• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Faulty 260

SH4RKY

SH4RKY

SH4RKY

Associate
Joined
28 Aug 2004
Posts
766
Location
b'ham
Just after some opinions/advice about my GPU really:

I have an EVGA 216 GTX260 SC. its one of the ones before the 55nm cores.

Basically, playing L4D the PC will hard crash, both screens go into pink checkered crazy patterns, sound will loop and the only way to restart the pc is to hold the power button for 5 secs. reset doesnt work - the screens will still be mullured.

I originally thought this was the GPU, so requested an RMA and sent it off, EVGA tested it and it came back as fine. (side story - i had already tested it with GPU tests as they had requested i do so to check temps before sending it in - it was fine in the tests, no errors etc. So as they tested it using the same programs im not entirely surprised they didnt find a fault.)

anyway... PC still crashes when playing L4D (does crash the same way with other games, L4D does it most frequently). I had done everything i could think of to test other parts of the PC just in case, but its stable, OC or not, every other part seems rock steady.

I declocked the card, put the fan on 100% and guess what, L4D still crashed but this time it took over an hour to do so. Put it back on stock settings and boom, crash within 10 mins. To really back my theory up I borrowed a friends pc to put the GPU in that.... L4D crashed. I put his GPU in mine and L4D was fine for 3 hours.


So... pretty conclusive? its the GPU thats faulty. Right?
 
I was having a very similar problem with my XFX GTX 260 Black Edition, turned out my PSU was not up scratch, I had a 720W Hiper PSU but it wasn't outputting enough amps on the 12v rail, replaced with Corsair 750W and it's been perfect ever since.
 
Have u tryed any other games apart from L4D could be somethink as simple as the driver
version your using dont like L4D
 
Yup, its crashed in WoW (much less frequent though), it does it in fallout quite a lot, The Trine demo on Steam.... Ive tested it with old drivers, new drivers, ati drivers.... haha ;)

I have a BeQuiet Darkpower 650w. The first one blew up (faulty) as i originally thought it may have been the PSU, but the replacement is perfect. The PSU is overkill for my system. (quad 9450, 4gb, 2 HDDs, 1 dvd).

The friends PC has a corsair 650 (....or is it a 750. Checking on that) but yeah, a 650w PSU is ample for a 260. The bequiet has separate 12v rails and its spec is more than the 260 needs so technically it is fine....
 
Check the amps though, the GTX 260 requires 38amps on the 12v rail, my Hiper PSU was giving out 16 amps on each of it's four 12v rails, whereas the Corsair is 60 amps on one single 12v rail.
My symptoms were pretty much identical to yours, GPU was passing all stress testing as was every other component but it was just randomly crashing and rebooting itself during gaming, since replacing the PSU it's *touch wood* been perfect.
 
Taken from BeQuiets site:

4 seperated 12V rails#
Specification: +3,3V: 26A, +5V: 30A, +12V1: 20A, +12V2: 20A, +12V3: 20A, +12V4: 20A, -12V: 0,8A, +5VSB: 4A (peak 5A)#

So... the 260 has 2 power inputs and needs 38A max, the bequiet has 20A per 12v Rail. so is giving it 40A. Technically it should be fine?

If my friends Corsair is a 650w - it has 52A per +12V rail! (or 52 on one and less on the other?) either way, thats more than enough?

What is interesting is even if I get a 750w bequiet, it still only has 20A per 12v rail so it would provide the same power to the GPU...


Anyone on here have a GTX 260/280 and a bequiet?
 
They amps dont stack like that, I don't have a very good understanding of it all but from the hours I spent Googling this exact same issue it seems you can't just add the 2 amp values together for the 2 seperate rails, it's all very confusing!
I just looked to see what most people were recommeding to use and the Corsair kept getting highly recommended so I took the risk and bought one and it sorted me out.
 
Yeah :)

I was the same, BeQuiet where highly recommended from people with similar set ups to mine so took the plunge on that one.

Hadnt looked in detail at the Ampage situation though, so thanks for the tip. However I still feel its the GPU at fault as it has the exact same crash on a Corsair PSU. The same friend is getting the XFX 260 this week (cheap as chips at the moment! i paid 250 for mine!!) so ultimate test will be putting that in my machine and see if it crashes.
 
are you using two seperate 12v rails for each input on the gtx 260 though? i have 3 12v rails for pci-e, but each one has two connectors (a 6 pin and an 8 pin) on it. Not using a BeQuiet though so i dont know if its the same on yours.
 
