- last edited on by
ROGBot
This problem is really starting to vex me...
The History:
Purchased a new ASUS GTX 780Ti 13 months ago. Card has been flawless up until a few weeks ago, when just about every 3D application I have would start, then immediately hang, die, and this error message would pop...
No hardware change whatsoever immediately preceded this issue. It just started happening, and then getting worse. At this point I can only see a few frames of BF4 before the entire thing crashes.
The Details:
This problem now persist through:
The RMA Aftermath:
A physical RMA - sent to Indiana a ~week ago, received back today. Same problem as soon as I fired up any 3D intensive program. ASUS provided no documentation (written or online) of what diagnostics were performed, their results, or what (if anything) was done to repair the card. Or any assurance that the card was now in good, working order.
RMA Number: USPC573197
Serial Number of Part in Question: E3YVCM088270
Part: GTX780TI-DC2OC-3GD5
Relevant System Specs:
ASUS Maximus VI Impact (BIOS 1603)
Intel 4670K
GSKILL 2 x 8GB DIMMs DDR3 2400 @ XPS CAS timings (10-12-12-31)
ASUS GTX780TI-DC2OC-3GD5
Seasonic Platinum Series 860 Watt Modular PSU
2 x 256GB Samsung 840 Pros in Intel RAID 0 (OS Drive)
The Bottom Line:
I've been building computers since the 1990s; every single personal rig I've ever built has contained a major ASUS component. I've heard the horror stories about ASUS post-sale technical support, but this is the first time I've really experienced its quality in person. There is something very wrong with this graphics card; it needs to be repaired or replaced. Apparently something further than a standard RMA is necessary to accomplish this. That is ridiculous.
What More Can I Do(?):
What more can I do to isolate the malfunction?
Who do I need to speak to at ASUS to get this resolved?
The History:
Purchased a new ASUS GTX 780Ti 13 months ago. Card has been flawless up until a few weeks ago, when just about every 3D application I have would start, then immediately hang, die, and this error message would pop...
No hardware change whatsoever immediately preceded this issue. It just started happening, and then getting worse. At this point I can only see a few frames of BF4 before the entire thing crashes.
The Details:
This problem now persist through:
- Multiple driver version installs (and complete wipes between) from nVidia's official site.
- Clean installed, fully updated versions of x64 Windows 7, 8, & 10.
- Multiple card reseatings.
- Multiple PSUs, single/multiple rails, and combinations of power cables.
- Underclocking card (and memory) to the very lowest setting the ASUS GPUTweak tool will allow. (With DirectCU II fans at 100% blast). Card has never been overclocked (beyond factory settings) and temps are well within expected thermal envelope; cards fails well before it even warms up (as in a second or two).
- Card fails all games, despite multiple OS and DirectX reinstalls.
- Card fails RealBench v2.4 OpenCL portion.
- Driver becomes unresponsive under moderate load with GPU assisted programs, like multiple tabs in stock Chrome open.
- TDR time delay and cancellation entirely doesn't fix (or even alleviate) issue.
- System is entirely stable otherwise; Memtest86+ (5.01) runs 10+ passes without issue; Handbrake has transcoded 10 Bluray rips to .mp4 back to back without issue or error.
- Entire system was rock solid for the first 13 months in this hardware configuration.
The RMA Aftermath:
A physical RMA - sent to Indiana a ~week ago, received back today. Same problem as soon as I fired up any 3D intensive program. ASUS provided no documentation (written or online) of what diagnostics were performed, their results, or what (if anything) was done to repair the card. Or any assurance that the card was now in good, working order.
RMA Number: USPC573197
Serial Number of Part in Question: E3YVCM088270
Part: GTX780TI-DC2OC-3GD5
Relevant System Specs:
ASUS Maximus VI Impact (BIOS 1603)
Intel 4670K
GSKILL 2 x 8GB DIMMs DDR3 2400 @ XPS CAS timings (10-12-12-31)
ASUS GTX780TI-DC2OC-3GD5
Seasonic Platinum Series 860 Watt Modular PSU
2 x 256GB Samsung 840 Pros in Intel RAID 0 (OS Drive)
The Bottom Line:
I've been building computers since the 1990s; every single personal rig I've ever built has contained a major ASUS component. I've heard the horror stories about ASUS post-sale technical support, but this is the first time I've really experienced its quality in person. There is something very wrong with this graphics card; it needs to be repaired or replaced. Apparently something further than a standard RMA is necessary to accomplish this. That is ridiculous.
