High CPU Temp

[Veinres]

I noticed that there are similar cases that has been posted here but the solution has always been different, i.e., case by case scenario. Mine sometimes had somewhat high CPU usage (8~9%) and the CPU temp reached and idled around the 70s (ryzen 7 7800x3d). I tried to use Process Lasso to limit hwinfo's CPU usage, and last time the temp spiked it had around 6% cpu usage so I'm not sure what's causing it.

how do I share the log to show what's happening with my hwinfo64?

 

[Martin]

Enable the "Profiling Time" column in sensor settings, that will then tell you which sensor takes most time to read (causes higher load).
Then disable monitoring of that sensor if you don't need it.

 

[Veinres]

Currently I'm gaming so this is not supposed to reflect the usual idle spikes. But out of curiosity, is 88ms (for the GPU read), high?

 

[Martin]

NVIDIA GPU drivers are known to cause lags or spikes when software reads GPU power usage sensors.

 

[Veinres]

so you reckon reading gpu data, in general, causes the spikes? funny that while gaming the cpu temp does not even reach 60°C

 

[Veinres]

NVIDIA GPU drivers are known to cause lags or spikes when software reads GPU power usage sensors.

So I noticed that usually the spikes (I could hear it from my CPU fans) happened when my PC is in idle and screen savers is running. Not sure if that helps with anything

 

[Martin]

That could also be Windows background tasks like scheduled search/indexing, virus scanning, etc. Windows does a lot of "weird" things when the PC becomes idle..
AFAIR Windows XP was the last Windows where the system remained really idle..

 

[Veinres]

That could also be Windows background tasks like scheduled search/indexing, virus scanning, etc. Windows does a lot of "weird" things when the PC becomes idle..
AFAIR Windows XP was the last Windows where the system remained really idle..

So you reckon there's not much I can do?
I tired disable monitoring for my NVIDIA gpu, just before my last reply there reporting the spike.

But then again, the cpu temp and fans went down right away after I close hwinfo64. Any other suggestions?

 

[Zach]

Try enabling "Snapshot CPU Polling" from main settings and check what is your polling period from sensor settings




Also, what power plan are you using?
This can affect how the CPU responding to loads
Win11 has 2 separate settings

I keep this always at balanced


And changing this one sometimes


When you observe the "issue" of elevated CPU thermals while running HWiNFO how do you see it in task manager?

 

[Veinres]

What does CPU Polling Snapshot do? I don't see anything changed on the main monitoring window. I set it to this, following your post, unsure if I need to change things other than the global polling

When you observe the "issue" of elevated CPU thermals while running HWiNFO how do you see it in task manager?

I now use Process Lasso to prevent hwinfo from using all the cpu threads and lower the cpu priority. And when there's a spike, in task manager it shows "only" ~6% cpu usage. before that it was close to 10%. But I don't reckon Process Lasso helped because the spike still happens until I terminate hwinfo







Other than that, I now just set the power plan to be as you suggested on your post. I wished I could just prioritise performance since I have no concern about electricity use

 

[g9v]

I'm having the same exact issue as yours with the 7800X3D too..
im guessing its having issues reading some sensor at a random time and it just does that, i wonder if i could just log whatever that's happening :/

 

[Veinres]

I'm having the same exact issue as yours with the 7800X3D too..
im guessing its having issues reading some sensor at a random time and it just does that, i wonder if i could just log whatever that's happening :/

What's your idle temp? mine hovers around high 40s~mid50s

 

[g9v]

What's your idle temp? mine hovers around high 40s~mid50s

also about the same as yours, usage from task manager is around 8-11% when the issues occurs.
im also using afterburner 24/7 so i disabled the gpu readings from HWINFO in case its causing interference issues with afterburner, i also disabled most other readings aside from the CPU since thats what im using HWINFO for. so far the issue didnt come back but its been only like ~2 hours

 

[Zach]

Everything is in the description

 

[Veinres]

Everything is in the description

View attachment 12217

Ah didn't know there's a tooltip. But I don't see any other thing on the monitoring table? as in, there's no extra column or anything. What am I missing?

also about the same as yours, usage from task manager is around 8-11% when the issues occurs.
im also using afterburner 24/7 so i disabled the gpu readings from HWINFO in case its causing interference issues with afterburner, i also disabled most other readings aside from the CPU since thats what im using HWINFO for. so far the issue didnt come back but its been only like ~2 hours

I have nothing else for hardware monitoring except fancontrol for fan curves so I don't know what's going on

 

[Zach]

Ah didn't know there's a tooltip. But I don't see any other thing on the monitoring table? as in, there's no extra column or anything. What am I missing?

Its all about how the software is polling the CPU, respecting its low power states with minimum disturbance.
No new sensors or values. I'm using it since it first came out.

 

[Veinres]

Its all about how the software is polling the CPU, respecting its low power states with minimum disturbance.
No new sensors or values. I'm using it since it first came out.

Ah I see. So far after I set it up to do that I have not experienced any spikes yet. Even in "heavy" gaming my cpu stays at ~60°C. I'll see what happens in the next 2-3 days. Thanks Zach

 

[Veinres]

Its all about how the software is polling the CPU, respecting its low power states with minimum disturbance.
No new sensors or values. I'm using it since it first came out.

I just checked it last night leaving it on idle. then suddenly the fan became faster, I checked on task manager hwinfo's cpu usage is at ~6% and temp at ~70°C, I terminated it and the temp goes down right away. So I suppose that still does not fix it.

Anymore suggestins?

 

[VictorVG]

Well, the first question suggests itself - and the installed cooling system copes with its work? There are two options: poor thermal contact “processor - radiator” is eliminated by the replacement of the thermal interface with a more thermal pipeline - the higher the heat conductivity value and the thinner layer of the composition, the better it conducts heat; Or simply this cooling system does not cope with the load - we change to more effective. I don’t advise you to take DeepCool products from experience - their parameters are overstated at times. For example, for DeepCool Gamma Archer Pro, they promise to scatter 115 watts of heat , in reality this cooler can take no more than 70 watts. Checked on many processors. I have a SNOWMAN M-T6s Dual Fan LGA-2011 which allocate to 160 W, which is enough for my Xeon E5 2697V2 releasing up to 138 W heat.

 

[Veinres]

Well, the first question suggests itself - and the installed cooling system copes with its work? There are two options: poor thermal contact “processor - radiator” is eliminated by the replacement of the thermal interface with a more thermal pipeline - the higher the heat conductivity value and the thinner layer of the composition, the better it conducts heat; Or simply this cooling system does not cope with the load - we change to more effective. I don’t advise you to take DeepCool products from experience - their parameters are overstated at times. For example, for DeepCool Gamma Archer Pro, they promise to scatter 115 watts of heat , in reality this cooler can take no more than 70 watts. Checked on many processors. I have a SNOWMAN M-T6s Dual Fan LGA-2011 which allocate to 160 W, which is enough for my Xeon E5 2697V2 releasing up to 138 W heat.

Don't reckon you're replying to the correct post?
But just in case, my cooling system works fine. On gaming the CPU is at 60-70s