[Jolting]

I have a 4850 GPU, e5200 CPU @ 3.9.
When playing MW2 my computer will randomly either black screen and lock up or the screen will turn into long vertical lines of different blended colors of red. Both of which require me to flip the switch on the back to turn it off. The cpu is not overheating. It stays below 45c always. The issue ONLY occurs with MW2. However I had something similar in Cod4 with the same hardware except I only had a 4670. I don't get artifacts that I notice. The game does really well then just stops. I have no idea what it is, so I'm going to go on a limb and ponder if its heat related.

Overheating is not the issue.
After extensive testing I have moved on to thinking it could be a lack of power?
Additional comments towards or against this theory are welcomed.

 

[Rewindlabs]

What are the card temps and which 4850 is it exactly? does it have vram on the rear?

Have you manually adjusted fan speed and observed temperatures like you should with ANY graphics card?

 

[Jolting]

http://www.newegg.com/Product/Produc...82E16814150337
Not sure where the memory things are at. I'll have temps at crashing point here in a few mins. I always forget to set my g15 screen to temps to see them at crash. Once it crashes again I'll post the results.
Also I didn't adjust the fan speed manually, until now. I just put it at 100% to see how it does.

 

[Benny99]

Quote:

Originally Posted by Jolting http://www.newegg.com/Product/Produc...82E16814150337 Not sure where the memory things are at. I'll have temps at crashing point here in a few mins. I always forget to set my g15 screen to temps to see them at crash. Once it crashes again I'll post the results. Also I didn't adjust the fan speed manually, until now. I just put it at 100% to see how it does.

I also have a 4850 and they run hot.

On the Stock cooler i hit 105 Degress Celcius yesterday on MW2 but it didnt crash LoL

Ive got a Zalman VF1000 coming in a few days.

For some reason these cards seem to get gradually hotter over time.

For the first 4-5 months i had it it never went of 70 on load.

Now its up to the 100 degree mark mostly 85-100

 

[Rewindlabs]

Benny have you put any thought as to why the temps are increasing? have you checked to see if the cooler is filled with dust etc?

As for you OP no your card doesn't have vram on the rear so thats good...NICE card BTW...you need to keep the temp within a safe range whilst of course not ringing your years out with 100% fan speed

 

[Benny99]

Quote:

Originally Posted by Rewindlabs Benny have you put any thought as to why the temps are increasing? have you checked to see if the cooler is filled with dust etc?

Ive put alot of thought.

And no its not filled with Dust ive cleaned it out many times and replaced the thermal Compound aswell temps stay exactly the same.

I just think the Stock cooler is a piece of crap.

Will be intresting to see how the VF1000 does.

My idle temps are the low 40s its just the Load that is terrible.

Btw in not even overclocking it all stock settings .

 

[Rewindlabs]

Quote:

Originally Posted by Benny99 Ive put alot of thought. And no its not filled with Dust ive cleaned it out many times and replaced the thermal Compound aswell temps stay exactly the same. I just think the Stock cooler is a piece of crap. Will be intresting to see how the VF1000 does. My idle temps are the low 40s its just the Load that is terrible.

Well its just i don't see the GPU getting hotter for no reason over the course of a few months..

 

[Benny99]

Quote:

Originally Posted by Rewindlabs Well its just i don't see the GPU getting hotter for no reason over the course of a few months..

I dunno but it seems that way..

Ive already gone through one 4850 and it had a similiar issue (HEAT)

Onto my Second one again issue is HEAT.

Really put off by this probably by Nvidia as my next card.

 

[Jolting]

Well I don't think its the temps. My GPU was at 60c for all 3 temp readings when I crashed. My cpu was at 37c with the cores at 60c... I'm kinda surprised it read 60c across everything. My core temps kept bouncing between 40c and 60c in very rapid quick changes. I didn't notice it do that before. Is there a possibility of software issue? I am still running Win7 Build 7077.

This also happens when my cpu is not overclocked. I'm doubtful its the cpu because I have not been getting any random crashes or anything. I was able to run every game so far on my 26in tv with a hdmi connection with 1920x1080 res and not have any issues (I havn't tried MW2 yet). That or maybe my hardware just hates the COD series? I'm willing to reformat to Win7 Professional (I get it for free from my school), if you think it may help. Also when I go to test different overclocks for the gpu it recreates the lines but it doesn't crash. I noticed it after the issue happened and I decided to give it a try and see if it just needed a boost.

