First, some quick specs on my homebuild:

Rosewill Case (cheap)
Antec 480W PS
Gigabyte p965-S3 Mobo
E6600 cpu
NVidia 7900GS video card
Seagate 500GB Barracuda HDD
Corsair XMS2 2GB RAM

So, I turned off my PC for the first time in a while last night and when I later tried to turn it on, it began to turn on for 4 secs, then it turned off, then 4 secs later, it turned on again, then 4 secs later, back off. It is turning off so quickly that it is not even reaching the POST. So, I disconnected everything piece by piece until i was only left with the CPU and tried to turn it on and it still was not stable. Also did not hear any Missing RAM beeps as there should have been. So, I checked the PSU on a diff compy and that worked. Also, I then went and put in a diff 965P mobo that I had and it turned on for 4 secs, turned off for 4 secs, and then turned on and stayed on and went all the way into my OS before I decided to shut back down.
Now I'm really confused, b/c if it were my mobo then replacing it should have fixed it. If I had fried my CPU (it is overclocked) then replacing the mobo shouldn't have fixed it (which is sorta didn't). A buddy suggested maybe it is the power switch itself but that seems a bit weird to me. Any help? Please let me know any thoughts at all!!!
Thanks!

Disconnect your hard drives, dvd/cdroms, and other peripherals devices. Run your computer with the following:

a. processor
b. motherboard
c. one memory module/stick
d. video card
e. power supply

Boot your computer and check if the problem still exists.

Soon-to-die memory can eventually cause behavior similar to over-overclocking. Also cheap or old PSU can do the same. (lack of voltage or watts)
BUT... 4 seconds delay points to ... (sounds silly, I know) stuck power button. Usually if you press and hold the power button for 4 sec, you will shut down your PC. Try this:

- change power-off behavior in BIOS. It is usually Instant OFF/4 sec delay/ do nothing . If available, select "do nothing" (I think those are the options)

- Increase memory voltage to the max if not already done. Corsair memory should be equipped with heatsinks, so no overheating worries.

- Increase CPU voltage to the max if not already. Be aware of the CPU temperature once you do that. That heat could damage your CPU.

- If all that fails, try setting core clock at lower rate.

Also, check other voltages (12v, 5v, 3.3v) in BIOS and/or monitoring software in OS. One of those could be too low. That would point to dying PSU.

hi, you do what chaky said check your PSU, make sure also that there is enough thermal compound on your processor and heatsink

Be a part of the DaniWeb community

We're a friendly, industry-focused community of developers, IT pros, digital marketers, and technology enthusiasts meeting, networking, learning, and sharing knowledge.