XFX Radeon HD6770 - random freezes / display problems / crashes

[ConcreteDonkey]

Hello,
So recently i bought a new PC and im having problems with video while playing games. Everything happens randomly:

- most often monitor display is turning off with audio still being played properly (got to reboot pc)
- second to that freezes happen (got to reboot pc)
- rarely "Display driver has stopped working" error pops up (no need to reboot pc)
- freezes for few seconds and then everything is back to normal
- few times little window popped up on the screen saying "Signal frequency out of range. Please change signal timing. etc"

I noticed that i have more chances of experiencing 1 of the above just after reboot from previous crash and starting a game.

Okay, now to the part what i have done sofar.

Already tried a dozen of different Catalyst drivers, previously removing old ones with Driver Sweeper. Overdrive is turned off and i didnt touch any clock. Default setup.

I tried PSU and GPU on another PC for 3-4 hours - no problems, but it doesnt say much anyway. I can play single game for few hours on my PC without any problem as well, then few in a row may happen.

Some of the problems directly imply that refreshing rate can be wrong and GPU is overheating. Note that i have CRT screen with DVI-VGA adapter and refreshing rate is set to 85Hz (already tried multiple ones in the range of 50-115) with resolution set to 1024-768 ingame/desktop, nothing helps. About overheating - i ran the Furmark test as suggested in another thread, it was running for roughly 14 min and temp got stable at 86-87 C - no artifacts. Monitor is just fine (worked like a charm on previous PC) and tried 2 different DVI-VGA adapters.

I ran a Memtest from boot level twice for whole night, no errors.

Re-attached the GPU into the slot to be sure its settled properly together with power cable.

Voltages checked in AIDA64:
- while on desktop - https://dl.dropbox.com/u/43525724/czujnik.png (sorry for no-english here)
- while playing - no change at all
(and im aware of software based sensors not being too accurate, its just all i can do for now and im unlikely to get the proper one)

Latest BSOD log (i had just a couple, this one clearly happened ingame):
privatepaste.com :: Paste ID 36316f055a - "Nie można odnaleźć określonego pliku." means "Couldn't locate "this" file"

By the way, there is so many topics like this in the web, especialy for 6770HD. I found 1 topic where someone had exact same problem on some Geforce GPU and fixed it by imroving voltage by just a little. Not that it has to have anything to do with my problem, but worth a mention i guess.

----------
So, those are my PC specs:

CPU - Phenom II X4 Quad Core 965 Black Edition 3.4GHz
RAM - DDR3 4096MB PC1333 CL9
GPU - Radeon HD6770 1GB DDR5 128BIT 2xDVI+HDMI BOX
HDD - Caviar Blue, 3.5'', 250GB, SATA/600, 7200RPM
MOBO - ASRock 970 PRO3
PSU - OCZ-CXS500W-EU
CRT Display - NEC FE791SB with DVI-VGA adapter
System - Windows 7 Ultimate x64
----------

Hopefully theres solution other than "give it back and ask for refund". Thanx in advance guys and greetings.

 

[Indianatone]

Hi and welcome.
I have a Gigae Byte silent HD 6770 and my temps are 47.5C, 47.5C and 51C at idle. Similar to yours. Download Download TechPowerUp GPU-Z v0.6.5 | techPowerUp and run it whilst gaming and set the options to log to a file and show maximum temperatures. This will show you if it is overheating. Do you have plenty of cooling in your case? I suspect you need to keep the VGA temps card below 80C maximum. Mine don't really go much over 77 to 77C.
Hope this helps Tony.

 

[ConcreteDonkey]

Thanks for reply!

Do you have plenty of cooling in your case?

No, only default cooling as for GPU/CPU/PSU on SilentiumPC Brutus case.


Log: https://dl.dropbox.com/u/43525724/GPU-Z%20Sensor%20Log.txt

2012-10-16 12:20:21 - somewhere there freeze happened, had to reboot
2012-10-16 12:45:11 - Signal frequency out of range popped up, had to reboot
2012-10-16 12:50:17 - blackscreen with music being played for few seconds followed by freeze, reboot

Note that times i gave are not most accurate, problems happened somwhere within 30 seconds.

 

[Indianatone]

OK As far as I can tell your temps on the card are fine, It is not getting anywhere near overheating. Just for giggles, does it make any difference if you run the machine with the side off of the case?
Your CPU temps look fine at 38C.
The out of range bothers me. I know you are using a good old NEC CRT monitor for awesome "pictures" at all screen resolutions (oh I miss that) and that is the only reason you can see that message. Honestly I suspect you have a bad video card.
If it was me I would contact the supplier and RA the card under warranty. It is obviously possible to get a faulty card that does weird things at normal temperatures.