Hi there, I bought an ASUS GeForce GTX 560 1GB six years ago, and it's been keeping me company for a long time without any issues. I'm on a Gigabyte Z77-D3H board with 24GB RAM (as one RAM slot is broken; the PC won't boot; but all other three work fine and PC boots fine; never bothered to RMA it). Three months ago, I received these white lines on the monitor on the line ... but a simple reboot fixed the issue. Today, I came from work to play my very intense modded Cities Skylines, as I normally would, and the card failed. So I started doing some good old googling to find out whether the GTX is fried, or whether it's a driver issue. The PC boots fine with the card, but the lines appear even prior to Windows. They appear in BIOS too: Here's what I did: Ran DDU. Uninstalled the drivers, I do not remember which version I had. Ran into safe mode, running DDU there too, multiple times. Opened the case, took the GTX out of the PCI slot, opened it up to give it a dust clean. Not much dust was there. I resealed it with new thermal paste, and put it back together. Ran sfc /scannow which returned some errors, but it cannot fixed them, referring me to a log. I can open the log, but I don't really know how to interpret it. During my multiple driver installs, I had the following results: 314.22 (as recommended by Reddit as a stable driver) worked only with 640x480 resolution. I was not able to change my resolution. But I get no white lines! 347.88, I cannot install. I get an error that it is incompatible. 359.06 (recommended on another forum) produces same results as 314.22 391.15 (latest WHQL) works only with resolutions lower than 1280x1024 (where my monitor can support up to 1920x1080) and produces the white lines effect. On some drivers, the card is disabled with Code 43. As a result, I took the card out of the computer, and I'm running the integrated card on the Gigabyte Z77-D3H board - which is far better for the timebeing than white lines and annoying, warped resolution. Here is also a shot from GPU-Z (which shows 0 MB Memory Size, does that mean anything?): But the question remains, is it time for a new graphics card, or is it a drivers issue? What could have caused it? Temperatures were never high, and the PSU was able to cope with the card for a long time. I have been mining bitcoins on it lately though so maybe that was why? I presume it's also way beyond warranty period, so there's no option of RMA (unless I'm wrong).
If the issue persists in BIOS display, it isn't a driver problem. Looks like it's a GPU component issue, probably memory. Baking the card may or may not help. I think it's time to shop for a new card.
Have you been playing with your vbios? I remember having a similar issue with my GTX 580 a few years ago after a bad flash (fixed by reflash) if not, then perhaps it's damaged.
classic symptom of memory errors, probably the solder joints connecting the memory chips or gpu(on the graphics card), as mentioned above "baking" might get you a bit more time, but its probably time to get something new.
If you have 2 or more RAM, check one by one. Remove all of them. Install one at a time and check. Or if u have only 1 RAM , change slot. try Else: Disassemble your gpu and bake
Classic memory corruption, As said baking may or may not revive it for a short while so worth a try until you can get something new I had a amd r9 280 that lasted 6 months and 2 bakes
Even though it looks like it's a VRAM issue, make sure it's not the monitor first. Use the monitor with another GPU to check. You can also try to downclock the memory in Afterburner and see if that helps.
It's impossible because he/she got artifacts in BIOS post before starting OS despite all GPU stock speeds!! lool Solutions: he/she should bake videocard during 10 minutes at 200 ºC in oven or he/she can buy new card. ONLY!
That happened to me in past and returned during warranty for another. In your case I believe one of the memory chips is bad.
Definitely bad VRAM, I've had a few cards with that sort of corruption. Last was a GTX 680 I bought 2nd hand a few years back, baking it made it error free for a good 6 - 8 months then I'd have to bake it again but it was a case of diminishing returns. I'd just get a new graphics card if I were you.
Not the RAM, its the vram that is the issue. You can easily run a memory diagnostic test to rule it out.