What More Can I Do(?):
What more can I do to isolate the malfunction?
Who do I need to speak to at ASUS to get this resolved?
- It cannot be a software/driver/DirectX issue - it wouldn't persist through multiple, independent, clean installs of Windows, nVidia driver versions, and game software.
- System is completely stable in all respects when discrete card is removed and Intel CPU graphics are in use.
- System (with 780Ti DirectCU II) was completely stable in all respects for 13 months - negating the possibility of some kind of hardware incompatibility. Problem occurred suddenly and quickly (albeit in a graduated fashion) rendered all 3D, GPU based games and applications unusable.
- This particular error message is closely associated with graphics cards which are failing on the hardware level - underclocking the GPU can sometimes resolve the issue, but not in this case. Nor have excessive (70+) thermals ever been an issue with this card at idle or under load.
- Card arrived back from RMA today with zero change in card's condition/performance and zero explanation of testing/results/repair details beyond 'Card Repaired OK.'
That's horror for you probably.
I would say first some conflict with DX but if problem persist after OS installation than it's definitely hardware problem and source only could be graphic card. Similar or even same message show up when clock is too high and card unstable.
It's not something wrong, card dying as you sad.
I have 780Ti in computer and same driver 353.62. He is better than 2-3 drivers before that... 350.12 WHQL was good and this last driver.
NVIDIA fix issue with rear crashing in Chrome in idle or when Windows Explorer is open in same time with few browsers.
I would say first some conflict with DX but if problem persist after OS installation than it's definitely hardware problem and source only could be graphic card. Similar or even same message show up when clock is too high and card unstable.
It's not something wrong, card dying as you sad.
I have 780Ti in computer and same driver 353.62. He is better than 2-3 drivers before that... 350.12 WHQL was good and this last driver.
NVIDIA fix issue with rear crashing in Chrome in idle or when Windows Explorer is open in same time with few browsers.
Yeah, it's not been fun. (Especially when the card has functioned so well until this point).
I think this DX error can usually be fixed by slightly underclocking the card and/or slightly overvolting it (on a practical level, the error is essentially the card 'disappearing momentarily in software,' crashing the driver/DX); in my case the former does nothing for stability, and only when I overvolt it to the red line in GPU Tweak (in conjunction with underclocking) does it begin to make the slightest bit of difference (but is still too unstable for regular/consistent use).
Here's the error that pops after a single runthrough of Unigine Heaven fails, even with underclocking/overvolting.
That's why I'm absolutely certain at this point that it is, in fact, a hardware issue with the card itself, despite the RMA making no appreciable difference. It is frustrating.
This card isn't even half-way through its stated ASUS warranty period.
And proof of date of purchase and that this card was purchased new, from Newegg.com - not second hand from eBay.
I think this DX error can usually be fixed by slightly underclocking the card and/or slightly overvolting it (on a practical level, the error is essentially the card 'disappearing momentarily in software,' crashing the driver/DX); in my case the former does nothing for stability, and only when I overvolt it to the red line in GPU Tweak (in conjunction with underclocking) does it begin to make the slightest bit of difference (but is still too unstable for regular/consistent use).
Here's the error that pops after a single runthrough of Unigine Heaven fails, even with underclocking/overvolting.
That's why I'm absolutely certain at this point that it is, in fact, a hardware issue with the card itself, despite the RMA making no appreciable difference. It is frustrating.
This card isn't even half-way through its stated ASUS warranty period.
And proof of date of purchase and that this card was purchased new, from Newegg.com - not second hand from eBay.
Friend that is cruel reality. Your graphic card is fault. I know that, you know that, all advanced users know that here.
That's problem you are in moment where 99% of chance is that problem cause ASUS hardware. Because of that you will not get more answers or help.
Today no such brand who produce low percent of fault units...With every brand I always leave possibility that something maybe not work. Because of that I didn't bought R5E from Hungary for 390e, I payed 460-470e in Belgrade. Because It's much easier with ASUS distributor near me. I could saved lot, but't that's not memory, SSD, cooler, graphic card, on motherboard many things could go wrong.
After production cross in China all of us are victims and we can't do nothing to change that... But we can do something else. Before upgrade we can choose someone who react instantly if problem show up, not someone to back you fault hardware. I have no such problem to bring two PC in store and show them what is problem and ask them to try in front of me in their PC to show me how that work if they say it's fixed.
