Random Flickering Pixels

Axianamos

Honorable
Sep 20, 2013
10
0
10,520
So the other day I bluescreened. I still Haven't discovered why really. The error was

0xa0000001 (0x0000000000000005, 0x0000000000000000, 0x0000000000000000, 0x0000000000000000).

The day after this happened I hopped into a game of League. It was artifacting all over the place. Then I noticed when I was out of game that there were flickering pixels all over the place. I could click move something over them to erase them for a few seconds, so I did some looking around for causes. I noticed the flickering pixels didn't appear when I was in any other game. So I figured it wasn't my monitor. I tried rebooting the system to see if it happened in BIOS. The way I figure it, if it was the card itself and not the drivers then it would appear during startup. They didn't so I thought it might be drivers, but reinstalling them didn't fix the issue. (Thinking about it now, it was a pretty stupid assumption anyway.)

I looked through the event viewer a bit more, I figured the BSOD was related.
I got this a few times.
The driver \Driver\ATITool failed to load for the device Root\*ATITOOLDEVICE\0000.

and I've seen this more than once, different variations. 4.1, 4.2. 4.01 at one point.
The AODDriver4.1 service failed to start due to the following error:
The system cannot find the file specified. (sometimes it was saying the path could not be specified.)

I checked my temps.
2egfsdl.png


At this point I'm just praying it's a driver issue and not hardware. What are the chances of that? Pretty crap. I don't overclock at all either, just to clear that up. Everything right now is stock.

Windows 7 64 bit
AMD FX 4170 4.2 GHZ Quad Core
ATI 7850
8gb RAM
GA-970A-UD3 Motherboard
http://www.newegg.com/Product/Product.aspx?Item=N82E16817182200 PSU
 
Solution
So a few months back the problem came around again. I used a multimeter to test my power supply voltages and they tested just fine. I did a ram check to see if that was the problem and no errors came back. So I figured it was time to send the card back. Just today I took the "new" card out and put my old one in, figuring I would just use that till the replacement came in. The old card of course had no problems, much as I expected.

That is till I loaded into a game of Titanfall. It was difficult to notice but I was getting flickering pixels there as well. Unlike my newer card, this was barely noticeable. This leads me to believe the problem was there all along, but I could never see it, and the new card I got just made the problem...

Axianamos

Honorable
Sep 20, 2013
10
0
10,520
my old 6670 is still kicking around, but I don't think that would be a whole lot of help. I think I also have an Nvidia 6600. I'm sure that will be helpful someday. Maybe during the apocalypse and we REALLY need to play quake for some reason. And no I don't have another computer to shove the card into.
 

Xyloriuphon

Honorable
Nov 17, 2012
507
0
11,010
its not a matter of performance were looking for, where just wondering if tha GPU is faulty, if it was, then when you slap in any other card, your shouldn't get any flickering,

Just slap in a different card, nothing bad will happen...
 

Axianamos

Honorable
Sep 20, 2013
10
0
10,520
I did none of that and might I mention I hate it when problems go away for no reason? I did a virus scan, caught a few trojans. But I feel uncomfortable thinking it was that simple. If the problem comes back sometime soon I'll test out the other cards.

And I apologize if I seemed abrasive. I was just being sarcastic, it's what I do.
 

Xyloriuphon

Honorable
Nov 17, 2012
507
0
11,010


Yeah, I kinda have the same trait, But, glad you fixed it, just repost if something bad happens (That means ----admins---- leave the thread open, for atleast a year...)
 

Axianamos

Honorable
Sep 20, 2013
10
0
10,520
So a few months back the problem came around again. I used a multimeter to test my power supply voltages and they tested just fine. I did a ram check to see if that was the problem and no errors came back. So I figured it was time to send the card back. Just today I took the "new" card out and put my old one in, figuring I would just use that till the replacement came in. The old card of course had no problems, much as I expected.

That is till I loaded into a game of Titanfall. It was difficult to notice but I was getting flickering pixels there as well. Unlike my newer card, this was barely noticeable. This leads me to believe the problem was there all along, but I could never see it, and the new card I got just made the problem visible enough for me to see.

So I moved my gpu to my second PCIe slot. I know electrically it's only a 4x, but the card fits and nothing blew up on me. I just wanted to see if perhaps the slot was bad. But the problem persisted. I've yet to send my card back and at this point I'm not sure I should bother.

Im reading back over my earlier posts. I had no idea that had been going on since november. But I notice I mentioned a bluescreen. There were no problems prior to that, and I haven't Bsod'd since. I never managed to track down what the error actually was, so I'm wondering if it's related. Does anyone know if it could just be an OS error?

SUPER EDIT OF EDITING AWESOME. I SOLVED THE ISSUE AND I WANT TO PUNCH SOMETHING

So as it turns out my temps? I posted those if you recall. My video card was cold and needed a blanket. I overclocked it and set the idle speeds to roughly twice what they were, and the high performance mode to 3 times what it was. My temps idle around 30 - 35 now and go up to about 40 when gaming, which eliminates all my problems. Figure that one out yea? I'll likely lower the clock speeds so I can control the temperatures a bit more.

To clarify, AMD cards need to be within a certain temperature range to operate properly. the CCC tends to underclock the card when it's not being used. THe problem is that windows 7 and 8 use 3D environments for just about everything, which causes problems when the card isn't warm enough to function. I had to set the idle times by creating a profile (or preset depending on your version) in the CCC and editing the XML file of that profile.

The end result of it all is just that I fixed my computer by overclocking. Personally I think this is a good problem to have.
 
Solution

Xyloriuphon

Honorable
Nov 17, 2012
507
0
11,010
Bah, I would send it if you want to, like... It seems its still a decent card, have you overclocked it? If not, as we thought a while ago, It could possibly be just the GPU is faulty... Saying that you changed the card to a different slot and nothing happened, shows more evidence its the GPU alone, I'd still try it in another PC, but, Send it back if you still have warrenty. (Sorry to Bump thread Mod's)

EDIT: Sorry, just realized, A inanimate object needs to be in a certain temperature to function properly? usually the Max-min is around -20 degrees, but, I've never heard of a problem like this, I think there is something else causing this problem, though, were glad its gone... Respond if it comes back... Though, some solutions do seem a bit weird :D

EDIT2: Remember, OC'ing to high will cause worse artifacts then you originally had,

EDIT3: if you get more problems, just send the GPU back, As i dont think its worth more time figuring out the problems then actually playing,

EDIT4: Also, Apparently, The PSU is only intended for light uses only, So, the whole problem could actually be the PSU, I'd recommend going to EVGA or Corsair. Anywho, Sorry for all the edits :d