Jan 18, 2014
Motherboard: AROCK H97M Pr04
CPU: i5 4690k

Problem: This started occuring about a month ago, specificially with the game Apex Legends, however rather quickly I discovered it started to happen in other games as well. At first this problem would happen in the middle of playing a game. The first time it happened while playing Apex Legends and I had been playing for quite some time that day, several hours worth. The first time it happened both of my monitors went blank, and then went blue and displayed the "No signal" and then powered off into a sleep state because there was no input to them. When this happens the game is STILL running I just have no output from the card to see anything, in fact the fans on the card still run! The only way to get out of this was to power-off my machine. This happened several times more, and would not wait until I had been playing a game to do it. Now it was just shutting off my monitor as soon as I had any load on the card at all. I tried reseating the card then, and made sure the power cables were seated well. Seemed to work, everything was fine.

Fast-forward about month till this past weekend. I'm about to play rocket league with a buddy and bam, it's back again. I don't even get into the match and my monitors cut to black and then go to sleep. I can still hear my buddy in discord and the game sounds. I am at odds with this problem. If I open any game at all the monitors will cut and I'm left scratching my head. The only game this does not happen on is Minecraft of all things.

My thoughts: I've gone into MSI afterburner and looked at the temps and compared them to my buddies rig (very similar) and there is nothing that really stands out, the temps are normally what they are under load. I haven't had problems with airflow in this case at all. I started digging a little bit deeper and went to my system event log viewer. Whenever this problem happens and I reboot and look at the logs I see: nvlddmkm stopped responding and successfully recovered. Event ID 4101. I've had this card for almost 3 years now, with really no problems at all, I'm not sure why now it's starting to act up. I'd like to avoid buying a new card because I'm a little low on cash being in college and all, but if it comes down to it, I'd like that to be the last option I needd to take.

What I've done so far: I have used DDU and completely unistalled drivers related to Nvidia. Reinstalled and was still having issues with it. I have also did a reset of my windows in hopes that maybe it was a windows update that was conflicting with something. No dice there either. I saw another recommendation on another thread that maybe I should update my BIOS. So I tried that as well, still nothing seems to help. I have reseated the card, tried different PCI-e cables from the PSU. I've got rid of all dust in the case (really nothhing to note there, there was not much.)

Extra: I've already tested the monitors to make sure that they were not the problem. I hooked one up to my motherboard and can run it fine off of integrated graphics. I did this and tested them both this way. No issues.

This has been very frustrating. Any help would be very much appreciated.


nvlddmkm stopped responding and successfully recovered. Event ID 4101.
The problem related to overheat, hardware and driver.
And you did use the DDU so that the driver is fine, also the BIOS and the temp are OK too.

So the problem related to the gtx970 or the PSU.
For the PSU: 1) go into the BIOS, in pc health monitor or hardware monitor section ( may check the MB manual) to check the +3.3V, +5V, and +12V to see they are within +/- 5%, like the +3.3V, it should be within +3.1V to +3.4V. If they are within the +/- 5%, the PSU is fine. otherwise the PSU has problem.

For the GPU: If the PSU is fine, try to use the MSI, 1) set the power limit for the gtx970, like 80%, 90%. Or 2) set the gpu voltage little bit high, or 3) under clock the gtx970 little bit, those will help the gtx970 runs more stable. If those don't help, try other GPU, because the gtx970 may have problem.