Best temperature monitoring software

bazzabear

bazzabear

bazzabear

Soldato
Joined
2 Nov 2013
Posts
4,372
Can anyone recommend a good temp monitor?

It really needs to be one which will keep a log. My computer is suffering sudden shut downs, so if the monitor doesn't write to a log, I won't see whether it has overheated just before the shut down.
 
Thanks!

Using that, it's reading my max CPU temp as 100C while playing Rocket League (a game it has never crashed during), about 80C when not running a game. I haven't had a chance to check during a game it has crashed on (Beat Saber, Borderlands 2) to see how hot it goes then.

I'm not clued up on these things, but that seems hot!

On Intel's page for my CPU (i7-8700K) it seems to be saying that's the maximum allowable temp ("T-junction"), so this does maybe suggest overheating is the reason for my woes. Can anyone more knowledgable confirm?

If so, I guess this has turned into a hardware question. Fans and AIO cooler pump all seem to be running fine, so why is it getting too hot?
 
Hello,
100C is way too high and it is the maximum allowed temperature by Intel spec before automatic throttling to preserve the CPU.
First thing to ckeck is the voltage which has the biggest impact on the temperature level.
If it's normal (around 1.25v during load) then the problem is probably with your cooler.
Try to remount it after applying some new paste, or try another cooler if you have one.
I have myself a 9900k on air OC at 5Ghz all cores (without AVX offset) stable at 1.25v, and temp is around 60C during games, so your temp is definitely not normal...
 
Voltages are showing a low of about 1.3 and a high of about 1.43 across all lines, is that OK? Re-mounting sounds like a good idea as a quick and cheap way to rule out a possible issue. I shall try that at the weekend.

The temperatures do seem to be showing spikes that last a fraction of a second. Having reset the min/max readings, if I sit watching the readouts, I never see a current reading out of the 30s, but the max is still managing to read at 60C after a few mins.
 
How do I get it down? The computer isn't overclocked (unless it has some kind of automatic overclocking through the bios).

I'm looking at the data sheet now, but it's 134 pages and I haven't a clue what I'm looking for. :)
 
It's something you set in the BIOS called CPU Vcore. Motherboards have a tendency to apply too much voltage when set to Auto (the default).

A reasonable strategy might be to put in 1.28 and see if it's stable. Then reduce by 0.01 increments until it becomes unstable, then go up 0.02, and leave it there.

You should watch some overclocking videos with the same motherboard manufacturer as yours so you're familiar with navigating around the BIOS. They tend to be aiming for achieving as high an overclock as possible, but your goal is to get your temperatures down so it's a bit different, but it's the same settings you change to achieve it.
 
Thanks very much for the help. It's an Asus motherboard and they seem to use non standard names, but I seem to have managed to find the right setting, and HWMonitor is now reporting VCORE at 1.28v.

Since then, highest temperature listed under the CPU is 82, and I've had no crashes playing Borderlands 2 (previously tonight, it was crashing every 5 mins when trying to play).

Still weird that this has only just started to be an issue. (Well, it has previously crashed very occasionally, but only now started to do it very frequently.) And is 82C still rather high - just not high enough to crash it? So maybe still worth re-seating the cooler.
 
bazzabear said:
Thanks very much for the help. It's an Asus motherboard and they seem to use non standard names, but I seem to have managed to find the right setting, and HWMonitor is now reporting VCORE at 1.28v.

Since then, highest temperature listed under the CPU is 82, and I've had no crashes playing Borderlands 2 (previously tonight, it was crashing every 5 mins when trying to play).

Still weird that this has only just started to be an issue. (Well, it has previously crashed very occasionally, but only now started to do it very frequently.) And is 82C still rather high - just not high enough to crash it? So maybe still worth re-seating the cooler.
Click to expand...
Good! That's better:)
Yes you should do both: continue to decrease your voltage while it's stable and reseat your cooler with good quality paste.
Temperature below 80c is safe for long term.
You are almost there.
 
Great, I can carry on experimenting with voltage throughout the week then re-seat the cooler at the weekend.

Thanks again, to everyone in this thread - you've been absolute stars! :)
 
It's a Coolermaster MasterLiquid Lite 240.
I wouldn't necessarily be averse to replacing it if there was something that would definitely give better cooling performance. But the case doesn't have space for a longer radiator.
 
How are you powering the pump and fans? Make sure the pump is on a 12v supply. If it's a motherboard fan header, it could be running at a lower speed that it should.

Something seems amiss, is it mounted correctly? Thermal paste? Fans running? Pump speed..
 
Last edited:
Pump speed is showing up (from memory) at just over 2000rpm. Fans are running. (Before lowering the voltages, I used the BIOS to set them all to 100% constantly to try to control things. Now turned down to 'turbo' setting instead.)

I didn't manage to re-seat the cooler after all this weekend, so I can't guarantee that it's seated perfectly. Still to be done.

The computer is sitting in a cupboard within a desk which probably doesn't help. But the cupboard is open backed, has a drawer sized opening at the top at the front, and has plenty of airspace around the PC.

I have had no crashes at all since the voltage change. But I'd agree it still wouldn't hurt to reduce the temperatures a bit further. Voltage now set at 1.25v, and temperatures rarely going above 70, but a very occasional spike to about 80. I'm not sure whether to continue going lower than this.
 
Ok.

If you're not overclocking, I'd just set the voltage to b default/auto.

Location of pc sounds far from ideal..

Would be fine for an office spec pc, but bring a gaming pc I'd either have a radiator outside of that area if watercooling or mount a 140mm fan to the front of the drawer/cupboard.
 
The default voltages were the problem. The BIOS was setting them at not much less than 1.5v and leading to it running much hotter. It's moving it away from auto which has solved the crashes.
 
You must log in or register to reply here.
Back
Top Bottom