AW M18X R2: Black Screen after Windows 7 Splash Screen

Hey guys, not sure what happened, but last night before bed my laptop seemed to have crapped itself.

I was running FireStrike and stepped away to use the bathroom. When I got back, the screen was black. I powered off and back on and once the Starting Windows splash screen appears, it sits for about 20-30 sec then the fans all stop and the screen goes black. It's still lit up, but it won't go to the desktop.

I am able to go into Safe Mode and I have tried restoring to a previous restore point, but the same thing happens. I've googled the issue, but haven't found anything that might resolve my issue. I would try a clean install, but I'm currently deployed and don't have my disks with me.

I have run MSE, but nothing is detected. Has anyone run into this issue before?

 

I would completely uninstall the GPU driver in Safe Mode with DDU and see what happens. If you're able to get into the desktop in Safe Mode where the GPU driver doesn't load, I'm thinking driver issue. In addition to that, this happened while running a GPU stress test.

If you are able to get into the desktop without the GPU driver installed, install the driver again and see if the problem crops up again.

For some reason it feels like I've been through this same scenario. Ended up being something to do with the display driver just going haywire for no reason. I have an R1, but they're close enough.

 

After posting I decided to do a clean install and it seems to have fixed my issues. I'm now able to get to the desktop. It seems to take a while now from boot up to desktop (I've got an 840 evo as my main drive). It appears now that perhaps my secondary GPU has fried as it's not being recognized in Nvidia control panel. Not sure why it would have died. Cards haven't been OCd and I just repasted with ICD7 which brought my temps way down. Little bummed out right now.

I don't have a screwdriver with me, so I can't swap the cards. I'll probably take it to work with me tomorrow and check.

Edit: Under device manager my second 780M has a yellow exclamation point and when I look at it's properties under device status it says "Windows has stopped this device because it has reported problems. (Code 43)".

 

I have taken a couple of screenshots of GPUZ showing each GPU. As you can see they both display different revision numbers and GPU2 doesn't have the UEFI checked off. Also the BIOS versions are the same except GPU1 is in all caps and has the extra portion in perenthasis. I have uninstalled all display drivers and reflashed both GPUs with the same vBIOS. When I log back into Windows, the generic VGA drivers will install for GPU1, but fails for GPU2. When I open up GPUZ it shows both cards as being revision A2 and both have the UEFI checked off. Once I install the nvidia drivers is when everything changes.

If I go into device manager, I can disable GPU1 which will then remove the yellow exclamation mark on GPU2. I'm hopeful that GPU2 is not fried, but that's still in the back of my head. It's strange how it is being recognized by just about every program. Perhaps there's an option in the BIOS that may need to be changed?

 

I ran into a similar problem not long ago. Somehow one of my 780M cards mysteriously had the vBIOS corrupted. I cannot prove it, but I think some underhanded shenanigans NVIDIA is pulling with cancerous drivers is what caused it. I have never seen GPU firmware get corrupted during normal use before it happened to me. The only times I have seen this happen was due to a failed flash operation. Hopefully, what worked for me will take care of your issue.

Revision A2 is correct, but FF is not. Use GPU-Z to dump the vBIOS from the card showing A2 and flash that ROM to the one showing FF. That instantly cured the problem for me and the problem has never resurfaced.

That was about a month ago and both of those 780M cards are running like a Swiss watch as long as I don't use the diarrhea-780M SLI throttle-cancer drivers NVIDIA has been producing consistently since March. If you are using anything newer than 345.20 drivers with 780M SLI, stop and go back to 345.20 or 344.75 drivers.

 

I've flashed both BIOS' with the one from TechInferno. I'll try what you suggested, but I'm doubtful it'll do anything as both should have the same vBios on them. I'm leaning towards doing a clean install of Windows. I went and got a USB drive and downloaded a copy of 7. I just don't understand how everything was working just fine, then last night right before bed it decides to crap out on me. Are the 345.20 the best drivers to use? I wasn't able to install the newest drivers as it said my system wasn't compatible.

