HackManSD Posted July 9, 2014 Share Posted July 9, 2014 Tried a lot of troubleshooting to solve this issue. Client had a single 780 ti and ax760 and was fine, high FPS and scores in benchmarks. The card was able to overclock with no issues at all. Client purchased a second 780 ti and AX1200i to run them, causes immediate crashes in games and benchmark. Crash is not correct, PC completely restarts as if nothing happened. Each 8 pin has it own cable from the supply. Tried each individually...no problems CPU overclocked at 4.8 or stock, same restart Ram running XMP or Auto 1333, restart Everything is brand new so it shouldn't, but could be, the power supply. Rampage IV Extreme with latest BIOS 16GBs Corsair 2400 Ven Asus DC II GTX 780 Ti X2 AX1200i Help please! Link to comment Share on other sites More sharing options...
jameyscott Posted July 9, 2014 Share Posted July 9, 2014 Try using the AX760 and see if it does the same thing. the AX760 will be able to handle everything at stock just fine. At least long enough to run a benchmark and see if the same problem occurs. This will help isolate it to see if it is really the PSU. Is there OVP built into the AX1200i like there is with the AX1500i where it will only allow a certain amount of amperage per rail when it is in multi-rail mode? I'm not extremely familiar with the PSU since i don't own it, but doesn't it have a single rail mode and a mutli rail mode? If so, try switching it to whatever mode you are not currently in and see if that helps as well. Link to comment Share on other sites More sharing options...
xclock Posted July 9, 2014 Share Posted July 9, 2014 i think the problem here is your Ti's and NOT the AX1200i. most of Ti's in sli mode had this kind of problems. it looks nvidia drivers are not fully optimizing sli. did you test each card separately? Link to comment Share on other sites More sharing options...
HackManSD Posted July 9, 2014 Author Share Posted July 9, 2014 i think the problem here is your Ti's and NOT the AX1200i. most of Ti's in sli mode had this kind of problems. it looks nvidia drivers are not fully optimizing sli. did you test each card separately? Each card runs fine on its own. Stock or overclocked but as soon as we put them in SLI....nothing but restarts. I don't understand why it would restart the PC altogether. No blue screen, nothing in logs just flat out restart. Link to comment Share on other sites More sharing options...
Ahtlon Posted July 9, 2014 Share Posted July 9, 2014 Are the 780ti from the same brand and series, it could be that the bios on the gpu are very different, did you also tested the pci-e on the mobo, it could be that one is faulty. It's not the psu I think, but I never use the multi rail, also try if it could be the sli-bridge , did you tried the gpu's in a other pc. Link to comment Share on other sites More sharing options...
Corsair Employees jonnyguru Posted July 11, 2014 Corsair Employees Share Posted July 11, 2014 Each card runs fine on its own. Stock or overclocked but as soon as we put them in SLI....nothing but restarts. I don't understand why it would restart the PC altogether. No blue screen, nothing in logs just flat out restart. Make sure that, under advanced system settings, you have "automatically restart" unchecked. It's checked by default. If each graphics card has it's own cable from the PSU, it wouldn't matter if you had one, two or three graphics cards as each connector has it's own OCP set point. Additional cards aren't increasing the loads on other connectors... only onto additional connectors. Besides, you can turn OCP off or increase it in the Link software. If you turn off "automatically restart" and still don't get a BSOD, then I would suspect a problem with the motherboard. Link to comment Share on other sites More sharing options...
HackManSD Posted July 12, 2014 Author Share Posted July 12, 2014 Make sure that, under advanced system settings, you have "automatically restart" unchecked. It's checked by default. If each graphics card has it's own cable from the PSU, it wouldn't matter if you had one, two or three graphics cards as each connector has it's own OCP set point. Additional cards aren't increasing the loads on other connectors... only onto additional connectors. Besides, you can turn OCP off or increase it in the Link software. If you turn off "automatically restart" and still don't get a BSOD, then I would suspect a problem with the motherboard. I went ahead and unchecked the automatically restart option. Loaded up 3dmark and ran Firestrike, it gets through the first test but always restarts on the second test. Weird thing is, after it restarted and booted into Windows, 3dmark re-opened and tried running the second test on its own. I literally did nothing and watched this thing restart the test again. I haven't tried to use the Corsair Link software yet to either increase or turn off the OCP. I really don't think it will help much though. The cards are both exactly the same. Asus DC2 GTX 780 Ti 3GB. I have tried them both individually and they work but together....they restartr. I did notice something pretty strange while testing. I haven't noticed it before as I was only using 3Dmark for testing. I loaded up the Valley benchmark to see if it would restart, it didn't, but both cards cores are running at 1361MHz! This happens with or without tweaking software loaded/installed on the system. I have no clue as to why they are clocking themselves so high but I'm sure it is causing the restarts. Link to comment Share on other sites More sharing options...
