Overclock.net › Forums › Graphics Cards › NVIDIA › Temporary Artifacts?
New Posts  All Forums:Forum Nav:

Temporary Artifacts? - Page 2

post #11 of 23
Hey Booty,

Quick question, do you leave your computer on 24/7?

Also I'm assuming (cause I know you know your stuff) you switched your cards so your secondary is running as your primary and you tested that out?
     
  hide details  
Reply
     
  hide details  
Reply
post #12 of 23
Thread Starter 
Quote:
Originally Posted by Arizonian View Post
Hey Booty,

Quick question, do you leave your computer on 24/7?

Also I'm assuming (cause I know you know your stuff) you switched your cards so your secondary is running as your primary and you tested that out?
Yes and yes.
post #13 of 23
There is a little bug with the 500 series for some reason that when left on 24/7 after a long time, each card varies but it's approximately 30 days or more, artificating happens. All it takes is a complete shut down with complete power off. Card cools off. Restart your computer and the problem is gone. Try that first, then give it the gaming test and see if you have that same problem. I'll be very insterested in knowing. So far that little bug has only effected the 570 and 580 as far as I know.

It never happens to me because I shut down my computer when I go to work each day. Only left on during the weekend and I power down to sleep mode when I go to sleep at night.

Lastly, I saw spotting in Crysis when entering a certain checkpoint and I thought for the longest time it was my card. It never happened in any other game. Turns out it was a software bug in Crysis.
     
  hide details  
Reply
     
  hide details  
Reply
post #14 of 23
Thread Starter 
Quote:
Originally Posted by Arizonian View Post
There is a little bug with the 500 series for some reason that when left on 24/7 after a long time, each card varies but it's approximately 30 days or more, artificating happens. All it takes is a complete shut down with complete power off. Card cools off. Restart your computer and the problem is gone. Try that first, then give it the gaming test and see if you have that same problem. I'll be very insterested in knowing. So far that little bug has only effected the 570 and 580 as far as I know.

It never happens to me because I shut down my computer when I go to work each day. Only left on during the weekend and I power down to sleep mode when I go to sleep at night.
Hmm, I hadn't heard about that. Though I don't think that's the problem here.

I generally leave my comp on, but over the past couple of weeks I've shut it down and unplugged it to add/remove various SSD boot drives.

It's worth a shot though, considering there's nothing else to lose
post #15 of 23
Yeah. Funny thing is it also happens again around the same time. For instance if your card artifacts after 45 straight days of being on, after the cold reboot, around the same time 45 days later the problem crops up. It's like a timer reset of some sort.

It's not a serious issue but the folders were the first to discover it on EVGA. We started a spread sheet and not only EVGA but other card owners like ASUS, MSI, Gigabyte, Sparkle came over started posting on our website in hopes to find a solution for their cards as well. The EVGA forum is a very good place and active compared to other companies forums btw. Part of the great support offered but it's free for all to use.
     
  hide details  
Reply
     
  hide details  
Reply
post #16 of 23
Thread Starter 
Update #2: Yeah, it's definitely the card. Rainbow artifacts in any 3D app now. I've never had a card degrade so fast , but I guess if it was never really stable to begin with it was just a matter of time.

I'm not too upset though. The card that died was a terrible clocker (took 1.087v just to run 950 core without crashing). I'm actually looking forward to RMAing
post #17 of 23
Quote:
Originally Posted by Booty Warrior View Post
Update #2: Yeah, it's definitely the card. Rainbow artifacts in any 3D app now. I've never had a card degrade so fast , but I guess if it was never really stable to begin with it was just a matter of time.

I'm not too upset though. The card that died was a terrible clocker (took 1.087v just to run 950 core without crashing). I'm actually looking forward to RMAing
Well Booty, sorry to hear that but I'm glad your taking it with your chin up and looking at the bright side. At least you just improved your chances at getting a better over clocker where you didn't have one before.

Let us know how the RMA process ends up. At least you have a second card to keep you going until the RMA process is complete.
     
  hide details  
Reply
     
  hide details  
Reply
post #18 of 23
So the card artifacts up constantly now and still has 0 errors in OCCT? Man that is weird.

