Sign in with
Sign up | Sign in
Your question

System Freeze/Audio Stutter HD4850 *update* Multicolored Bar Crashes

Last response: in Graphics & Displays
Share
December 27, 2008 4:38:40 AM

Hey all,

This might not even be a video card issue, but I figured this was as good a place as any to post this.

My Specs:
Corsair 650 PSU
2 GB Patriot
e8400
P5Q-E mobo
4850 gpu

Vista Ultimate 32-bit


In every 3D application I've tried to run seems to lock the system up requiring a reboot.

If what I've been running has audio, the audio will continue to play stuttering the same 2 seconds or so over and over.

Its a hard lock, CAD won't work.

Initially, certain sections in the Witcher, would cause a crash, sections that, once identified, I was able to avoid to keep playing the game.

Then running 3Dmark06 would crash after a couple of minutes.

Now the latest crash happens in Crysis 1-5 minutes into the game.

Those are the only programs I've tried to use so far.

Memory is fine, memtest passed.
CPU is fine, but runs a little hot.
GPU is not overclocked at all.
PSU gives more than enough juice.

I've seen this problem happen to tons of people using my google-fu, but no solutions, and they might not even be having the same issue.

Using latest drivers for everything (already updated them all trying to solve the issue myself)

I've read that a voltage did in the 3v rail could be an issue, but the Corsair should be fine.

Prime95 has been run for 8 hours stable.

I've tried running tons of different videos and music and such at the same time, in an attempt to recreate a crash, but I'm at a loss. If Prime95 doesn't crash it, I just don't see how a game could unless it was the video card's fault.

Anyway, suggestions would be good. There are no artifacts to precede the crash.

Thanks.
----

After doing a fresh cleaning/install of the drivers and getting rid of all the third party software stuff, using only the catalyst control center at stock speeds and forcing the fan to stay on at 80%, instead of locking up with a black screen, it now likes to crash with vertical multicolored bars.

Running Crysis in windowed mode so I could monitor CPU and GPU temps/fan speeds, both temps topped at 58 cpu/71 gpu and the game lasts about 30 mins to an hour before crashing occurs. There are only two other things I can think of to try before outright saying this card is a piece of ****. Perhaps the voltage is too low on the 3v rail (not likely considering the corsair psu), or maybe the RAM has errors which I will discover after running memtest tonight when I go to sleep.

I have also been using Furmark on the burn in setting to try to get the temps up on the card at a full load (did the renaming of the exe of course), but it still tops at about 85.
December 27, 2008 12:23:02 PM

Blue Screen + stop code 0x116 directing to the atikmdag.sys error?
December 27, 2008 12:37:11 PM

uninstall any and all video driver programs in add/remove
uninstall driver in device manager
restart and boot into safe mode, run a few iterations of drive sweeper to clean out the junk
restart and install latest drivers
test.
Related resources
December 27, 2008 6:22:18 PM

No blue screen.

Just a plain ole' black screen lock up.

Maybe I should note that I have the various software from ASUS installed and running, the 5 system speed management, smart doctor, etc. Perhaps that combined with the catalyst control panel is creating a conflict?

Ex. in the catalyst control panel, the highest you can set the mem clocks is 1200 even though the stock is 1986. Smart doctor, however, shows that the card is set at the correct 625/1986 speeds.
December 28, 2008 4:18:56 AM

Followed montyuk's advice and am having crashes still, only now it looks to be signs of corrupt vram :( 
February 19, 2009 10:17:15 PM

Someone have same prob and solved it pls??? I have one and I think it can be something about VMR overheating. Iam using accelero S.1 with no fan on my GB HD4850 512MB card (I like quiet comp) and temps of GPU are around 85°C max. (with stock cooler were 95°C and more sometimes but system was stable!!!!). Crashes started when a small memory and VMR passive coolers start falling from card but this I solved yet and some games falling still. NOT ONLY LATES graphic bombs!! Old Far Cry too Btw...
I try scan temps with GPU-Z ver.0.3.2 and there is some value named "GPU Temp.(MEMIO)" which is around 15°C higher than GPU temp. My system go black screen and sound shutter when this one reach 90+ °C... BTW when it crash, red light go on on top of card and I found on i-net that this one named D1601 pointing to overheating! But GPU on 80-85°C isnt "overheated" IMHO :) 
Point: If I look at stock cooler the fan is situated near VMR, so I sugest that temeratures should be lower in this part of card with stock then with only small passives without fan.... Tomorrow Iam going to switch to stock and make some tests. I hope it will work cause Iam in it about 3 weeks omg :(  Hope my sugestions going to be well and help me and someone!!! :)  Write pls if know more about this.

I think it doesnt matter but my config is:
C2D 8400 + freezer7
2x2GB A-Data DDR2 800
GigaByte HD4850 512MB + AC S.1 no fan
ASUS P5Q-SE/R
ENERGMAX Liberty 500W

DOES MATTER: Have 2 almost same computers with SAME PROBLEM (2nd with PSU Bloue Storm 500W and ASUS P5E-SE)

Both passed tests like: memtest86, video card stab. test, OCCTPT... didnt pass some games and "FurMark (but its maybe the point cause many watercooled cards didnt pass with temps around 40°C as I found on net.. But It was temp of GPU not VMR..."

P.S. sorii fooor maaaj poooooor ingliiish ;) 
February 19, 2009 10:20:31 PM

Oh I forgot: new drivers, latest chipset drivers etc. nothing didnt work for me. I almost Flash card Bios but its pointless IMO....
March 15, 2009 3:10:06 AM

I've been having this exact problem! I've been totally at a loss. Started with Empire: Total War, and I thought that was the cause since it's a new and undeniably buggy game, but it has since happened in Medieval 2: Total War, Rome: Total War, and the World in Conflict Demo. All updated drivers, flashed a new BIOS, passes all testing programs I've run, but lock-ups in the games continue.

Specs:
Core 2 Duo E8200 OC'd @ 3.2Ghz
HD 4850 running catalyst 9.2
EVGA 750i FTW
2x1GB RAM
Vista Ultimate 32-bit
March 24, 2009 5:56:46 PM

I could write the same as my preposter - I've got exacly same issue and did everything I could to solve the problem but no luck. I even reinstalled systems (both Vista 64 and XP 32).
March 24, 2009 6:42:58 PM

What are your PSU specs? It all sounds like a power supply issue to me...
Anonymous
April 4, 2009 6:27:17 PM

Im having this problem also, has anyone fixed this yet? last remnant (only crashes on certain maps) and f.e.a.r 2 (crashes at random) are not working for me. but i managed to play wanted from start to finish with no problems, though it is a short game.
April 6, 2009 1:19:55 PM

Same Problem different card Spec as below:

Sapphire PE-AM2RS740G Mobo
AMD 7750 (2.7)
Gainward Golden Sample 4870 1GB
Kingston CL8500 (5-5-5-15) (4GB)
Antec HE Neopower 550W PSU (3 off 12v rails @ 18W, reviews state PSU capable of 650W @ 83% efficiency)
Samsung Spinpoint F1 (640GB)
Vista Ultimate 64 (SP1)

World of Warcraft will freeze, will sit for a while before Vista restarts the machine. Vista Blames ATi Display driver.

Tried 9.3's, hard lock after 15 minutes, 9.2's lasted an hour or so. Will try 9.1 and 8.12 later.

Temps on GPU around 50 C on load, CPU around 52 C.

I've got an ASROCK 790 Board arriving tomorrow if I don't get a resolution the AMD/Ati stuff is getting RMA'd and i'm buying Nvidia and Intel!

!