I advice you to nice explain problem on paper to ask RMA again. Explain them over phone or e-mail or personally or write on paper that they didn't do nothing, they back you hardware in same fault condition and if you bought in some local store ask them to try in front of you in their working PC, if they think that something else is wrong with you. If problem gone and you see graphic work normal with own eyes than you need to deal with problem on different way... If card start to behave same, crash with DX message they can't do nothing else than to try other sample in front of you and replace your card.
I talk about this with possibility that you tried everything and that really no other solution after Windows reinstall, DX update, Microsoft Visual C++ installation, NET Framework all version and updates, good PSU, normal temps, fabric clock, different drivers and if other graphic card work in your PC normal.
That's problem you are in moment where 99% of chance is that problem cause ASUS hardware. Because of that you will not get more answers or help.
Today no such brand who produce low percent of fault units...With every brand I always leave possibility that something maybe not work. Because of that I didn't bought R5E from Hungary for 390e, I payed 460-470e in Belgrade. Because It's much easier with ASUS distributor near me. I could saved lot, but't that's not memory, SSD, cooler, graphic card, on motherboard many things could go wrong.
After production cross in China all of us are victims and we can't do nothing to change that... But we can do something else. Before upgrade we can choose someone who react instantly if problem show up, not someone to back you fault hardware. I have no such problem to bring two PC in store and show them what is problem and ask them to try in front of me in their PC to show me how that work if they say it's fixed.
I advice you to nice explain problem on paper to ask RMA again. Explain them over phone or e-mail or personally or write on paper that they didn't do nothing, they back you hardware in same fault condition and if you bought in some local store ask them to try in front of you in their working PC, if they think that something else is wrong with you. If problem gone and you see graphic work normal with own eyes than you need to deal with problem on different way... If card start to behave same, crash with DX message they can't do nothing else than to try other sample in front of you and replace your card.
I talk about this with possibility that you tried everything and that really no other solution after Windows reinstall, DX update, Microsoft Visual C++ installation, NET Framework all version and updates, good PSU, normal temps, fabric clock, different drivers and if other graphic card work in your PC normal.
Hi guys I have the same problem with a 1 day old 980 ti strix, bought it to replace my 770 dc2oc 2gb sli. Bf4, path of the exile, haven't tried other games all with settings from nvidia experience, not using custom, with 4k DSR and without at FHD.
I am thinking of bad drivers instalation I had removed only the driver and not all of the nvidia stuff.
P.S. path of the exile had the same issue when I had enabled mfaa to try it out, it gave me that error returning to off the error didn't came on the next tries.
On bf4 even I disable mfaa still the same issue.
I had the card with manual OC, tweak OC, tweak gaming OC, same issue on all. The temperature didn't go over 63C
I am using the 770 as a physx card on pcie slot 5 on a pcie 4x. The issue remains even the 770 is removed.
Or is it a hardware issue the pcie had the yellow part on a place removed out of the box. Here a photo
I7 2600k with corsair H100i
P8p67 deluxe
Corsair 16gb 1600
Samsung 840 pro 256
Samsung 830 256
Corsair AX1200i
Win 10 pro
Old gfx SLI asus dc2oc 770 2gb
P.S. on first time using the 980ti I had the second 770 on the pcie slot 3, as I had it on sli setup. After work today I will try to repair the game. Remove all driver with 770 removed and restart and install new. Dl dx from Ms and install it. And if still the same I will reinstalling windows.
I am thinking of bad drivers instalation I had removed only the driver and not all of the nvidia stuff.
P.S. path of the exile had the same issue when I had enabled mfaa to try it out, it gave me that error returning to off the error didn't came on the next tries.
On bf4 even I disable mfaa still the same issue.
I had the card with manual OC, tweak OC, tweak gaming OC, same issue on all. The temperature didn't go over 63C
I am using the 770 as a physx card on pcie slot 5 on a pcie 4x. The issue remains even the 770 is removed.
Or is it a hardware issue the pcie had the yellow part on a place removed out of the box. Here a photo
I7 2600k with corsair H100i
P8p67 deluxe
Corsair 16gb 1600
Samsung 840 pro 256
Samsung 830 256
Corsair AX1200i
Win 10 pro
Old gfx SLI asus dc2oc 770 2gb
P.S. on first time using the 980ti I had the second 770 on the pcie slot 3, as I had it on sli setup. After work today I will try to repair the game. Remove all driver with 770 removed and restart and install new. Dl dx from Ms and install it. And if still the same I will reinstalling windows.
