My GPU Died and Took The Whole Computer With It
A couple of weeks ago, or days, time has lost all meaning to me, I wrote about how surprisingly difficult it was for me to switch to Linux. It essentially required me to switch from an NVIDA GPU to an AMD one. I picked one up on eBay for $190 about a month ago and until yesterday it worked perfectly.
Until yesterday.
I turned on my computer and was greeted by a black screen. Turned it off and back on again, still a black screen. After many deep sighs, I got to work. I was pretty sure it was that GPU, I had bought it used, after all, but I hoped it was something else. Anything else, to justify everything I had went through. The CPU could be a smoking crater, the motherboard could be a smoldering heap of scrap, the RAM sticks could be fighting each other to the death in a cage match, all would be preferable than the GPU failing. This used graphics card that I knew was a huge risk when I bought it, but did so anyway because I just wanted to use Linux.
But no, it was the GPU.
I plugged one of my monitors into the HDMI out on the motherboard and it worked fine. Well, there was another problem we’ll come back to, but I was at least getting an image. I have an older machine lying around, so I plugged it into that and again, no dice. The card is showing no signs of life – no lights, the fans aren’t spinning up, and it did nothing when I showed it a picture of puppies. It’s a dead gym, as Spock always said in Star Wars.
I contacted the seller and have yet to hear anything. I’m sure I’m not going to, but we’ll see how that goes.
However, another even more significant issue has reared its ugly head. One that in all my years of working on computers I have never encountered before. Sans GPU, I went back to my desktop and tried to turn it on. Turning my back to look at some cheese or whatever, I was surprised when I turned back around and saw a Windows 10 recovery screen. After staring at this screen for sixty seconds trying to process what was happening, I restarted the machine.
This time I was paying attention and went into the BIOS. For some reason, it’s not detecting my M.2 SSD as a bootable drive. It’s listed as a peripheral, so I know it still works and that it’s being detected in some capacity, but I can’t boot from it. Instead, my machine was trying to boot from an old SATA SSD I’ve got plugged in, (which apparently I never deleted the Windows 10 boot files from, but that’s an issue for another day).
I spent hours trying to figure out what was going on until I stumbled upon a post somewhere telling me that I should disable secure boot and enable CSM, which should allow the motherboard to detect the M.2 drives as bootable. That made sense, but it occurred to me that if I was booting from this drive before than that shouldn’t be the fault. Surely the GPU not being present wouldn’t cause CSM to disable or anything, that’d be silly!
I checked the settings and sure enough, secure boot was enabled and CSM disabled. I changed the settings, rebooted… and nothing. I checked the BIOS again and the CSM setting was still disabled. The BIOS wasn’t saving the change. I looked this issue up and, somehow, it has something to do with the Intel CPU I’m using and its integrated graphics. I’m not going to pretend I understand what’s going on here, but from the help I found online, CSM cannot be enabled without a GPU because it doesn’t work with Intel’s integrated graphics.
So to sum up: my GPU died, which told my BIOS not to recognize my M.2 drives as bootable, which means I effectively can’t use my computer at all right now. I hate computers so much.
I’m going to dive more into this when I have some time off work, and when I’ve had the chance to calm down. I have to keep telling myself this isn’t an absolute disaster, only a partial one. The drive still works as far as I can tell, I haven’t lost any of my files since everything was backed up onto an external hard drive and the cloud, and I’ve got a laptop I can use in the meantime. The only things I can’t do on it are play games, which I haven’t been playing many games lately anyway, and edit videos, which is a slightly bigger deal since I was in the middle of working on one.
I’m in a holding pattern right now. I want to dig more into this to make sure this CSM/GPU thing is the problem. I’m waiting to hear back from the eBay seller about the graphics card (though I’m sure I’m screwed), and I’m looking up new GPUs I can buy that won’t break the bank, but I can’t afford one anytime soon since I spent so much on this dead one. Repairing it is far beyond my abilities, if its even repairable. For now, I’ll have to make do with my laptop until whatever happens happens.
What can we learn from this? For starters, if you think something isn’t a great idea… maybe don’t do it? I should have been more patient, saving up for another month or two so I could buy a brand new card. I don’t need anything top-of-the-line, so I wouldn’t have had to spend too much more. Newegg has a 7700 for $400 right now, twice as expensive but brand new, under warranty, and better than the 6600 I got.
And I’m going to get hate for this, but you should seriously consider if you really, truly want to run Linux. I think about all the problems I’ve had trying to switch over, and while I still believe it’s worth it to not be spied on or have malware randomly installed, maybe a Mac would be a better alternative if you’re not willing to go through this struggle. I’m sure I’ve been massively unlucky, but any Linux install will be harder than simply getting Mac. I believe it's only a matter of time until Apple pull the same shenanigans, but still. I don’t blame Linux for this, by the way. It’s a stupid turn of events that could have happened to anybody.
(I know I said today was going to post the blog recommending Peertube channels today, but, well, yeah. That blog will be up tomorrow.)
