Amareis Posted May 11, 2014 Share Posted May 11, 2014 Hi guys I'm hoping someone can point me in the right direction here. My specs: GFX: Asus GTX 780ti Mobo: Asus Gryphon z87 CPU: Intel i7-4770k (stock speeds - no oc'ing here) RAM: Corsair Dominator 1866MHz (CMD16GX3M2A1866C9) HDD: Samsung 840+830 SSD's PSU: Corsair AX1200 Windows 7 SP1 64bit Hardware is a little over a year old, except the 780ti which is brand new (and the 2nd card since I RMA'd the first card, thinking it was the problem, it was not). Issue: Whenever I start one of the three following games, 10 seconds to 1 min later the screen goes black and my computer reboots: World of Tanks Mechwarrior Online Battlefield 4 Exceptions: I can play Diablo 3 onboard and with the 780ti without problems. I have had a GTX 670 that has worked flawlessly in this system. I can run FurMark both windowed and fullscreen without any problems. So I'm on my 2nd 780ti and have tested the following: Tried switching to the 2nd PCI-E slot on the motherboard. Reseated RAM and gfx card. Tried moving the modular cables on the AX1200 to other sockets. Tried a myriad of different drivers. Updated drivers and BIOS on motherboard and gfx card. Tried various settings in the BIOS. Nothing has even remotely changed the behaviour of my system. I have logged the card's usage through ASUS GPU tweak-up and Techpowerup GPU-Z. This revealed that the only difference between the games that crash and furmark is GPU memory usage. Furmark does not use a lot of memory, but the games do. I will try and test my PSU and hopefully the card as well in a friends system. He also runs an AX1200 but has a EVGA GTX 780 (not TI). If his system is fine with my AX1200, then I'm out of things to test. What do you think? Link to comment Share on other sites More sharing options...
Corsair Employees jonnyguru Posted May 12, 2014 Corsair Employees Share Posted May 12, 2014 Not the PSU. PSU would shut down. Hopefully, it's just the RAM. Do this: Go to "system properties". Go to "advanced system settings". Under "startup and recovery" click "settings". Under "system failure" uncheck "automatically restart". Now what happens when you run one of those games? Link to comment Share on other sites More sharing options...
Amareis Posted May 12, 2014 Author Share Posted May 12, 2014 Not the PSU. PSU would shut down. Hopefully, it's just the RAM. Do this: Go to "system properties". Go to "advanced system settings". Under "startup and recovery" click "settings". Under "system failure" uncheck "automatically restart". Now what happens when you run one of those games? I tried what you suggested and the screen goes black, I can hear my two 3.5" drives going click as if they turn off and then nothing. No reboot. Then I can click my power button and the system immediately shuts off as if windows isn't running. You know, usually you have to hold it for 4 seconds. Also I tried a friends AX1200 with exactly the same result. Then I tried his EVGA GTX 780 (not a TI) and the system is flawless. No problems at all. Played Battlefield 4 for 30 mins, no issues what so ever. The weird thing is I can run Diablo 3. I confirmed that the log from GPU-Z says that Diablo use around 300 MB of video ram whereas the games that crash use around 1800 MB. EDIT: Found a lot of bluescreen reports that link to DirectX and nvidia display drivers. Others have them as well. I'll dig deeper into this to see if it has anything to do with my windows installation or driver installation. One guy mentions a mobo replacement (http://www.techsupportforum.com/forums/f299/solved-bsod-due-to-dxgkrnl-sys-604177.html). Link to comment Share on other sites More sharing options...
Corsair Employees jonnyguru Posted May 12, 2014 Corsair Employees Share Posted May 12, 2014 Hmm... Well, I can see how you would think the graphics card is at fault. But you're already on your second card! Could be the motherboard, but I have no idea why. Not get enough power through the PCIe slot? Though both cards should have the same TDP. Also, remember: The power button on the case goes to the motherboard. Not the PSU. So if you press the button and it turns off immediate versus having to hold it down for four seconds, that variation is coming from the motherboard.... not the PSU. Typically, it should take four seconds. But that's not because the PSU waits four seconds. It's because the motherboard waits four seconds before cutting off the signal to the PSU. Link to comment Share on other sites More sharing options...
DCult Posted May 13, 2014 Share Posted May 13, 2014 Hello... I have the exact same problem as you. My card is a ASUS GTX 780 ti DCUII o/c. I had a GB Windforce 770 GTX o/c which worked flawlessly in my system, and then found a chance to upgrade to a gigabyte wf GTX 780 ti o/c which would freeze in games in factory clocks (1020MHz) and artifact in Heaven but would work without crashes when downclocked at -100MHz. So after I tried everything, I exchanged it for the ASUS a few days ago. I have stability in all benchmarks, 3D mark 11, 13, heaven, valley, etc, max temps 72c, but the games for me that force the screen to go blank without signal (I haven't tried many) are Crysis 3 and Batman Origins. Things I tried: - Took the card back and got the same with a consecutive serial number (shouldn't have done that), kept the second one though, because it had a higher ASIC quality (79,8%) and would boost by itself at 1097 MHz - Changed PCI-e 16 slot and reinstalled latest WHQL. - Borrowed an identical PSU (Corsair AX 860) from a friend, who runs the store where I bought the card and PSU. The PSU powered 2 Xeons for a long time flawlessly. - Changed cables in my PSU, from 2 (8pin to 6+2 pins) to only 1 (8pin to 8+8 pins) and to 2 (8pin to 8pin+8) - Tried different 8pin (CPU+pci-e) connectors in the PSU. - Removed a Belkin 1200VA UPS from the power line that wasn't even connected to the PC. - Updated motherboard BIOS to latest f15q beta (it's a BETA for the last 1.5 years) There are some options in the BIOS about GFX overcurrent protection that varies from auto to standard, normal, low, medium, extreme, but god knows what it actually means... extreme protection or extreme allowance? So I haven't messed with these. Weird thing is that the first card that had a lower ASIC quality (69%) would reset/go blank instantly upon loading a saved game in Crysis3, whereas the second one needs a few minutes. I have run out of ideas, just like you. My pc specs: i7 2600k stock freq. Corsair A-50 cooler Gigabyte Z77X-UD5H 16 (4x4) GB Corsair Vengeance Cas 8 ASUS GTX 780 ti Direct CU II o/c Intel SSD 520, 240GB WD Black 2TB Windows 8.1 Pro 64bit with Media Center (update1) Silverstone Fortress FT02. Excuse my errors, English is not my mother language (greek is) Link to comment Share on other sites More sharing options...