 

What happened to you sounds exactly like what I ran into. They both had the same vBIOS on them for almost 2 years, LOL. Then, suddenly, one of them had corrupted firmware. Don't have a clue why and there was no reasonable explanation for it. I was as puzzled by it as you are now. It scared the crap out of me because I thought that one GPU had died, so you can imagine how relieved I was to find that flashing the vBIOS fixed it instantly. Coincidentally, right before it happened (less than 24 hours earlier) I had tested a new GeForce driver.

344.75 and 345.20 are basically the same driver. Both are excellent. Any and every driver version NVIDIA have released after them causes SEVERE 780M SLI throttling problems on my Alienware 18, M18xR1 and M18xR2, as well as a friend's M18xR2 that I worked on recently and tested to verify it had the same behavior. The throttling happens only with SLI enabled and it occurs with the stock vBIOS or modded vBIOS. It is absolutely caused by the GeForce drivers. Some newer drivers throttle only when overclocked and other versions throttle even at stock clocks. It varies by driver version, but literally everything they have released from March forward is pure garbage for 780M SLI.

 

I'm transfering a bunch of files over the my external right now. I'll give it one more shot with the vBIOS from the primary GPU. If that don't fix it, I'll do a clean Windows install. Should I remove the drivers before flashing the gpu?

 

No, that is not required. You only need to disable the GPUs in Device Manager if you are flashing Maxwell GPUs, and then only if you are using an older (non-UEFI compatible) version of NVFLASH. However, disabling the GPUs in Device Manager before flashing may reduce the risk of something interrupting the flashing process in Windows. There is no harm in doing that, and it only takes a couple of seconds to do it if you want to do so.

Either of these driver mods should work really great for you.

Mr. Fox's GeForce 345.20 Desktop Driver Mod for Alienware and Clevo Mobile GPUs

Mr. Fox's GeForce 344.75 Desktop Driver Mod for 780M, 880M and 980M Mobile GPUs

 

The strangest thing just happened. I opened up GPUz and now both GPUs are showing correctly. Both are now revision A2, but in device manager the second GPU is still showing an error.

Edit: Now it went back to showing GPU2 having revision FF.

 

I'm not sure what's going on now. I reflashed the secondary GPU with the primary GPUs vBIOS. When it logs back into the desktop, after about 20 sec everything freezes. No inputs will work, have to power down with the the power button. I then reflashed both GPUs, same thing happens. I'm able to get to the desktop in safe mode no problem. I've uninstalled the display drivers, but the problem is still here. Looks like I'll be spending the next while doing a clean install of Windows. Fingers crossed that this fixes my problems.....

 

Looks like the clean install didn't help at all.

Before installing the Nvidia drivers, GPUZ shows both cards as revision A2. Even after installing the Nvidia drivers, if I choose to not restart, GPUZ shows them both as A2. Once I restart though, that's when revision FF shows up on the second card.

I'm gonna take it to work in the morning and swap cards to see if the card is dead. It's strange that the card is still showing up everywhere and GPUZ even shows its clocks and temps.

I'm wondering if I perhaps put too much ICD7 on when I repasted, or if something is causing it to short out somewhere, similar to the X-Brackets on the 580Ms back in the day.

 

I'm at a complete loss right now. I brought my laptop into work today and swapped GPUs, fired it up and just got a black screen then one beep. Ok, I guess the GPU is bad. Swapped them back, turned it on, and to my suprise, the same thing. Black screen and one beep. The LCD lights up, but nothing shows up on screen.

 

So anyone had the one beep on startup before? I'm guessing it's the motherboard, but not sure why it would just die all of a sudden. Any help would be appreciated.

Update: So I'm at work and decided to pull both GPUs out. It now powers up like normal and boots into Windows. I'm hoping I don't have 2 dead GPUs......

Update 2: I swapped GPUs again (original primary GPU back in slot 1) and it now boots up just fine. Would drivers cause any issues with secondary card not working in the primary slot?

Update 3: The GPU is indeed dead. At least my system is still functional, although it does feel crippled with only one GPU.....looks like I'll be in the market for another 780M, or perhaps just getting a single 980M for the time being and selling off my 780M.