[SOLVED] Older PC reboots under certain loads.

[Fenric]

Old i7 4790k and 980Ti I built years ago. I've been lazy and assumed the CPU is overheating and needs a thermal paste change, but after installing SpeedFan, the highest core temperature reading I saw before the system rebooted was 62C. I don't think this is high enough to trigger a thermal protection reboot is it?

This only happens when I use a GUI frontend for ffmpeg/avisynth to convert video files. CPU usage can spike to 100%, and sometimes (more frequently lately) the system reboots. Event Viewer just says the system restarted unexpectedly.
Other than this I haven't noticed anything, even when running a game and recording or streaming via obs/discord.

If the core temp. isn't getting really hot what else should I be looking at? Failing power supply? It's got a EVGA 750W 80+ Gold PSU. I'd like to figure it out so I don't unexpectedly have a dead desktop later on with nothing to use.

 

[CountMike]

Old i7 4790k and 980Ti I built years ago. I've been lazy and assumed the CPU is overheating and needs a thermal paste change, but after installing SpeedFan, the highest core temperature reading I saw before the system rebooted was 62C. I don't think this is high enough to trigger a thermal protection reboot is it?

This only happens when I use a GUI frontend for ffmpeg/avisynth to convert video files. CPU usage can spike to 100%, and sometimes (more frequently lately) the system reboots. Event Viewer just says the system restarted unexpectedly.
Other than this I haven't noticed anything, even when running a game and recording or streaming via obs/discord.

If the core temp. isn't getting really hot what else should I be looking at? Failing power supply? It's got a EVGA 750W 80+ Gold PSU. I'd like to figure it out so I don't unexpectedly have a dead desktop later on with nothing to use.

Use another program like HWinfo or CpuId HW monitor to check for temperatures, SpeedFan is slow to react (it's obsolete) and a second two of overheat is enough to shut CPU down.

 

[Fenric]

Use another program like HWinfo or CpuId HW monitor to check for temperatures, SpeedFan is slow to react (it's obsolete) and a second two of overheat is enough to shut CPU down.

With HW Monitor I saw 65C for a split second, mostly stayed around 62/63, also restarted again. It seems to be restarting every time I try now.

Edit: I never felt the need to overlock this system, but I recall looking into undervolting. I checked the BIOS just now and everything is still set at default values.

 

[Fenric]

Well I thought it was the PSU fan, but I forgot this one has an Eco mode switch, which lets the fan shut off if the PSU temps are low. Fan works fine.

Anyone have any suggestions on other things I can test?

 

[Fenric]

Thanks to the wonderful support team at EVGA I was able to troubleshoot and determine that the #1 RAM slot on the motherboard is failing. I'll switch to slots 3 and 4 for now and see how much more I can get out of this rig built in 2015.

I took out the GPU and the slot 2 stick of RAM and didn't immediately get a restart under load, but I didn't test it long enough apparently. I assumed the RAM stick itself had gone bad, and by chance had some new unopened RAM to replace it with. After putting the 2 new sticks in, it restarted again under load, so I switched to slots 3 and 4 and haven't been able to get it to happen again.