![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
Expensive Woes, Maybe
Jun. 22nd, 2015 07:49 pmHad some random black screens of death on the PC since mid yesterday.
Kept testing to see what was causing it, seemed to always be upon opening or closing a program, or shortly after opening. Don't think I ever saw it die on idle.
When crashing, the PSU would keep everything running except for the peripherals and presumably the hard drives. The light on the motherboard itself indicating power up until that point would still be on.
I ran a memory diagnostic and there were no errors.
Then, I reformatted every hard drive and tried installing windows on each. On each attempt with each SSD, I'd black screen when windows was just finishing, ruining the installation. So I reformatted those again and tried the standard 7200rpm HD.
That managed to finish installation. I installed my asus network drivers and started downloading google chrome. Mid-installation I black-screened again.
I had to assume at that point that the hard drives were not to blame, nor was the RAM. So it must be the PSU/CPU or MB. The three most costly items. I can't figure it's the GPU because I believe windows would still have a crash log of some sort when graphics drivers or rendering fails, or I'd get a blue screen instead - I've had graphics cards burn up before, they definitely don't do this. Plus there's the whole peripherals shutting off issue - I am pretty sure that even in the event a graphics card completely fails, the computer is still running and receiving input - there's just no graphical output to reference from.
So instead of trying to figure out what exactly I was going to need for a new build, I decided I'd do the one other thing I could do without purchasing more shit. I pulled the heatsink and unseated the CPU.
First look at it, I put way too much thermal paste on it when I first built this - it leaked around the edges a little, but not down into anything important thankfully. I cleaned that off with rubbing alcohol. The motherboard looked fine really - no scoring, no black marks, no bulging capacitors. I don't get it. The PSU was never hot, and I never smelled burnt plastic - both are signs of a bad PSU.
At any rate, I reseated the CPU (I ended up getting some thermal paste on one of the pins - I used a toothbrush soaked in rubbing alcohol to get it out.) Then put some new paste on and reseated the heatsink - I think I put too much paste on this time, too, but I'd be doing more harm than good by checking. My temperatures are somewhere between 5-10c better at idle already (never was in a danger zone anyway, but eh?)
I haven't crashed in about two hours now. Makes absolutely zero sense. Only thing left to do is start switching windows back over to an SSD, installing all my audio shit again (will take me about 2 hours) and trying to overload the damn thing to see if I can get it to crash again.
I used Prime95 to load the CPU - got it up to 56C and it held steady. Really confused about this. I've seen all sorts of hardware issues over the years, but this is a gigantic mystery at this point. Still just waiting for the damn thing to crash again so I can blame something reasonable, but if it holds up, I literally just saved myself anywhere from $200-800USD
Kept testing to see what was causing it, seemed to always be upon opening or closing a program, or shortly after opening. Don't think I ever saw it die on idle.
When crashing, the PSU would keep everything running except for the peripherals and presumably the hard drives. The light on the motherboard itself indicating power up until that point would still be on.
I ran a memory diagnostic and there were no errors.
Then, I reformatted every hard drive and tried installing windows on each. On each attempt with each SSD, I'd black screen when windows was just finishing, ruining the installation. So I reformatted those again and tried the standard 7200rpm HD.
That managed to finish installation. I installed my asus network drivers and started downloading google chrome. Mid-installation I black-screened again.
I had to assume at that point that the hard drives were not to blame, nor was the RAM. So it must be the PSU/CPU or MB. The three most costly items. I can't figure it's the GPU because I believe windows would still have a crash log of some sort when graphics drivers or rendering fails, or I'd get a blue screen instead - I've had graphics cards burn up before, they definitely don't do this. Plus there's the whole peripherals shutting off issue - I am pretty sure that even in the event a graphics card completely fails, the computer is still running and receiving input - there's just no graphical output to reference from.
So instead of trying to figure out what exactly I was going to need for a new build, I decided I'd do the one other thing I could do without purchasing more shit. I pulled the heatsink and unseated the CPU.
First look at it, I put way too much thermal paste on it when I first built this - it leaked around the edges a little, but not down into anything important thankfully. I cleaned that off with rubbing alcohol. The motherboard looked fine really - no scoring, no black marks, no bulging capacitors. I don't get it. The PSU was never hot, and I never smelled burnt plastic - both are signs of a bad PSU.
At any rate, I reseated the CPU (I ended up getting some thermal paste on one of the pins - I used a toothbrush soaked in rubbing alcohol to get it out.) Then put some new paste on and reseated the heatsink - I think I put too much paste on this time, too, but I'd be doing more harm than good by checking. My temperatures are somewhere between 5-10c better at idle already (never was in a danger zone anyway, but eh?)
I haven't crashed in about two hours now. Makes absolutely zero sense. Only thing left to do is start switching windows back over to an SSD, installing all my audio shit again (will take me about 2 hours) and trying to overload the damn thing to see if I can get it to crash again.
I used Prime95 to load the CPU - got it up to 56C and it held steady. Really confused about this. I've seen all sorts of hardware issues over the years, but this is a gigantic mystery at this point. Still just waiting for the damn thing to crash again so I can blame something reasonable, but if it holds up, I literally just saved myself anywhere from $200-800USD