Corsair Employees jonnyguru Posted July 12, 2014 Corsair Employees Share Posted July 12, 2014 Motherboard. Link to comment Share on other sites More sharing options...
HackManSD Posted July 12, 2014 Author Share Posted July 12, 2014 Motherboard. What on the motherboard could be causing this type of behavior? This is a first for me and really don't look forward to replacing a $400 board but if it is the culprit....no problem. What do you think failed on the board to cause this? Link to comment Share on other sites More sharing options...
Corsair Employees jonnyguru Posted July 12, 2014 Corsair Employees Share Posted July 12, 2014 Poor distribution of power to the PCIe slots. Faulty controller. Bad slot. A number of things. Of course, if replacing the power supply is easier (I imagine it probably is), then do that. If it continues after you do that, you know what the next step is. Link to comment Share on other sites More sharing options...
HackManSD Posted July 16, 2014 Author Share Posted July 16, 2014 Poor distribution of power to the PCIe slots. Faulty controller. Bad slot. A number of things. Of course, if replacing the power supply is easier (I imagine it probably is), then do that. If it continues after you do that, you know what the next step is. OK, replaced the R4E with a Black Edition and still having the same problem. Used the AX760 from old build, same restart. Can't figure out which card is causing the restart as both run fine when single. I am at a loss as to why its doing this. First time I've had any issues with Asus stuff. Is their anyway to tell which graphics card could be at fault as they run/overclock fine when run alone. Any help would be much appreciated. Link to comment Share on other sites More sharing options...
xclock Posted July 16, 2014 Share Posted July 16, 2014 can you take both card to run them in another system or local pc shop to let them run in sli mode? or borrow PSU from someone to let it run in your pc in sli mode? Link to comment Share on other sites More sharing options...
HackManSD Posted July 17, 2014 Author Share Posted July 17, 2014 can you take both card to run them in another system or local pc shop to let them run in sli mode? or borrow PSU from someone to let it run in your pc in sli mode? Semi-Solved! We tried numerous motherboard/psu combos in attempts to find the cause of this very annoying issue. Swapped all sorts of hardware, everything but the case itself, trying to eliminate each piece as the problem. Blindly stumbled upon a thread with some random files Kingpin uses when doing his LN2 runs and finally caught a break. Disable.exe----it disables the over current protection on the Asus graphics cards when using more than stock voltages....high overclocking with LN2. As a last ditch effort, we ran the disable.exe in a command prompt window to see if it was something on the cards themselves causing it. Sure enough, we were able to run Firestrike without issue. Added 75 to the cards cores and 100 to the memory. No more instant reboot, she went through and gave us the highest score his system has ever seen. 18449 overall 22871 graphics 17765 physics 7713 combined Will update later as we will be pushing the CPU back up to the 4.8-4.9 mark. Also going to find the max core/memory clocks for the cards and re-run. Should prove to push out an excellent score! Had to add the disable file to Windows start up so it runs when it boots. Why on earth did Asus have the OCP so low on factory overclocked cards??? We were still on stock voltages and stock clocks for all of the testing and adding this disable allowed us to fully utilise these cards. I'll be passing the word along to everyone having these problems in hopes to keep others from going through all the hardware we did for testing. Thanks for all the help guys!:laughing: Link to comment Share on other sites More sharing options...
Corsair Employees jonnyguru Posted July 17, 2014 Corsair Employees Share Posted July 17, 2014 Wow. Just, wow. Link to comment Share on other sites More sharing options...
HackManSD Posted July 17, 2014 Author Share Posted July 17, 2014 Wow. Just, wow. That was my thought exactly.. I really thought I was loosing my touch, couldn't figure it out for the life of me. Then bam! a small file running in the background took the issue away. It wasn't just the cards he had either as he swapped both cards for new ones and still had the same issue. We put all this hardware through a gambit of tests and this was the only thing that worked. In my searches, I have found loads of threads about the same issue.(Usually with an Asus card) I'll be spending my day trying to help the rest of the guys that are scratching their heads with this problem. We will keep it this way until I put them under water, then I will use the Skynet VBios to get around it totally. Such a limitation on such a great, and high priced, product......shame on them Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.