get an idea of what is causing the problem.

    Did you buy your 290s used?
     
  3. Sukovsky

    Sukovsky Guest

    Messages:
    967
    Likes Received:
    76
    GPU:
    GTX 1080
    I've seen a few, but they are different every time. One I took a picture of is 0x0000001a. I think I've also seen something like 0x00000050 and 0x0000007a

    And I bought the 2 cards new.
     
  4. Fender178

    Fender178 Ancient Guru

    Messages:
    4,194
    Likes Received:
    213
    GPU:
    GTX 1070 | GTX 1060
    Hmm It could be the card or it could be the PCI-E slots are faulty but since you moved the cards around into different slots and it still did the same thing even with the good card taken out. Does the card artifact when you are in 2d mode just surfing the web and not putting any stress on the cards such as gaming or mining? The reason why I ask this is because it might be a power supply problem even though you have a 1000W PSU club3d might be a terrible/cheap brand.
     

  5. Sukovsky

    Sukovsky Guest

    Messages:
    967
    Likes Received:
    76
    GPU:
    GTX 1080
    No, Club3d is pretty sold. Also, like I said, I've used 2 Reference 290's in crossfire without issues, so one from Gigabyte should roll without problems. Also, the stuff happens without any load, during or after load. Temps aren't an issue, even idle it shows it. Actually, idle the problems are more than at load.
     
  6. Fender178

    Fender178 Ancient Guru

    Messages:
    4,194
    Likes Received:
    213
    GPU:
    GTX 1070 | GTX 1060
    OK then I would try running memtest to test your RAM.

    And if that is OK then it could be a falty memory module on your graphics card.
     
    Last edited: Apr 16, 2014
  7. Sukovsky

    Sukovsky Guest

    Messages:
    967
    Likes Received:
    76
    GPU:
    GTX 1080
    I actually disabled the XMP profile for my ram. It's Corsair Vegeance 1600 ram, but it now runs at 1333. Seems to make a difference. Artifacts from system memory, wut.
     
  8. AcidSnow

    AcidSnow Master Guru

    Messages:
    438
    Likes Received:
    24
    GPU:
    Sapphire 5700XT
    I was going to say RAM, because I've also been getting some BSOD over the past few weeks, which I wasn't getting... I just assumed it was my drivers, or used 290. Anyway, my PC is getting old, and I hash all my Firefox pages though my RAM - not on my SSD C: drive.
     
  9. Sukovsky

    Sukovsky Guest

    Messages:
    967
    Likes Received:
    76
    GPU:
    GTX 1080
    Soooo...everything was fine and dandy after I replaced my mobo with a Asrock Extreme 4, but since yesterday the problems seem to have returned in the same way. And it actually happens on both cards. I'm starting to think that every once in a while when mining causes a driver crash, it completely corrupts every attempt after that to reinstall a driver or whatever. Really at a loss here
     

Share This Page