• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

GTX 570 artifcating and drivers crashing constantly

R

royrules22

Junior Member
Hi all,

I apologize if this is not the right place to post.

I have a GTX 570 that I purchased around summer of 2011. It had been working great up until earlier this year when I noticed some weird issues. Generally speaking my nVidia drivers crash quite often (and occasionally it will bring down my network driver as well). Usually when using Firefox (especially if there's something Flash going on).

But it also represents itself via artifacts in games, most notably with The Walking Dead.

Oddly enough I can go many weeks without seeing an issue (usually enough time to finish one game or so), before it resurfaces.

I've done the following since I noticed this:

- Bought a new PSU (SEASONIC X750 GOLD 750W)
- Bought a new case (Corsair 600T - upgrading from an Antec 900)
- Cleaned and installed new drivers (many many times - the latest being last Thursday to 306.23
- Opened the case and pointed a fan at it
- Waited for it to hit winter

Still the crashes continue and the artifact continues. I don't have anything OC'd and memtest and all diagnostics pass fine.

The error message from the crashes say:

"Display driver nVIDIA windows kernel mode driver version 306.23 stopped responding"

I tried searching for it and I applied all suggested fixes but to no avail. Unfortunately the link to nVidia's forums is gone, so I don't know what was suggested there.

Anyway here are some screenshots from GPU-Z (it recorded while I played some Walking Dead):

Fnuny.gif


Sensors showing average:
DPCmc.gif


Showing max:
fT4cF.gif


And I just noticed this weird little thing in Firefox right now (it fixed itself after the driver crashed again and no it's not a user style):
JzDlI.png


So does anyone know what the issue is and how I can finally solve this? My warranty is long gone and I really don't want to buy a new GPU (because at this point it's my old LGA 755 mobo and Q6600 I should probably replace first).
 
I believe this is a long known bug.If possible raise a ticket to Nvidia or you can pm keys with your issue, he can probably help you regarding this.This is his forum name "Keysplayr"
 
Ok will do. I'll PM Keysplayr. Thanks!

Edit: Unfortunately I cannot send him a PM until I have 10 posts, and I don't want to spam the forum. Hopefully he'll see this thread soon.
 
Last edited:
Welcome to the forums, royrules22. Jaydip PM'd me about your issue.

My very first question. Did you buy this 570 new or used. I ask because you mentioned warranty was up and it's been only 1 year since your purchase in the summer of 2011.

I've also looked up your PSU and a significant number of the reviews say the PSU died soon after installation. Not a good sign.
http://www.newegg.com/Product/Produc...82E16817151087

Could you provide your original PSU's make and model?

Make and model of your motherboard. Do you have more than one PCI-e slot? Is it an SLI board?

Next, do you have another PC? A family member or friend who has a PC you can try the GTX570 in?

Ok lets start with these.

Keys
 
Last edited:
The easiest and one of the most effective solutions to non-uniform graphical problems is to reseat the graphics card. Pull it out of the computer, blow out the PCIe slot to make sure there is not chunks of debris or fur in it (if you have pets), and firmly (but not too firmly) re-seat the graphics card into the PCIe slot. I have had 3 separate problems in the past, with different motherboards and graphics cards that have been corrected by doing exactly this.
 
Thanks everyone!

I bought the 570 new on May 18, 2011. I believe the warranty was only for 30 or 60 days because I forgot to register.

My PSU seems to be fine for everything else. In fact I bought this PSU because I was having this issue. How can I find out for sure?

My original PSU (before this one) was a Seasonic SS-550HT, purchased around summer of 2007. The current one is the exact same one listed in your link.

My motherboard is old. It's an Asus P5K-E/WIFI-AP that was purchased also in summer of 2007. It technically has two PCI-E x16 but only one is really accessible for me. It's a crossfire board and I only have one GPU.

I unfortunately do not have another PC and no one I know has one that would be willing to let me try this in. Also the issue is somewhat sporadic and it's hard to repro.

Tviceman:

- I did reseat the GPU. Once when moving it over to a new case and once because I had the same thought as you. It helped for a month or so. I have gone through multiple cans of compressed air on the system to clean it out (I don't have pets so it's a pretty clean place).

Thanks everyone!
 
royrules22 said:
Tviceman:

- I did reseat the GPU. Once when moving it over to a new case and once because I had the same thought as you. It helped for a month or so. I have gone through multiple cans of compressed air on the system to clean it out (I don't have pets so it's a pretty clean place).

Thanks everyone!
Click to expand...

Above in bold. This is a clue, or a red flag if you will. Reseating the card helped for a month or so. You say you have a second PCI-e slot on your motherboard. Why isn't it accessible? Can it be made accessible for the purposes of troubleshooting, especially since you've no access to any other computers?

Keys
 
royrules22 said:
- I did reseat the GPU. Once when moving it over to a new case and once because I had the same thought as you. It helped for a month or so. I have gone through multiple cans of compressed air on the system to clean it out (I don't have pets so it's a pretty clean place).

Thanks everyone!
Click to expand...

