I'm writing in here because I just had a very weird experience.
When entering Windows, the CPU temperature was running constant at high temperatures (81-87)... when I first saw it my thought was that it was a bit excessive, but maybe it was just working "harder than usual" on starting everything on booting Windows.
But the temperature stayed there long after it was done booting... It wasn't even (really) working, as it was hovering in between 0-10% use all along, and from what I remember the power draw didn't go above something like 20W.
Unfortunatly I do not have screenshots of any of this, as I was scurrying around different monitoring programs to see if I could find something...
It confuses the heck out of me on how it can even push it's temperatures this high, without utilising the CPU, especially without real power-draw...
After a while, it just "stopped being weird", and went back to it's usual Idle temp around 40-45 (which I've also always thought was high, but the IHS is known to be... problematic)
All this occurred after a sudden black-screen and restart, while I was just sitting in Google Chrome and messing around with Foundry (Virtual Tabletop for games like D&D).
To give you a reference for cooling, I'm on a custom watercooling loop, so it should not be an issue whatsoever...
It's sharing a 360x60mm and a 280x30mm radiator with my RX 7900XTX, and has 4 120mm fans besides the ones on the radiators, creating airflow in the cabinet.
I'm normally running this for "absolute silence", but I immidiatly pushed my fans and pump to 100%, when I noticed the temperature...
It did not change anything.
Plus...
Considering it was just randomly temp-spiking without powerdraw and load, kinda rules out the cooling issue.
Solved! Go to Solution.
I've had this happen on a Gigabyte board. For some reason it was Ryzen master, some bug that keeps the voltage high to the cores.. uninstalled and reinstalled ryzen master then reconfigured my settiings. My issue was it locked my Core voltage for some reason and wouldn't let the clocks come back to a lower idle clock. I've also had it were 6 cores was used instead of all 12 couple months ago. LOL..
lordcrisp, I do not think this is weird much less very weird. When Ryzen processors are give work to do by the OS, the processor goes full bore to get it done. I suspect the real culprit here is your
"scurrying around different monitoring programs".
Please uninstall all the multiple applications, install Ryzen Master (RM) and post a screenshot of RM running Cinebench R24 Multicore. Image, of CB only. John.
Did you not read my post?
It wasn't doing anything...
It wasn't heating up when I opened things... I opened things because it was running at almost temp-threshold, without doing anything real.
I noticed the temperature on my motherboard debug LED, and jumped into Adrenaline to see if it was saying the same...
And then I opened Fan Control to spin up my fans to try mitigating it...
I know it might have sounded like I opened a million monitoring programs, but that was not actually the case...
It's not like I didn't trust what my motherboard and Adrenaline told me... opened Adrenaline for more information than just the temp from my motherboard.
It was doing nothing and was just running at very high temps...
When playing games it's usually sitting around 65.
Of course something like Cinebench can push it to the temp-limit, but that's kinda what it's made for...
It was not normal behaviour.
I've been having this system for half a year, and use it daily... and been building my own (and others) computers for 20+ years.
I could run Cinebench and screenshot Ryzen Master for you, but as I said... it went back to normal behaviour, so you won't see anything out of the ordinary.
Edit:
Realized you'll just request the screenshot anyway, so here it is...
And I even forgot about this... it's not even running at temp-thresh when doing that.
So... the CPU was running at a higher temperature, without doing anything and not drawing any power, than when it's running CPU benchmarks.
I've had this happen on a Gigabyte board. For some reason it was Ryzen master, some bug that keeps the voltage high to the cores.. uninstalled and reinstalled ryzen master then reconfigured my settiings. My issue was it locked my Core voltage for some reason and wouldn't let the clocks come back to a lower idle clock. I've also had it were 6 cores was used instead of all 12 couple months ago. LOL..
Thank you, that actually makes sense.
When I opened Ryzen Master to give misterj his screenshot, Ryzen Master crashed and I couldn't get it running. I exactly had to remove and reinstall it.
Here's to hoping that is all it was... although it's very concerning if that is the case... if Ryzen Master isn't more bulletproof than the occational "I'll just fry your CPU now for no reason" situation, I might not want it installed at all...
Everything I do "with it" I do in the BIOS anyway... Only used Ryzen Master to keep an eye on behaviour.
