PC freezes under load (usually)
 

PC build:
CPU: AMD FX-9590 (not OC) running at base 4.7 Ghz (Currently have it cooled with Thermaltake 3.0 water cooler and liquid metal paste instead of artic silver).
Mobo: Asus Sabertooth 990FX (Most recent BIOS 1604)
GPU: MSI GTX 1070 seahawk, water cooled
RAM: 16gb of 4gb crucial balistix
PSU: 800W silent pro GOLD
Hard Mem: SSD 850 EVO

Here's my problem... my PC freezes. It does not shut down or reboot due to hitting temp cut offs, it just freezes. It usually does this under load, but not with every game. For example, I play Paragon, and it often freezes with this game, but I play with Witcher 3 on ultimate with 4k mods, and nothing. It has rarely frozen when not under load, but it has happened.

The GPU is new, and the same problem happened on my GTX 970, so I know it's not that. I have recently replaced two of the RAM sticks, and they are all matching frequency and type. (Mind you it could be the two older sticks, or the MOBO RAM slots). SSDs are new, and the same problem happened on my older HDD.

My only thought is it's the motherboard, or the PSU at this point. I have run many tests, including Prime95 torture tests for 8 hours. I never get an error. I have run the tests multiple times. However, I ran the "Blend" tortue test that uses a lot of RAM and the PC did freeze, but reported no errors. I'm in dire need of help here for anyone who can help me.

Try [URL="http://memtest86.com/"]memtest86[/URL]

If it fails the tests, try 1 stick at at time until you find the culprit (assuming it is not a memory socket problem.)

:smile:

Any way to discern if it's the socket? I have my two old sticks running now. Will test the two new ones next.

Not sure if your system will boot without a stick in socket 0. If it does you can take it from there. :smile:

It's running memtest as we speak.

What operating system? Assuming MS Windows, are there any events in the event log (control panel -> administrative tools -> event viewer -> windows logs -> application/system) after rebooting around the time it freezes?

When it freezes, I assume you mean that all input is totally unresponsive, but it is still displaying whatever was up at the time it becomes unresponsive?

While frozen, have you tried unplugging the mouse/keyboard, and re-plugging them back in, or if you have a optical (cd/dvd) drive, opening that, inserting a disc, and closing it. Those events might trigger the operating system into further action depending on why it is frozen.

Win 10, and the problem also occured with win 7. Yes the PC is completely frozen and unplugging the mouse does not do anything. I have yet to try the Disc drive. I will search for any event log error, but when I've checked in the past I couldn't find one.

I pushed my PC today with only running two RAM sticks at a time, using different slots and no crashing... Could my mobo not be setup properly for using all DIMM slots?

I'll post event logs in the morning and BIOS data for voltage with RAM

This is a long shot, but recently happened with a friend's gaming machine:

I can't figure out why or how, but the monitor was causing the hang.

He brought me the system (without the monitor) and I went through it heavily with its installed Windows OS and with an alternate install of Linux. I could find no failures. There was a lot of dust, etc. inside, so I cleaned it and even resinked the CPU. Since I had no failures, I gave it back and it immediately started failing again for him. Someone else suggested the monitor and he switched to another and it hasn't failed since.

I can't see why/how the monitor was the trouble, and mention it only because I've witnessed it via my friend. He said it would even hang when switching scenes, even when not gaming, simply using the web browser.

I recently upgraded my monitor, same problem so I doubt that 😥 ty for the info though

.

After you have convinced yourself that neither RAM nor RAM sockets are the problem and you can run memtest86 with 4 sticks for 24 hours say, install some temperature monitoring software -- it maybe that the board's bridges chips are getting too hot.

Next try with turbo turned off in BIOS. Then try underclocking the beast. Also try upping the memory voltages a very tiny amount -- you don't what fried chips. Ditto the CPU.

Diagnosing the problems could take weeks, only to find you need to replace the motherboard or some other component.

Does the case have excellent cooling? :smile: