|
My system: OS: Windows 7 64-bit (latest updates) CPU: i5-2500k GPU: GTX 560 Ti PSU: 800w Apower RAM: 4gb 1600 (running at 1333) DDR3 Corsair brand Mobo: gigabyte ga p67a ud3 b3
Tests / Benchmarks I've Run:
- Furmark: Ran through a half-hour test (as recommended) flawlessly. - Prime95: Ran through a four-hour test flawlessly. - Memtest86: Ran through four passes without issue or error. - CrystalDiskInfo screenshot: + Show Spoiler + (the badass Blastoise is absolutely necessary.) - HDTune benchmark screenshot: + Show Spoiler +
Images:
+ Show Spoiler +
Hey guys, got a problem I hope you can help me with.
Twice in the last two days my computer has frozen, which is something I've been dealing with for a while (this older thread I wrote goes more into that and more in detail to other, related issues I've been having http://www.teamliquid.net/forum/viewmessage.php?topic_id=324998), but these last few times it has been a full-stop crash.
First the screen and keyboard/mouse stop responding, with the screen's pixels flashing blue and red green in almost a checkerboard pattern (images above). Then the sound skips, stopping a moment later. After that it will sit as long as I let it, unresponsive in every way.
What is this kind of event typical of? I've run tests on all the important components of my computer and the tests all came back fine, so I'm really at a loss as to the cause. Could it be a software issue, or could there be an issue with my hardware the tests wouldn't pick up?
|
For some reason the title didn't save properly. It was supposed to be "Computer Freeze, Monitor Pixels flashing RGB / Frozen in Pattern"
|
Can anyone give me some insight to this strange type of computer freeze / crash? I feel like my system is getting worse and worse over time.
|
Used to happen to me too. It's overheating bro... 1. Either you are playing too hard and/or overlocking too much? 2. You did not install your graphics card in PROPERLY. You probably jammed it in and it isn't in the slot fully and maybe as a result the fan is not spinning properly? (That happened to me).
Open your PC up and take out and place back in your Graphics Card. Make sure the fan is spinning. Make sure you don't have any software that overclocks.
|
With those visual effects you've got to lean towards a GPU fault. Did you ever test with that other card? Furmark doesn't test much except heat/power.
A seating (slot connection) fault is possible, but I doubt it's overheating if it survives Furmark.
|
On April 25 2012 22:51 jaj22 wrote: With those visual effects you've got to lean towards a GPU fault. Did you ever test with that other card? Furmark doesn't test much except heat/power.
A seating (slot connection) fault is possible, but I doubt it's overheating if it survives Furmark.
Finding somebody willing to trade cards for testing has been real difficult. I haven't been able to yet, and the local comp shop said they'd basically do a stress test like i've done and no more. :/
The card we thought we could test with turned out to be lost somewhere.
|
Possible GPU errors. PSU is most likely trash; shitty voltage regulation could also be causing the problem. Replace it with a quality 500w unit before you risk frying a card that isn't yours.
|
On April 26 2012 05:52 SoulWager wrote: Possible GPU errors. PSU is most likely trash; shitty voltage regulation could also be causing the problem. Replace it with a quality 500w unit before you risk frying a card that isn't yours.
What he is saying!
It saddens me to see so many people choose for el-cheapo power supply's with high end hardware. The PSU is one of the most important components of your system, it supplies the life line after all!
|
On April 26 2012 19:45 theincrediblemachine wrote:Show nested quote +On April 26 2012 05:52 SoulWager wrote: Possible GPU errors. PSU is most likely trash; shitty voltage regulation could also be causing the problem. Replace it with a quality 500w unit before you risk frying a card that isn't yours.
What he is saying! It saddens me to see so many people choose for el-cheapo power supply's with high end hardware. The PSU is one of the most important components of your system, it supplies the life line after all!
It saddens me that most of us are broke, haha. Ill be taking it back soon then, its not over a month old yet.
|
5930 Posts
If you can't afford something good don't buy it. Its that simple. What were you going to do if that PSU fried your GPU and CPU?
|
|
On April 27 2012 11:32 Killcycle wrote:Show nested quote +On April 26 2012 19:45 theincrediblemachine wrote:On April 26 2012 05:52 SoulWager wrote: Possible GPU errors. PSU is most likely trash; shitty voltage regulation could also be causing the problem. Replace it with a quality 500w unit before you risk frying a card that isn't yours.
