15 replies to this topic

[Sunnyalex]

Hey there!
Since a couple of days (maybe since patch of April 1st, 2014) I got some graphics issues:
In game it shows green and purple artefacts (see picture).

I tried running the repair tool -> no change.
I update graphics driver (now it is latest 9.18.13.3523) -> no change.

My graphics card is GeForce GTX260. Win7 / 64 bit OS.

Sometimes framerate drops to < 10 fps (usually i got 40 - 80 fps) and then after a couple of seconds I get a "monitor is going to sleep" (black screen) and shortly after monitor restarts, windows prompts "Graphics driver were restarted after an error" and then the game is fine again...

Sometimes in game resolution also changes to maybe 640 x 480 (until again graphics drivers crash and are being restarted).

This happens since a couple of days now, I've been playing > 1 year without any problems...

What could that be?
What can I do?

Best regards,
Alex

[Shamous13]

sounds like a thermal problem to me. clean your computer with compressed air. then download msi after burner and check your video card temps, and check your cpu and mb temps too. it also could be your psu going south too but statrt with the cleaning.

[Werewolf486 ScorpS]

Blow it out with an air compressor on less then 10 psi. Pay for Compressed air, what a racket! I'd say that you need to buy a new Titan Black and be done with it.

[xWiredx]

Could be heat-related like others mentioned. Time for a clean-up (and possibly replacing the thermal compound on your GPU). Worst case scenario is that your ~6 year old card is finally on it's way out.

[9erRed]

Greetings,

Are those coloured blotch's, leftover foot prints from Mech's?
The dissolving foot prints we have in-game are only temporary and the engine is designed to remove them and update the ground texture. These look like the terrain surface is not being cleared and the buffer refreshed. Does this happen on all maps or only the snow locations?

As a note here, your Vid card is at the low end of being able to actually play this game.
- And when the error occurs that you listed it is the vid card shutting down and rebooting to protect itself from overheat.
The requirements for your card are rather extreme, compared to even a 400 or 500 series card, which require much less power and run at lower temperatures. And that 200 series really does not have enough or adequate memory on board.
- Just from a quick check a Zotac GeForce GT630 SE (ZT-60413-10L) with 4 Gb DDR3 lists for $89.99 Cdn at "TigerDirect", and there are quite a few 500 series out there for under $50.

Thermal and Power Specs GTX 260

- Max GPU Temp (in C) = 105C
- Max Graphics Card Power (W) = 182W [this is crazy as newer cards run on around 45W]
- Max System Power Requirement (W) = 500W

- And yes, if you have never changed the "thermal past" on that card it's time is up. but going to the trouble of changing it, it may just be easer to upgrade.

As a note here are the latest WHQL drivers for your card from Nvidia. (as of 10Mar2014)
http://www.geforce.c...s/results/73780

9erRed

Edited by 9erRed, 09 April 2014 - 05:06 PM.

[Sunnyalex]

Dear all!
Thank you a lot for the replies.
I will clean whole PC and especially graphics card. I will also renew (if possible) thermal compound.
Thank you for the advices!
I will keep you informed.
And no: These are not only footsteps and it does not only happen on snow maps (I've seen it on canyon network, too).
Greetz!
Alex

[Sunnyalex]

Here are two more screenshots (including GPU temp):
1st is directly after drop. Temp is 62 degC, however I've got severe graphics issues as you can see.

2nd is after death: Temp is 80 degC, GPU load is 97 %. However, no issues...


I still need to clean PC/graphics card, however, I do not see a direct correlation of temperature / errors...
Anybody an idea?
Best regards,
Alex

[ThatBum42]

I'm thinking it's blown VRAM or something like that, unfortunately. I've seen similar purple and green spots from dying cards or cards clocked too high.

Is the card overclocked? Does this happen in other games?

Edited by ThatBum42, 10 April 2014 - 03:04 AM.

[Sunnyalex]

ThatBum42, on 10 April 2014 - 03:02 AM, said:

I'm thinking it's blown VRAM or something like that, unfortunately. I've seen similar purple and green spots from dying cards or cards clocked too high.

Is the card overclocked? Does this happen in other games?

Card is not overclocked. Actually I have not played any other games for years ;-)

So maybe you are right and it is time for a new card... :-(

[ThatBum42]

Sunnyalex, on 10 April 2014 - 11:47 PM, said:

Card is not overclocked. Actually I have not played any other games for years ;-)

So maybe you are right and it is time for a new card... :-(

I suggest running FurMark to stress the GPU, it will force the problem to the surface assuming it's a hardware issue.

For a new card, I recommend the higher end Nvidia 600 series cards, they're pretty cheap for what they can do. The 660 Ti or 670 can max out just about everything at 1920x1080 if there's no bottlenecks. Specifically the ASUS versions, as they have a wonderfully designed cooler.

[Sunnyalex]

Hi there!
So actually FurMark "Burn In Benchmark" passed with max. temp. < 90 degC.
However: FurMark showed it was only using 15% of GPU RAM, so I downloaded another tool called "video memory stress test" (VMT). This tool revealed that approx. 64 MB of my graphics RAM (what would be 1/16th of total memory) is corrupt and cannot be set to "1". So I suppose one the RAM ICs on the graphics card is broken (I assume that there is a total of 16 RAM ICs on the card).
...
Too bad... actually I do not want to spend money at the moment...
Guess I need to keep playing like this...

Thank you all for your helpful hints and support!
Alex

[ThatBum42]

Yup, I was correct in my original assessment. Too bad

Might I suggest a GTX 760 or GTX 660 Ti for your next card, if you so choose to upgrade or the 260 dies completely.

The 700 series is basically a re-badge of the 600 series with some minor engineering tweaks. They're both based on the Kepler architecture, so consider that if you find a 600 series card that's significantly cheaper than its 700 series counterpart.

Try GPUBoss for comparing cards.

[Sunnyalex]

Here in germany a Asus GTX 660 is about EUR 150,- (US-$ 200,-) and the GTX 660 Ti is EUR 200,- (US-$ 275,-).
However, so far I always used MSI components (Mainboard, Graphicscard) and they always worked well. But they are approx. 1.5x more expensive then Asus...
In any case, I think I'll continue playing even with the graphics issues until the card completely dies...

[Sug]

It's been said but yeah, card's dying. I've seen those artifacts many times on nvidia cards. The cards just can't handle long gaming sessions at the temps they run at.

[ball0fire]

you could try using afterburner to lower the vram clocks see if that helps, 30 or 50mhz downclock might help stabilise it a bit
(tho prolly not)

should also increase gpu clock speeds a bit to help recover any lost performance (now is a good time to try overclocking the gpu since its already half dead)

[Neil Diamond]

Your video card is dying and more than likely the VRAM is doing the same. I would suggest as have others, looking into the 660ti. It is a good bang for the buck card.