I own an Alienware X51 R2 which came with an nVidia GTX 670. The 670 is probably a generic brand; it came with my build as ordered from Dell, and the specs according to GPU-Z match up with the default specs on the nVidia site.
My video card may be dying; one night while playing a game the screen went black, the video cards fans spun up suddenly to extremely high speeds, and then the computer rebooted itself. It did this maybe twice more that week, and some forums suggested my GPU may be overheating. I opened up my computer, blew dust out of the fans, and re-applied thermal paste before re-seating everything back in the computer and making sure all connections were good. I also started monitoring temperatures.
The random reboots continued. They were truly random - sometimes I would play for a couple of hours with no issue, other times it would reboot me within 2 minutes repeatedly when I tried to play the game. It became so frequent that I had to just keep my HDMI cable plugged into the Intel adapter since I didn't want to stress my 670 anymore.
I've seen a few posts from people with the same issue but they were typically a few years old and I've tried all of the suggestions. I have MSI Afterburner installed and I have underclocked my card significantly (-75 Core, -100 Memory) and tried several Power reductions from 95% down to 76%. I have turned my desktop Power settings to Balanced and turned off the CPU Turbo boost; I have an aggressive fan speed profile set. I have tried V-Sync in games. I have done a clean re-install of multiple driver versions. My PC BIOS is up to date. Temperatures are under the maximum; at idle they range from 40-50c, and during games they are around 87-91c. An MSI log showed my GPU temp at 79c right before a reboot occurred. Multiple 3D games I've tried (e.g. Smite, World of Warcraft) are rebooting during game play.
I have also read that this could be an issue with the power supply that came with my Alienware X51 and that it may not be able to provide enough power to the computer and GPU, or that the power supply itself is possibly overheating. The power brick is free from clutter and not hot while active; the reduced power consumption settings have helped some people but not me. The only thing I haven't tried is hooking up a separate power supply to my GPU directly.
Are there any fixes that I have missed, or any other suggestions people may have? Is there an easy way to test for sure if it is the GPU or Power supply? I don't have the funds to just buy a comparable GPU right now, and it would be horrible if it turned out the issues were not from the GPU anyway. Sadly I am outside of the 1 year warranty window as the computer is about 2 years old.
Thank you for your help and suggestions, I am open to just about anything at this point.
i have your same exact problem, i've seen your post on alienware arena forum (and i don't understand why you deleted it)
what is this problem?? we want to know!!
Hi. My post on Alienware Arena is still there (I just replied to someone else). I wasn't sure where would be the best place so I posted it to both forums.
Hopefully someone will have some suggestons for the cause and some possible solutions. It's been about a week where I haven't really been able to do any PC gaming so far and I'm open to trying different things to get it fixed.
I had a similar problem with random restarts that match your description. For me it was the RAM. Took them out, cleaned them and made sure they were locked back in properly. This fixed my problem for about a year, then again started getting random restarts. The same fix worked again, so I assume one of my rams might need to be replaced soon.
Not sure if this will help but it worked for me. Random restarts can be caused by a lot of random things, try checking if you get blue screens before they happen and track the error number down.
I haven't tried anything with the system RAM yet - thanks a lot for the suggestion. I am going to open up the computer again tomorrow to re-check the thermal paste and make sure the heat sink/fans are sitting correctly, so I will clean the RAM and re-seat them also just in case.
I'll be sure to update here once that's done.
Sorry for the late update, I wasn't really able to get back into my X51 and test it till yesterday.
I opened up the case and blew dust out of every spot I could, and also re-seated all the cables/plugs and RAM. I went ahead and re-applied thermal paste to the GPU and also adjusted the thermal gum pad on the GPU memory underneath the heatsink since it seems easy to move out of place. Also I noticed the fans seem to blow a lot of air towards the case panel that slides off, so I am keeping the side panel off for now.
Temperatures have improved dramatically, not sure if it's because the case panel is off or because I made sure the fan/heatsinks were aligned or just a combination of everything. Gaming temperatures on the GPU haven't gone over 67c after a few hours straight of gaming. The CPU temperatures are the same (close to 50c or so).
I thought I may have finally fixed it after playing about 90 minutes of World of Warcraft without issue. However.. after about 30 minutes of Smite my PC automatically rebooted itself and the GPU fan went to max again. MSI Afterburner didn't keep the log file unfortunately for when it rebooted but I played Smite for about another 20 minutes once everything was back and the GPU never went over 67c (no reboot this time either).
It almost seems like things improved but still aren't quite there. I am going to maybe let it run some synthetic benchmarks for awhile and let MSI Afterburner keep logging things to see if it reboots again.
Thanks again for any help or replies
So OP did you manage to solve this issue? I'm having sort of that problem for a year now and nobody could solve any issue with my alienware 😞 replaced my SDD, my RAM and even bought a UPS because i thought my problem could be energy supply. Still I'm running with the same issue and I'm not even able to run my computer for 15 minutes before it stops responding. No sudden reboots in my case but lots and LOTS of BSOD showing: