P5Q-E Troubles

[Kattivo]

I know it's considered old tech now, but it still performs great so....

Last night while changing a couple of fans I decided to stress test my system to see if the new fans were able to drop my temperatures. Upon running Prime95 core 1 failed after the third test, while the others ran strong. Finding this weird (as this OC has been stable for quite sometime), I decided to play with the voltages here and there to see if I could get it stable since I couldn't remember whether or not my last saved OC Profile was stable or not (The CMOS got reset a little while ago while i was changing a video card...battery popped out). After playing around for most of the night I simply couldn't stop core 1 from failing, so I gave up and decided to try tomorrow afternoon (today). I went on to play 3 scrims in CS 1.6 that same night with no issues whatsoever - not that it is a strenuous game for the computer to run, but I thought I'd add that in.

So I go to start up the computer today and I get a nice black screen (monitors in sleep mode, no POST) with all fans running. I can't even make it into the BIOS to try different settings. So far, I have tried:

1. Removed the motherboard from the case
2. Different PSU
3. Different videocard (6800GT that works for sure...use it to recover from bad video card flashes, lol)
4. Different sticks of RAM (one at a time)
5. Reseated CPU/HSF
6. Disconnected HDD's
7. Disconnected all USB peripherals
8. Cleared CMOS (CLRTC jumper and removed the battery along with main power)
9. Removed audio card and capture card
10. Checked NB heatsink to ensure proper contact was being made
11. Visually checked capacitors to see if any had leaked, etc.
12. Disconnected all unnecessary fans
13. Connected speaker to mobo to check for beep codes...none heard.

At this point, I think the mobo kicked the bucket (would be my 2nd dead ASUS mobo, and if it's the case expect to see the QX9650 and RAM in the BST haha), or I am missing something. One thing I have noticed is that when the system attempts to start up the CPU fan will spin at a lower RPM if it is plugged into the CPU fan header on the mobo than it does hooked up to another header on the mobo. I never had any fan controls enabled in the BIOS.

Your thoughts? Opinions? I'd rather not build a system right now since this one ran so well, but if I can't figure it out, so be it.

[_dangtx_]

try another battery, but see if the 3 yr warranty is up or not yet :)

btw you id reseat cpu too right? if the bttery doesnt do the trick let it sit all night without anything plugged in, battery included. leave the pins shorted

[Kattivo]

Originally Posted by _dangtx_

try another battery, but see if the 3 yr warranty is up or not yet :)

btw you id reseat cpu too right? if the bttery doesnt do the trick let it sit all night without anything plugged in, battery included. leave the pins shorted

I bought it in September '08 so the 3 year warranty would still be good I guess...still have the box and receipt for it too.

I did reseat the CPU (and cleaned the contacts on the back of it too with alcohol), still no luck. Right now it's sitting in my case with absolutely nothing plugged in, but maybe I should pop the battery out of it too...

If it still doesn't work tomorrow I will get my hands on another battery.

[_dangtx_]

ok, usually i glance upon the pcb and give it an afterthought. if nothing comes up in the good ole pumpkin, i go the easy way out : 'asus, another one to your collection' :)



edit : im pretty sure they don't have any in stock to replace it with, so what would you get is the name of the game

[boggtrotter59]

Maybe the cpu kicked the bucket... I know that's unlikely as cpu's rarely fail but the fact that one core was unstable may indicate this. If you have access to another cpu, I'd try swapping it out before rma'ing the board.

[_dangtx_]

it can happen, if the board dumps a load on it just before going. very rare indeed. specially with good psus and power protection.

[Kattivo]

I've had the one core fail thing happen before, usually it needs more FSB termination voltage (i used to run 1.24 and the profile had it at 1.22...which is why I thought it was failing the stress test). I've had low NB voltage do the same...

If I do decide to sell it I'll test it in a friend's board or something beforehand...I don't rip people off.

edit: I have an E4500 sitting here that I forgot I had so if it still doesn't work tomorrow I'll toss it in to rule out the CPU.

[_dangtx_]

cool, but i wouldn't try a messed up cpu in a good working board unless he doesn't care and is upgrading anyway.

i call them the black widows for a reason. my hammer has met one or two over the years. still got bits attached.

[Kattivo]

Originally Posted by _dangtx_

cool, but i wouldn't try a messed up cpu in a good working board unless he doesn't care and is upgrading anyway.

i call them the black widows for a reason. my hammer has met one or two over the years. still got bits attached.

I'll try the E4500 in my board first. If the board still doesn't work, then it's pretty safe to say the CPU is fine and the board is toast, at which point if I decide to sell the QX9650 I'll test it in another board just to be 100% sure it works properly.

[martin_metal_88]

Originally Posted by _dangtx_

edit : im pretty sure they don't have any in stock to replace it with, so what would you get is the name of the game

You could be surprised with that. 2 month ago they send me a P5K-E WIFI, they still send P5B and M2N-32 so they have a pretty large stock of old crap that probably don't work most of the time after along wait on the shipping tablet.

On topic.
I would try to get the voltage near the limit. If the core 1 still fail I would RMA the CPU before the mobo since it's always a certain core that fail. CPU are build solid but sometime after a long OC run they just can't handle it like in the first day.