When one thing breaks everything does.. Help me make Descions please!
I recently updated my rig with 8 core and a new 750w Fatal1ty psu. Put it all in and booted it up and had no signal. So tried again and it booted. I thought this was normal when upgrading the cpu. So did some prime 95 only to find out that my h100 died. Out of nowhere.. I'm rmaing that. Slapped the loud stock fan on and powered it on . It didn't boot at all this time. So i immediately thought the graphics card. I put the card into my next pcie slot and it didn't boot right away but it still booted. I ran some fur mark on it and i immediately saw the temps were staying at 38C. So i wanted a second opinion and looked at Catalyst Control and it said i was at 111C. Yeeeeah what? So i got a third opinion that said the same thing. So i took the card apart and i saw the worst thermal paste ever.. and i also noticed the gpu looked like it it was moving around. So i redid the thermal paste put it back together and it still didn't boot properly. But my temps went back to normal. So i lived with it even though i had to take it out and put it in the other slot. So Is a new Graphics card in order or should i just try to fix it..
The card doesn't let my system boot unless i have taken the card out and put it in the x16 slot underneath my main one. If i leave it for a night it wont boot again so i have to do the opposite. I booted it yesterday the temps were spiking again. It could be something to do with how heavy the Direct Cu heatsink is weighing on the slot. It does post because the motherboard beeps 5 times to tell me i have no video card..
Actually, it sounds like the MOTHERBOARD may be the problem.
If he tries the card on the top PCIe slot and it fails to boot but works in another PCIe slot then the graphics card is working. That only leaves the motherboard (don't ask me why it "switches" back and forth).
While it sounds like a HARDWARE issue, likely the motherboard, there's two things I think you should do:
1. Flash the BIOS if an update exists, and
2. check the System RAM using memtest ( www.memtest.org )
Also, I recommend getting a $15 (i.e. HD5450) graphics card unless you have a spare one to use for troubleshooting. I had a spare card and thought my new GTX680 was faulty so I swapped it. Turns out it was my motherboard.
silversuns said:Ya but that has nothing to do with the random heat spikes. One thing i forgot to mention is that accidentally might of shook my case and that might of broken something. So what you are saying is that the pcie chip might be dead.
I'm saying that it's likely the motherboard.
You need to:
a) find another graphics card (cheap $15 HD5450, or BORROW one)
b) RMA the motherboard if that appears to be the issue.
I still recommend flashing the motherboard BIOS and using MEMTEST to verify the RAM.