Reseat the video card again then. It sounds to me like it's slipping out of it's slot because either the slot is slightly defective (the contact pins have gotten too loose), the card's insertion fin is ever so slightly too thin, or it's a damn heavy card and gravity is getting the best of your situation. Make sure to push it firmly in and also screw in the output plate very tight.

Again, I have had this happen several times to me with several different cards with several different motherboards, including my current card and current motherboard. With my current card, the problem first manifested itself as detection issues (detecting it as a card that it isn't), then windows couldn't detect the card at all. I reseated it, and it fixed the problem for about one day, and then I reseated it AGAIN but making sure I screwed it really tight to prevent it (as much as possible) from slipping. That was about two months ago and it has been working fine since.
 
Its artifacts possibly due to heat. Make sure your fan is on manual and 100 percent when you play game. Then tell us if there is artifacts. Also turn off AA while we perform this.

Card is fine its VRAM getting hot. quick fix,, 100 percent fan like I said and open the side of your case and make sure your room is 74F or below. gl
 
What are your load temps? Have you took the card out and checked the fan/ fins for dust buildup? Please post your load temps. I just had all the same symptoms on my 670 and ever since I've cleaned off the old paste and applied new paste I've been running great. The stock paste was a done crappy and dried out. 46 degrees at near zero load is a bit high to start with.
 
Thanks everyone. I posted GPU-Z numbers in the OP with load temps (avg 45.7 and max of 50C).

I guess my mobo could be at fault. It is nearly 5 1/2 years old. I'll try reseating it tonight.
 
It doesnt look like your GPU is at any load. Run EVGA OC Scanner (stress test) or something similar and retake your loaded temps pics. I betcha your hitting max temp with those idle temps of mid 40's. Are you stopping the stress test then taking the GPUZ Pics?
 
Hey all I just wanted to update this thread.

I couldn't get my GPU to stay in the second slot and SATA cables blocked the bulky thing. But I did remove it, clean it up (as well as the rest of the system) and reseat it.

So far so good. I even played enough of the Walking Dead to be pissed off.
 
Yes Its the TDR issue (if NV forum was up you would have seen alot of ANGRY useres complaining about this)!
the majority of users are the GTX500 Family (the OC'd cards are the bigest victums).
(if you use any GPU teak software, simply UnInstall it and see if problem is fixed).
if not then simly go back a driver Untill NV will bring you a stable driver
-
(i had asus570 diver family 28x.xxx was total mess for me, so has the 30x.xxx family. - the only stable driver i found without any TDRs or memory leaks or freezups or lockups...... was the 296.10 Driver)!
-
(got asus 680 so the 300 family drivers are excellent for me now)
 
Last edited:
What do you mean by the TDR issue? My card is not OC'd nor do I have any tweak software.

I've been having this issue since early this year so I think the driver was probably 260? 290s all also crashed.
 
1stcowgirl said:
Yes Its the TDR issue (if NV forum was up you would have seen alot of ANGRY useres complaining about this)!
the majority of users are the GTX500 Family (the OC'd cards are the bigest victums).
(if you use any GPU teak software, simply UnInstall it and see if problem is fixed).
if not then simly go back a driver Untill NV will bring you a stable driver
-
(i had asus570 diver family 28x.xxx was total mess for me, so has the 30x.xxx family. - the only stable driver i found without any TDRs or memory leaks or freezups or lockups...... was the 296.10 Driver)!
-
(got asus 680 so the 300 family drivers are excellent for me now)
Click to expand...

Did you read any portion of this thread besides the first post?
 
I had a Gigabyte GTX 570 reference that began to show the excact symptoms described in OP. In the end the card died completely. Could not start my system anymore. Turned out it was a burnt VRM on the card.

As a note, reference GTX 570's barely have sufficient VRM setup for stock speed, let alone for overclocking. This is a poorly design from Nvidia. They tried to cheap out on the GTX 570 to separate it more from the GTX 580. Taking away some of it's needed VRM's was not very smart.

I always ran mine with a somewhat modest voltage and overclock. Was at 1.050v (My card was 0.975v stock) and 850MHz for the GPU. It lasted over a year before showing the excact symptoms described in the OP. I think my several hundred hours in Skyrim was the nail in the coffin for this card. And the thing that killed it completely in the end (After all the symptoms) was OCCT GPU test. It just blacked out and my system would not start anymore. After I removed the cooler I saw one of it's VRM's was burned.
 
Last edited:
The_Golden_Man said:
I had a Gigabyte GTX 570 reference that began to show the excact symptoms described in OP. In the end the card died completely. Could not start my system anymore. Turned out it was a burnt VRM on the card.
Click to expand...

Did you start a thread here about it when you were going through that?
 
I have the same problem with a 580 that just started out of the blue 2 days ago. Have had the card ~4-6 weeks with no problems, used from a friend. XCOM crashed and I had artifacting afterward. Blew out all the drivers and installed most current. Still get random artifacting in games and XCOM especially still crashes.

I ran GPUTool and it instantly shows lots of pixel errors. I used the find max function and getting the shader clock down to ~1200 was the fastest way to get it roughly stable. I have not tried reseating the card yet. Think it might help, or is this a clear case of RMA?
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top