 

[Rewindlabs]

What are you using to check GPU temps? the shader/core/memory should rarely all read the same

 

[Jolting]

Listed below is the temps gathered in the few mins before the crash and then idle after rebooted.
Everest
The temps were moving in the ranges listed below when it crashed.
GPU Diode (DispIO) 55-57
GPU Diode (MemIO) 59-61
GPU Diode (Shader) 59-61
CPU 35-40
Core 1 58-62
Core 2 58-62

Idle
GPU Diode (DispIO) 44
GPU Diode (MemIO) 48-50
GPU Diode (Shader) 45

CPU 19
Core 1 42
Core 2 43

 

[Jolting]

Could not enough power be the issue? I only have a 550W PSU. If I lower the settings it does take longer for it to crash. So maybe just if I play for a long time and have alot of things open then the power usage becomes to much and the card dies out?

 

[JMT668]

you shouldnt have issues with it over heating the XFX cooler is sweet.

Maybe try cleaning it all and re applying thermal paste?

I have the same card but its locked to 4830 spec as i couldnt afford the 4850 but with a reseat and a good old clean i got it to 4850 clocks.

 

[Jolting]

After a bit of testing I realized it is not overheating and have moved on to trying to figure out the real problem. I'm guessing I just don't have enough power. Also I have completely reformatted to see if maybe I just had some bad drivers/software doing it. The first thing I did was reinstall the game and after 3 hours the game crashed the same way as before.

 

[Rewindlabs]

Quote:

Originally Posted by Jolting After a bit of testing I realized it is not overheating and have moved on to trying to figure out the real problem. I'm guessing I just don't have enough power. Also I have completely reformatted to see if maybe I just had some bad drivers/software doing it. The first thing I did was reinstall the game and after 3 hours the game crashed the same way as before.

Well what are the specs on your PSU? it says it has enough watts but if your GPU isn't getting the amps it needs then there is your problem...

As you can see 550W's is pushing my rig just fine but i do have a ridiculous amount of amps

Your rig shouldn't even get close to consuming a full 550W's so it could be a PSU fault or lack of amps

I know when i ordered my 550VX from newegg that i got a bad one and it would allow me to game from anywhere to an hour under load to no time on my 260GTX....was really weird and took me a day or two to diagnose it was the PSU....got the one in my sigrig sent to me and its been running solid ever since

 

[G.E.Nauticus]

You should all trade in your 4850 for one of these 4850's. The cooler on this thing is nuts with 1 fault, it is stuck at 100% fan speed

. I personally own this model and have it clocked 24/7 @ 730/1050 and cannot get the thing over 60c..... fur mark extreme burning is like 57c max..... this is in an open case with 80F ambients.

 

[Rewindlabs]

Quote:

Originally Posted by G.E.Nauticus You should all trade in your 4850 for one of these 4850's. The cooler on this thing is nuts with 1 fault, it is stuck at 100% fan speed . I personally own this model and have it clocked 24/7 @ 730/1050 and cannot get the thing over 60c..... fur mark extreme burning is like 57c max..... this is in an open case with 80F ambients.

Whats the rest of your temps though? does 57c max include core/shader/memory?

 

[G.E.Nauticus]

Quote:

Originally Posted by Rewindlabs Whats the rest of your temps though? does 57c max include core/shader/memory?

Not 100% positive on the other temps, im getting my 57c from furmark temps, ill test it when i get home tonight. What do I use for telling other core temps?

 

[SporkofdooM]

I used my 4850 with a no name psu that was "450 watts" and never crashed due to power issues. So your Antec (a good quality PSU) not supplying enough power probably wouldn't be it.

My guess would be either a driver conflict of some sort, or possibly re-install the game.

 

[Jolting]

Unless the drivers from ati.amd.com are faulty then its not a driver issue. I even reformatted and installed everything from scratch.

Here is the PSU
http://www.newegg.com/Product/Produc...82E16817371020

What is a list of things that could cause a line pattern with MANY different shades of the same base color, majority of time its red others its blue. A few Black screens for long periods that go away. The video card was bought in the summer just before Aug. The cpu has been pushed to pretty high limits from overclocking has a slight glitch with the multiplier. The board was bought in the summer just before the video card. The PSU was bought from someone on here who had just got it back from a return.