And on a somewhat different note.
Even though they're the ones I've had the least trouble with throughout the years, I'm not on Gigabyte (running MSI this time).
So if you were scared it was Gigabyte's fault, I think/hope you can stop being concerned with that particular part of the problem.
Lol yeah i heard its been buggy on just about all MBs. Having voltage issues or higher temps from RMaster isn't a huge deal since the processor will just shut down.. Those safe guards do work.. Wonder how I know as my AIO pump died few weeks ago on a 5900x haha..
Thanks, lordcrisp, I have seen no evidence that the processor was not doing anything. The multiple applications as well as Windows was doing many things, probably causing the temperature rise. If we could see RM at this time I think it would be very clear. The next time this occurs, please post a screenshot of RM. John.
The 7000 series processors commonly keep the temps right near the maximum as it's how they are programed so your system is fine and operating as it should. I always recommend liquid coolers for these chips whenever I build them for someone and the liquid cooling loops for these processors are fairly inexpensive on amazon even though there not required.
Basically just gonna refer to my answer to misterj.
It's not normal behaviour... it's never done this before...
They only run to the limit, if they need it...
In games I'm usually around 65 degrees.... I only really ever see it being pushed to the temp-threshold from benchmarks.
I know a lot of people get weirded out by how Ryzen handles their clock/temperature balance (and that also being the reason why undervolting is better than "normal OC").
But that is not what is going on here.
lordcrisp, You are probably correct in saying
"It's not normal behaviour... ".
A much simpler and more likely explanation is that the cooling system did not start due to clogged loop or fan/pump control problems. This also explains the black screen. If the pump stops, the processor overheats and triggers PROCHOT quickly. I never trusted the MB fan control and run all my fans/pump on 12 Volts straight from the power supply. John.
Just to clarify your OP.
Are you using a single Custom AIO cooling system connected to both your GPU and CPU hardware or Custom AIO separate for each?
I have read that connecting a single custom AIO liquid Cooling system to both the GPU and CPU may cause cooling issues with one or both connected hardware.
For instance, the GPU output cooling tube is connected to the input of the CPU processor. Thus it is getting warm liquid to the CPU from the GPU output tubing. as an example.
Both of your GPU and CPU consume a lot of power.
But if you are connecting the GPU and CPU tubing separately to the Radiators it could be the position of the Radiators or the flow of the liquids from the GPU and CPU to the radiators which might be causing heating issues.
Sound like a cooling issue with your custom AIO since it is running almost at overheating temperature. Maximum Operating Temperature for your processor is 89c with a TDP rating of 120 watts which requires a powerful AIO CPU Cooler.
If you do have both the GPU and CPU connected to the same custom cooling system try to separate them, temporarily, to see the there is any change in the CPU temps.
I never said anything about AIOs, I said custom watercooling loop.
As in... custom build.
Running a D5 pump, with 2 big radiators (sizes also mention in OP), and waterblocks on my GPU and CPU, all in the same loop.
And yes, many people say that "the order of hardware" in a loop makes a difference, but it doesn't really...
Jayz2Cents has talked about it (and shown it) on several occations.
There might be a slight difference, but it's insignificant, if there at all.
It's an "urban legend" that seems to make logical sense, so it's easily spread everywhere. I also personally believed it until Jay got tired of hearing it and made a video on it (ages ago).
And as I've also already reitterated in several other replies, it wasn't doing anything outside of booting into Windows, and afterwards just being idle, except for me opening AMD Adrenaline to see what was going on, and Fan Control to see if my fans or pump was doing something weird, and then afterwards spool all of it up to high-speed, to try fixing the temperature (which didn't help at all)...
And as I've shown in a screenshot to misterj, even running Cinebench R24 it doens't normally get as hot as it did for that short weird temp-spike.
I know you are not using a regular AIO but a Custom water cooler system.
So, I let MisterJ take over in getting your issue fixed with his excellent troubleshooting and computer knowledge. (-:
I strongly advice you follow MisterJ troubleshooting methods for the best results. Do exactly as he says so that you will be able to fix your problem.
Take care.
Have a look at this it's very interesting Everyone!!! https://hardwaretimes.com/amd-ryzen-7000-cpus-are-built-to-run-at-95c-24x7-without-affecting-lifespa....