Amareis Posted May 16, 2014 Author Share Posted May 16, 2014 Still no luck. Have tried the card in a friend's computer. He has a Corsair HX1050. Same result. He did however discover that he could run Assasins creed IV in 1080p on low for some time, then he changed the settings to Ultra and BAM restart. This enforces my belief that the card does not have stable access to the memory. ASUS has made a big mistake somewhere in their production of this card. I've reported it to their technical questions department along with my findings from various logging software. They show that applications that access a lot of memory will crash the system. Furmark and Diablo 3 only use around 300-500 MB of RAM while BF4 wants 1800 mB and so will cause a crash after a few minutes to as little as a few seconds. Their support is great and competent, but he just said that he would report it to their R&D department and it could take a long time for them to come up with anything. If we could get Tom's hardware, Anandtech and Hardware Canucks to dig into this we get a response sooner. Word on the Internet is blaming PSU's left and right, but I don't believe 3 Corsair PSU's, that are running just fine without the card, could all be faulty. Link to comment Share on other sites More sharing options...
rene mauricio Posted May 16, 2014 Share Posted May 16, 2014 The black screen issue you are describing sounds an awful lot like the one addressed by the latest Beta Driver: 337.81. Contains fix for this the monitor display blank screen issue that started with R325 drivers - Source Link to comment Share on other sites More sharing options...
HOODedAssault Posted May 16, 2014 Share Posted May 16, 2014 Either driver or they sent u exact same card back lol. Knowing asus it's pretty likely. If u tested just the card and it did exact same thing there's no doubt it is for sure the card or graphics driver. Link to comment Share on other sites More sharing options...
Amareis Posted May 16, 2014 Author Share Posted May 16, 2014 Either driver or they sent u exact same card back lol. Knowing asus it's pretty likely. If u tested just the card and it did exact same thing there's no doubt it is for sure the card or graphics driver. I wish. I RMA'd to the vendor, not ASUS. I had the old and the new card in my possession so that's a no. I highly doubt that all the driver versions would act exactly the same. I've tried the ones that was on ASUS Support site and several others in between those and the newest WHQL drivers. I'll give latest beta drivers a go, but my hopes are not up... Link to comment Share on other sites More sharing options...
rene mauricio Posted May 17, 2014 Share Posted May 17, 2014 Make sure you use something like DDU to clean up all traces of your old drivers. Link to comment Share on other sites More sharing options...
DCult Posted May 17, 2014 Share Posted May 17, 2014 The black screen issue you are describing sounds an awful lot like the one addressed by the latest Beta Driver: 337.81. - Source I'm afraid the blank screen problem you are referring to has to do with unsuccessful installation of the driver, rather than mine and Amareis' problem. Anyway, I will wait for the whql driver, it doesn't hurt to try. I had the 770 windforce which was working with all 325+ drivers so it would have the same problem. I got an email from ASUS that my card is probably defected but I want to exclude all possibilities. Link to comment Share on other sites More sharing options...
DCult Posted May 17, 2014 Share Posted May 17, 2014 This is no longer a corsair PSU potential problem for me, not that I was convinced it ever was, since I put the same AX860 inside and it behaved exactly the same. I will keep an eye on the thread, thanks for the help. Cheers. Link to comment Share on other sites More sharing options...
Amareis Posted May 17, 2014 Author Share Posted May 17, 2014 Make sure you use something like DDU to clean up all traces of your old drivers. Tried it and nothing changed... Tried the 337.81 drivers. No change. I've ordered an Asus Maximus z97 Hero motherboad just for testing but I'm getting tired of doing ASUS's job and testing all of these things. I'm gonna send the Hero motherboard back and have the 780ti exchanged to a regular 780 or maybe just try a 780ti from another manufacturer. Link to comment Share on other sites More sharing options...
DCult Posted May 31, 2014 Share Posted May 31, 2014 Tried it and nothing changed... Tried the 337.81 drivers. No change. I've ordered an Asus Maximus z97 Hero motherboad just for testing but I'm getting tired of doing ASUS's job and testing all of these things. I'm gonna send the Hero motherboard back and have the 780ti exchanged to a regular 780 or maybe just try a 780ti from another manufacturer. Any news? Have you got a new 780ti? Any new suggestions what might be at fault? Thank you in advance. Link to comment Share on other sites More sharing options...
Dogleg1957 Posted June 2, 2014 Share Posted June 2, 2014 I too just had the same problem couple weeks ago with EVGA GTX 780 w/ ACX, standard not ti. I reloaded driver from GeForce experience and still got black (seemed grayish to me) and buzzing noise in speakers for 15 secs. but also windows opened window showing it was a gpu problem. so went back and downloaded from evga site 337.50 drivers and problem fixed. also all my black screens came while playing world of tanks, with settings at max my vcard runs constant 80C. my friend recommended using a driver sweep to clean out all old drivers, but didn't need to do that in my case. my tech skill are noobish at best so maybe I just got lucky Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.