Makes me question how accurate the test is now.

Edit:
I was looking up what your errors could be and according to this site it's a memory artifact. It could be that your memory is overclocked too high or the memory is going bad. OCCT probably doesn't really check/stress the gpu memory much. The only gpu memory test I know is VMT (Video memory stress test) although I haven't used it much.

http://www.playtool.com/pages/artifacts/artifacts.html

"Increasing your video RAM clock commonly causes artifacts. It can overheat your RAM enough to permanently damage it. The damage is cumulative and RAM seems especially vulnerable to it. I've seen plently of examples of people whose overclocks worked for a few months and then they started getting RAM artifacts. They discover that they have to reduce the RAM clock rate to make the artifacts go away. And then later they have to reduce their RAM clock more and more as their RAM deteriorates. "
Edited by gl0ry - 5/26/11 at 1:27pm
Bladestorm Z
(11 items)
 
  
CPUMotherboardGraphicsRAM
i7 3770k @ 5ghz [1.4v] MSI Z77 MPOWER MSI GeForce GTX 1070 @ +225/+450 Corsair Dominator Platinum 16GB @ 2200mhz [1.5v] 
Hard DriveCoolingOSMonitor
Crucial M4 SSD Custom Water Loop Windows 10 Professional x64 2x Debezeled QNIX QX2710 [77hz] 
PowerCase
Corsair AX860i Corsair 600T SE 
  hide details  
Reply
Bladestorm Z
(11 items)
 
  
CPUMotherboardGraphicsRAM
i7 3770k @ 5ghz [1.4v] MSI Z77 MPOWER MSI GeForce GTX 1070 @ +225/+450 Corsair Dominator Platinum 16GB @ 2200mhz [1.5v] 
Hard DriveCoolingOSMonitor
Crucial M4 SSD Custom Water Loop Windows 10 Professional x64 2x Debezeled QNIX QX2710 [77hz] 
PowerCase
Corsair AX860i Corsair 600T SE 
  hide details  
Reply
post #19 of 23
Thread Starter 
Quote:
Originally Posted by gl0ry View Post
So the card artifacts up constantly now and still has 0 errors in OCCT? Man that is weird.
No, the 2nd card is passing OCCT, the 'bad' card is now on its way to Palit's factory in Cali.


Quote:
Edit:
I was looking up what your errors could be and according to this site it's a memory artifact. It could be that your memory is overclocked too high or the memory is going bad. OCCT probably doesn't really check/stress the gpu memory much. The only gpu memory test I know is VMT (Video memory stress test) although I haven't used it much.

http://www.playtool.com/pages/artifacts/artifacts.html

"Increasing your video RAM clock commonly causes artifacts. It can overheat your RAM enough to permanently damage it. The damage is cumulative and RAM seems especially vulnerable to it. I've seen plently of examples of people whose overclocks worked for a few months and then they started getting RAM artifacts. They discover that they have to reduce the RAM clock rate to make the artifacts go away. And then later they have to reduce their RAM clock more and more as their RAM deteriorates. "
Hmm now that is interesting. My artifacts started out a lot like the CSS screenshots on that page.

The weird thing is, I only OC'd my VRAM to 4200MHz (4000 is the default). Not an extreme OC by any means... but maybe Palit cheaps out on memory quality on these double VRAM models?

Thanks for the info. Reps!
post #20 of 23
Thread Starter 
Update 3: My good card is starting to show the same symptoms as my RMA'd card. Sparse memory artifacts at stock clocks.

Its in the early stages where it's not much of a problem (I may see one flash of bad geometry in an hour or so of gaming), but if the memory is degrading like the last one, it'll only be a matter of time.

I has a sad. This card does 1GHz core on 1.06, OCCT/Crysis/Metro stable

Is downclocking the memory the only thing I can do to save this one? I reeeeally don't wanna send it back!

TL/DR moral of the story: For anyone considering the Palit 2GB 560 models, do NOT touch your memory clocks.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: NVIDIA
Overclock.net › Forums › Graphics Cards › NVIDIA › Temporary Artifacts?