The PSU has 2 PCIE power connectors on separate cables (assume separate rails).

Maybe its possible that the HDDs/DVD are sitting on the same rails? but from the technical blurb from bequiet, it sounds as if each GPU cable is on its own rail.

Ill disconnect the DVD and other HDD tonight and swap some cables around for the other HDD. I'll try the dual Molex to PCIE converter you get with the GPU too.

Although, clutching at straws? If the GPU was being underpowered, would I be able to play games at all? Or even run the stress tests across 2 screens (1920x1200 and a 1600x1200) with the rotating HD balls and ATI stress tester with no issues or even the GPU Folding at Home??
 
Same was happening for me, it didn't and still doesn't make sense why it would take all the stress testing but then randomly give up during gaming. I was close downing the bottle of Isopropyl I had near by to numb the pain!
I can only assume that it has something to do with the rest of the system kicking in and consuming lots more power when gaming which pushed the PSU over the edge.
 
Yeah, I thought similar about why it only crashed in games. as well as using the rest of the system - maybe its using more of the GPU (physics, rendering etc) all at the same time rather than just focusing on one aspect like a stress test does.

But that could apply to lacking power or a faulty/overheating GPU...


Im still more inclined to say its the GPU at fault at the moment as its crashed on 2 different brands of PSUs and 2 different PC setups.
Also, the way the crashes can be delayed by putting the GPU fan at 100% suggests its a temp related board issue...


.... maybe im sticking to that theory because I don't want to blow 130quid on another PSU, rewire my PC only to possibly find its not the cause... haha.



Still curious if anyone else has a 260 and bequiet 650 combo... gotta be some on here :)
 
Nvidia forums, so link is ok? if not please remove: http://forums.nvidia.com/lofiversion/index.php?t90352.html

Similar issue. There is also some stuff about it on the EVGA forums. Some people have mentioned that EVGA have reflashed their cards BIOS to give the cards a bit more juice.

image of PSOD (not mine but i have one thats similar): http://img177.imageshack.us/img177/4729/imag0014uc1.jpg


Edit: hmm.. ok, lots of it on the evga forums. Ill read up over there and see whats what.

Edit: Yeah loads about 260 SSC and SC cards having the exact same issues. The main suggestion is to try lowering the shader clock as the pre overclocks evga are applying might be too much. Ive set mine to stock 260 settings and it still crashed so that doesnt help me anyway... haha.


The other issue i see is, they dont make 65nm cored versions anymore? at least you cant buy them... Is there a chance it will be replaced with a 55nm one? is there any downsides to this?
 
Last edited:
Not the PSU if it crashes on two separate units. Either the BeQuiet or the Corsair should easily be enough for a GTX260 - crashing in both systems indicates that it's definitely the card.

Seems like it could be something to do with temps - what's your GPU core temp whilst gaming?
 
Known issue with the 260GTX - generally they are ok but a few don't seem to get adequate cooling to a certain part (unknown) of the card - or possibly some have components with a lower heat tolerance than normal - Left 4 Dead is one of the most common to cause it - but usually turning the fan too 100% for L4D will sort it.
 
Yeah, I can usually finish a game if i have the fan on 100%. Will still crash though, seems to be getting worse too.

Temps... Its on air in a P182 so not exaclty the coolest of cases.... 50-55 idle. Fan on Auto in a game it will reach 78-82. Infact, the Trine demo pushed it to 86 which freaked me out. Its ok though, the crash shortly into the demo cooled things down.... haha.

Im asking for an RMA tonight with a big explanation of all my testing proving, in my mind at least, its the card at fault. I just hope they dont be silly and test it again and send it back. That will annoy me considering ive already paid postage once to them...


Thanks for the info and tips. Ill update as and when I get a result.
 
Ah with those temps not a hope - the card I had that problem with ran 38-40 idle and 64-65C in furmarks (with more agressive fan settings than default) and setting 100% fan I could play through all 4 scenarios without it crashing - whereas on stock fan settings it would random lockup around 30 minutes in.

Deffinatly RMA it IMO its not acceptable.
 
Do you have the possibility of pointing an extra fan at the rear of the card, perhaps the VRM/Power chips are overheating?
 
You must log in or register to reply here.
Back
Top Bottom