Computer blank screens after hours of running

Hey I just build a new computer and I am having an issue. Everything runs awesome for 2-5hrs and then at random the screen goes to a multishaded curtain look. Typicaly the color has either been muilti shades of white or purple. Then if I let it sit for a minute the whole screen will turn to just 1 solid color. It never BSOD, just crashes to a blank screen of color. I reset the computer and it will boot properly and run again for a couple hrs b4 the same crash.

I am assuming this may be a graphic card failer? Everything in the computer is brand new, the GPU is a XFX 5770. I got the latest drivers from the site, if this has anything to do with drivers. I am still new at building PC's and need advice on what action I could take to troubleshoot my problem to figure out whats going on.

21 answers Last reply
More about computer blank screens hours running
  1. My first thought is that you have a temperature issue somewhere, but when you submit your system specs, we will be able to help you further.

    What are your specs of your new system? Please include all parts, including Case/PSU too.
  2. Ok I will list the parts

    Optic Drive - Sony Optiarc Black 24X DVD+R 8X DVD+RW

    HDD - Seagate Barracuda 7200.12 ST31000528AS 1TB

    Memory - G.SKILL Ripjaws Series 4GB

    CPU - AMD Phenom II X4 955 Black Edition ( Using stock heatsink )

    PSU - APEVIA Chameleon ATX 550W

    GPU - XFX HD-577A-ZNFC Radeon HD 5770

    Mother Board - ASUS M4A79XTD EVO

    Case - APEVIA X-CRUISER2-PK Pink ( its for the gf..)
  3. I have the temp gauge sensor hooked up to the cpu heatsink it reads right under 86F or 30C
  4. Well my gut tells me that your PSU can't handle your system. Apevia isn't what I'd call a good PSU manufacturer. The PSU you have is only rated at 25c (load) and only has ~27.5A on the combined rails (by the manufacturers website, 330w max). This makes this 550w PSU actually a 300-350w PSU. I would check out what your PSU's temperature is doing during gaming. I would bet that it's exhibiting alot of heat out of it during heavy loads.
    It's possible that the GPU has issues too, so don't rule that out. I'd just double check and make sure that your connections on your mobo and your GPU (6 pin PCI-e power connector) are all good too.
  5. Most likely your PSU, never heard of Apevia brand and it certainly is not one of the certified psu maker listed by ATI.

    It's actual power output is most likely too low to power your computer under stress, since it does not have power factor correction which is important for maintaining constant output.
  6. Apevia has been around a bit, but they make crap PSU's what what I've seen. I've built 1 system around their PSU's (before I knew what I was doing) and it worked out fine, but the system only needed about 150w to run :)
  7. Download OCCT and run the various torture tests....if you do get thru 'em, look at the graphs to see if voltages and temps are stable.
  8. Ok taking one step at atime, I changed the PSU to this one
    had it in another computer.

    Changing the PSU did not solve my problem. It still has the same problem but now it seems it is slighly worse? Right before it crashes to that curtain screen of random colors, the top 2-3 inches of my screen like seperate from the rest and shake at random times. Its very rapid and comes and goes.. after a few minutes of letting it do that off and on it will then crash to the blank screen of color.
  9. insane4cefed said:
    I have the temp gauge sensor hooked up to the cpu heatsink it reads right under 86F or 30C

    30c for your CPU? seems very low, verify that your heatsink is installed correctly. You may be overheating after a certain time passes.

    What does the CPU temp and GPU temps get to when under load?
  10. 30c can't be right...
  11. Ill see if i can find a program that shows a temp of the cpu and stuff. the case gauge could not be working properly idk.
  12. JackNaylorPE said:
    Download OCCT and run the various torture tests....if you do get thru 'em, look at the graphs to see if voltages and temps are stable.

    I started this test but it says my gpu doesnt work with it, it needs an nvidia card? also I have DX11 and it is saying it needs DX9. Do i need to install or update part of my drivers for this?

    Anyway I let it run on the CPU load for a few minutes to get an idea what this software did. Here is a few of the cpu logs. CPU 1-4 look the same so im just posting 1 and then the 3 & 5 volt charts.

    The temps on this seem to be close with the case gauge ( the case guage actually shows it running a few C cooler then the chart ).. im tryin to download asus temp utility now also to have another source of information on my comps status.
  13. If it is not the PSU, then GPU should be the next to be tested. I use FurMark to test for gpu stability. This program should work with both ATI and Nvidia cards.

    It works for 64 bit too, even though it may say 32 bit. Do a burn test and see at which temp your screen starts to get flaky. Then try monitoring your GPU temp real time as you do your regular activities and see if there are any connections there.
  14. Ok I ran that ozone program for about 1.5hrs on stability. The GPU got to 82C and held it for the entire time at that temp. The screen didnt glitch nor did the computer crash to its blank screen.

    What would be my next step in finding a solution?
    The comp has crash to the blank screen playing WoW the most, it only glitches in WoW and I seen it glitch in Spore aswell.

    The comp has also crashed playing this falling sands browser game.

    Other then that, I have had little issues with the computer.. just the ghost problem with the screen glitching the top 3 inches on WoW and Spore. Desktop and internet have not shown these glitches.
  15. Do you have another screen to test this out on?? Maybe you just have a bad LCD monitor??
  16. lunyone said:
    Do you have another screen to test this out on?? Maybe you just have a bad LCD monitor??

    Honestly before I got on here to check for replys I was thinking about it and I wondered if the monitor could have just been wiggin out. Seeing as even when it did crash to blank screen it never gave me any windows errors or BSOD.

    Maybe if that is the case, when I would reset my computer when it was "crashed" it could of just needed a signal reset for the monitor?

    Tonight I will test this out, I have another PC hooked up to a 26inch samsung lcd that has been running flawless for a month now. I will hook that up to the monitor I was letting my GF use and see if my pc has the same crash after hours of play.
  17. good, keep testing
  18. I don't think it has anything to do with the monitor. My GF demanded she get to take her comp home even tho I was still testing it out... she thought it worked well enough for her to get some use out of it lol.

    Well she had the comp for 2 days, IDK how long each day she used it but she said it wasnt doing anything that I discribed to her. Then on the 3rd or 4th day she had it at her place.. using her own monitor she got the blank screen like I was talking about to her.

    She was playing spore when it crashed to a blank screen. She said that it wasnt like the whole computer crashed because she was fighting some monsters in the game and even tho it was a purple screen all the sounds were still there like her character was still fighting.

    So I have no other thoughts on what it could be, since I have ran that ATI graphic test software previously posted. I just cant think of what else it could be besides the GPU if the visuals is the only part of the machine that is physicaly failing.

    I pulled the XFX 5770 out of her machine and replaced it with a Asus Gforce 9400GT. Erased all the drivers and refreshed them with nvidia drivers. She has had the computer for another 2 days since then and She has yet to run into that crash again.. I am thinkin it possibly was the card and I am going to attempt to RMA the part.. hopefully this thread will be proof that I tried to do as much as I could and could not get the part to function 100%
  19. Sounds like it's the GPU. Hopefully the GF's use will continue without any hitches and if so, it'll prove that the GPU was at fault.
  20. Yea, wait and see for a week. If nothing happens, then its definitely the GPU, hopefully if you do RMA it, they'll send you replacement or something real quick.
  21. They should replace it, if not easily fixable. It's a pretty new GPU, so shouldn't be too much hassle.
Ask a new question

Read More

New Build Computer Systems Product