What he is saying! It saddens me to see so many people choose for el-cheapo power supply's with high end hardware. The PSU is one of the most important components of your system, it supplies the life line after all! It saddens me that most of us are broke, haha. Ill be taking it back soon then, its not over a month old yet.
I know how it is you should totally bring it back if that is still possible.
What you might want to do, is check how much watt your system actually needs. I have a i7-2600K with 16GB internal and a 8800GTX (old, but uses a LOT of power) and my system runs super smooth on a zalman 660watt (including HDD's and proper cooling) with room to spare for two new video cards in SLI or Crossfire.
Try this site to calculate what you actually need (bear in mind, this calculator includes a safety margin, the result in W is always a little bit higher then what you actually need): http://extreme.outervision.com/PSUEngine
I took the liberty of filling one in according to your specs, and added some USB devices, fans and HDD's. The result that came in was 541W, so 600 should be more than enough.
I would advise to purchase one of the following brands: Corsair, Cooler-Master, Zalman. There are more, but Corsair and CM are very stable and relatively cheap. What you should look for is a efficiency certificate of 80 or higher. This indicates how stable/consistent it supplies its power.
Hope this helps!
EDIT:
Why are you advising a power supply that isn't even powerful enough? Antec is an OK brand and their PSU's are generally stable, but always stay away from budget models! If you want to save on your machine, save money on exteriors, in any case, buying a budget PSU for a High-End system is never a good idea, as this will probably cost you more in the end.
|
On April 27 2012 19:09 theincrediblemachine wrote:Show nested quote +On April 27 2012 11:32 Killcycle wrote:On April 26 2012 19:45 theincrediblemachine wrote:On April 26 2012 05:52 SoulWager wrote: Possible GPU errors. PSU is most likely trash; shitty voltage regulation could also be causing the problem. Replace it with a quality 500w unit before you risk frying a card that isn't yours.
What he is saying! It saddens me to see so many people choose for el-cheapo power supply's with high end hardware. The PSU is one of the most important components of your system, it supplies the life line after all! It saddens me that most of us are broke, haha. Ill be taking it back soon then, its not over a month old yet. EDIT: Why are you advising a power supply that isn't even powerful enough? Antec is an OK brand and their PSU's are generally stable, but always stay away from budget models! If you want to save on your machine, save money on exteriors, in any case, buying a budget PSU for a High-End system is never a good idea, as this will probably cost you more in the end.
are you being serious? It should be adequate even if he overclocks, As an example this is + Show Spoiler + 351 w from the wall(!) with Intel Core i7 975 clocked to 3,8ghz which is more than what his cpu will ever draw. Around 300w would be a normal max load for him.
|
Clearly theincrediblemachine doesn't know what he's talking by recommending Zalman and Coolermaster and talking about how 80PLUS = stable LOL
Let's not forget that a 2500k and a GTX 560 Ti is in no way even close to considered high-end.
|
It does seem like GPU-related, but it might be just a driver issue, not actual hardware defect. Especially considering how you passed Furmark stress-testing. Try other GPU-stressing tools, maybe 3dmark, etc. Does it happen only when playing certain games? Only for D3? If so then I'd relax and write it off as a driver bug.
Does the windows event log list the error? (rightclick computer, manage, navigate to events/system in the treeview, then browse the list on the right for errors around the time you had a crash)
|
It may be many things like they said, but I think the most important thing you have to check on is your cooling.
|
As far as my monitoring programs are concerned (coretemp for cpu and asus smartdoctor (as well as a side bar app "GPU Meter")), my hardware is never that warm, high 60's / low 70's seems to be the max I get to on average. Furmark nevers gets past 90C. If for some reasom the cards temp can be incorrectly discerned by the programs because of a poor seating...? Otherwise my stuff shouldn't be overheating. I made sure it was plugged in properly, so I don't think that's an issue. Then again I suppose it's impossible for me to be seeing my temp when it freezes in-game or where my monitors aren't visible, but the fan didn't seem to be working harder than normal when it went.
Ill definitely be returning the psu, probably today, and replace it with one of the mentioned brands, for the mentioned reasons. As far as drivers, I reinstalled all my graphics drivers a few weeks ago with driver sweeper, maybe I did it incorrectly?
Event Log just shows that it had an unexpected restart, whenever this happens.
The D3 failure was the one I happened to take pictures of, but it happened two other times as well, when just browsing, doing nothing intensive.
I'll try a few other GPU stress tests, and report back with results.
And thank you all for your responses. + Show Spoiler +
|
|
|
|