Using Core Temp itself fixed my problem, explain?

[NoBugsOnMe]

In short, after running Core Temp in logging mode in order to record the temp at point of failure, suddenly the problem was gone.

Longer story, for the past 3 months I was experiencing crashes consistently, multiple times a day, when there was a heavy CPU load, the PC would shutdown and start a reboot. The reboot would get stuck in the initial diagnostic boot page, I would have to unplug the power supply and let the PC rest for at least 5 minutes. If I rushed it and restarted under 5 minutes, the PC would be extra-susceptible to shutting down during CPU usage. I tried dusting with an air compression can, removing and reseating the RAM sticks. My theory was that the PC wasn't able to draw enough power from the power bar, so I unplugged everything else other than the monitor and PC. I looked through my PC drivers, updated old ones, removed unused ones, cleaned registry, virus scanned. Nothing worked.

I thought that since the crashes happened during heavy CPU load, it might be related to high CPU temps. So, I installed Core Temp to log the last recorded temp before a crash and loaded up the video game that had always led to a crash (X-Com 2, which btw had taken 8 hours to successfully install, but at relatively low CPU usage) and waited for a crash, particularly at a character loadout customization screen. And waited. I played and played, 24 hours and then 36 hours passed during which I used the customization loadout screen dozens of times. I also slept during that time and after waking up it still hasn't crashed.

So, the question I have is, what could Core Temp might have effected that solved the problem? Has this been reported to have happened before (I read a post about Core Temp temporarily causing 3 old CPU's to start working again)? Could it have updated some value that was corrupted, incorrect or missing? Using the Task Manager, I opened a Resource Monitor and looked at Associated Handles and Modules that Core Temp was accessing, other than a Bonjour and a Logitech Gaming Software dll I didn't see anything unusual.. maybe one of those is the problem? As a sidenote, this whole situation reminds me of the Observer Effect in quantum physics, how the attempt to observe an experiment itself causes the experiment to change, at least it's been for the better in this case lol

[The Coolest]

This sounds very much like a power supply that is on its last leg. If this is a PC (not a laptop), then the PSU would be the first thing I would check if I had similar issues.
Another cause for this sort of behavior in my experience is bad capacitors on the motherboard. When the processor is under load, it draws a lot of power from the voltage regulators by the socket, they depend on properly working capacitors to supply stable voltage and enough current to the processor. If the caps are bad, this can also cause random reboots or freezes.

[NoBugsOnMe]

This sounds very much like a power supply that is on its last leg...

Yeah, I'm sure you're probably right, but then how could Core Temp fix or compensate for a faulty PSU or bad capacitors? I don't experience random reboots though, they only previously happened during high CPU usage. Still no crashes yet since installing Core Temp, fingers crossed.

[The Coolest]

I doubt Core Temp has had anything to do with this, it's probably a coincidence.
Try running something that puts a lot of stress on both the CPU and GPU and see what happens.

[NoBugsOnMe]

I doubt Core Temp has had anything to do with this, it's probably a coincidence.
Try running something that puts a lot of stress on both the CPU and GPU and see what happens.

I've been doing that for the past 2 days, playing XCOM 2 which had previously crashed it 100% consistently at the character loadout screen. I doubt it's coincidence, which is why I wrote the first post.

[meimeiriver2]

The reboot would get stuck in the initial diagnostic boot page

This is all you needed to know to realize Core Temp couldn't possibly have anything to do with your problems (as it isn't hooked into your BIOS, of course, hence cannot affect anything you would do/see at your diagnostic screen).

Likely a bad PSU/caps indeed.