Hey guys, I've posted this problem on these forums (about 2 months ago) and am cross-posting this from the homebuilt forum set. I'm trying to diagnose the problems before I begin hardware removals, and I need to fix this problem before I return back to school.
Overheating isn't an issue: Xigmatek heatsink and HAF X case.
About 3 months ago, I woke up, and my PC had shut down. I tried to start it back up, and the computer attempted to (LEDs/fans started up except for LEDs/CPU fan), then it powered down. I tried to start it up again, and 0 response.
I decided to test my PSU and used the Corsair video where I disconnected it from my Mobo, bridged the cable via a paperclip, and found the PSU working fine. I reconnected it to the Mobo, and this time my computer successfully restarted. However, the computer's clock has been reset to 1/1/2010 at 12:00 AM, 12/1/2009 at 12:00 AM, or another rollback to some other point (it changes).
Since then, it's happened anywhere from three times daily to once a week with increasing frequency, sometimes when I'm using it (minute tasks: Microsoft Word//no auto save, lost a lot of work) and sometimes when I'm asleep. Occasionally it will happen while I game, but there's no evidence that overheating would be the culprit since shut-downs occur at any given time. It's most commonly occurs when I'm not using it. Temperature readings are all fine.
Every time I have it shut down, I have to go through the same process as I mentioned above. For instance:
1. Random shut-down mid essay
2. Try to restart: LEDs/Fan briefly turn on, then off. Reattempting the start up after this has no effects. CPU fan/LEDs do not light. Eventually, nothing will light up/no fans begin.
3. Turn off PSU
4. Disconnect 24 pin PSU and Mobo; Reconnect. Try to power-on
5. Power-up works. If it doesn't, I go through 2-4 until it works again (might have to do this up to 20 times).
What I've tried
I talked to a friend of a friend that is into home-builds, and he suggested a few things:
1. Run Mem86 to check my memory (I did; 10 passes with 0 fails)
2. Look at my CPU underneath a microscope and check if anything is bent. I have not tried this. I no longer have access to one either (finished undergrad/research until September). Sorry if this is worded incorrectly, but I wasn't really sure what I'd be looking for anyways.
3. Replace my CPU with a 3 year old quad-core (I will be trying this today or tomorrow). If the problem persists with the old CPU, it's mobo. If it doesn't, it's my i7.
I've only talked to Gigabyte, and they requested that I test each part independently to isolate the issue. I don't have an extra mobo, so this is difficult to do.
What I'm asking
Last time, the only suggestion was that I was in the hands of a failing mobo. Would that or a failing CPU be the most likely culprit to this issue?
One additional question, what is the best video/tutorial for removing/reseating a CPU? My friend helped me with the install, and he took over the original seating. I've checked on YouTube and others, and it's said the X method is the best way for spreading my thermal paste. Any suggestions to the best vid?
More about :home built shut downs cpu mobo related
Logic would suggest (and I'm loosely quoting both Sherlock Homes and Spock here.. ;-) ) that when you have eliminated the impossible, whatever remains, however improbable, must be the truth. You may have to, as far as possible, swap out every component that can be removed. The main focus will be CPU, RAM, and PSU. If these are all in working order, then the mobo is the only real culprit left.
By the way, you won't be able to replace your i7 with an older quad; it won't fit in the socket.