[SOLVED] New PC builder, can someone check my 9900K temps? Idle and 100% load stats inside!

[shagann96]

Hi all, thanks for looking. Build is;

Case - Lian Li PC-011D
Mobo - Z390 Aorus Ultra BIOS ver. F9
CPU - i9-9900K (stock)
CPU Cooler - Corsair H150i (3 ML120 fans, top mounted, exhaust)
Memory - 32GB DRR4-3200 Corsair Vengeance RGB Pro (XMP enabled)
GPU - RTX 2080 Super FE
Storage - Corsair MP600 1TB, Samsung 860 EVO 1TB
Fans - 6 ML120 Pro fans, side and bottom mounted, intake)
OS - Win 10 Home

Idle Stats - Individual Core Temps & Individual Core Voltages

https://imgur.com/a/HY3wiNZ

View: https://imgur.com/a/HY3wiNZ


So from what HWinfo reports, I think an average idle temp of 30-32C isn't too bad. Core voltage seems to be okay too, I see a lot of people on various forums saying the absolute maximum is 1.35V and I'm quite a bit away from that.

Can someone take a look through the Imgur and point out anything concerning? Outside of temps and voltages, I'm really not sure if there are any other red flags to watch out for.

Stress Test Stats - Prime95 v29.8b6 - Small FFTs - All AVX Disabled

https://imgur.com/a/G0HecBc

View: https://imgur.com/a/G0HecBc


I had AVX disabled in this test, which I ran for just over 1 hour. I read online that Prime95 generates unrealistic CPU load, so I am taking these readings with a pinch of salt.

Temps are obviously a lot higher here, and throughout the duration of the test the package temp increased from 66C to 71C, which I think is to be expected. More importantly, are these good temps for a stock 9900K? I read conflicting opinions online of a) ~70C is fine, 9900K runs hot and b) a H150i should be able to cool it better than that. Some clarity here would be nice.

Voltages look okay, but again, I don't know if the obvious increase from idle is too little, too much, etc.?

Supplementary Stuff

All case fans are running at 1000RPM, H150i on the 'Balanced' preset in iCUE, during all tests. I have done some brief experimentation with fan curves but, honestly, it doesn't seem to make a lick of different to my temps, at idle or stressed. More experimentation required here on my part.

I don't know my ambient temp, but I live in Ireland. It is cold here. I would estimate the ambient temp in my room is 15-20C.

I have tried to provide as much info here as I think is relevant, if you require anything else I can provide it.

Thanks guys!

 

[CompuTronix]

shagann96,

On behalf of Tom's Hardware Moderator Team, welcome aboard!

Thank you for providing thorough information ... nice rig.

You're on the right track. You have no red flags or concerns.

There are many variables that "a lot of people on various forums" don't bother to consider, so numbers just get flung around like gorilla poo in a cage. "A lot of people on various forums" typically compare apples to oranges, so all you end up with is thermal fruit salad in a blender. It's no wonder why this topic is so confusing.

In order to have the lowest possible idle temperatures, "idle" must actually be idle. The definition of "idle" is minimum activity at 1% CPU Utilization in Windows Task Manager, which includes the following conditions that are normal and expected when conducting an idle test:

• Default power saving features enabled in BIOS.
• Windows Power Options set to "Balanced".
• Minimum Processor State at default value of 5%.
• Computer in an undisturbed "hands off" state.
• No programs or screensaver running, and off line.
• No Dropbox or Folding or SETI or background "tray-trash".
• Case Covers removed with pump and all fans at 100% RPM.

Here's how "idle" should look:

These are best case scenario conditions which minimize as many variables as possible across all hardware platforms. With High end cooling, average "idle" temperature may be as little as 3°C above ambient.

Concerning Core voltage for 14 nanometer processors, 1.4 volts is considered by Intel's Engineers to be a prudent maximum value, with 1.425 being the limit. See page 7, paragraphs 7, 8 & 9: Inside Intel's Secret Overclocking Lab

Here's the Maximum Recommended Vcore per Microarchitecture from 14 to 65 nanometers since 2006:

Here's how the Degradation Curves correspond to Maximum Recommended Vcore for 22 nanometer 3rd and 4th Generation, which differs from 14 nanometer 5th through 9th Generation:

Regarding Prime95, while it's true that Prime95 can be an "unrealistic" workload, here again, what "a lot of people on various forums" are either unaware of or don't bother to explain is that this only applies when AVX test selections are enabled.

Here’s a comparison of utilities grouped as thermal and stability tests according to % of TDP, averaged across six processor Generations at stock settings rounded to the nearest 5%:

Although these tests range from 70% to 130% TDP workload, Windows Task Manager interprets every test as 100% CPU Utilization, which is processor resource activity, not actual workload. Note that Prime95 with AVX is a brutal 130% workload, while Prime95 AVX disabled is an ideal 100% workload.

The definition of "full load" is a steady-state 100% TDP workload. Conducting a thermal test using Prime95 Small FFTs with all AVX test selections disabled is the correct test. It conforms to Intel's datasheets as a valid steady-state 100% workload.

As per Intel’s Datasheets, TDP and Thermal Specifications are validated “without AVX”. See page 87, Section 5.1.1 - Thermal Considerations, 1st paragraph, 2nd sentence: 8th and 9th Generation Intel® Core™ Processor Families Datasheet, Volume 1

Here's the nominal operating range for Core temperature:

Core temperatures above 85°C are not recommended.

Core temperatures below 80°C are ideal.

Core temperatures increase and decrease with Ambient temperature, so during the summer months your Core temperatures, will of course, be higher.

We have a Guide that covers this topic. It's a "Sticky" located at the top of the CPUs Forum where you posted your Thread. If you look up there you can't miss it:

Intel CPU Temperature Guide 2020
​

Most of the above information is straight from our Guide. We encourage our Members to check the Stickies as they frequently contain the information you need. This can save you time searching for answers from questionable sources who are often "somewhat less than well informed".

Once again, welcome aboard!

CT

 

[shagann96]

shagann96,

On behalf of Tom's Hardware Moderator Team, welcome aboard!

Thank you for providing thorough information ... nice rig.

You're on the right track. You have no red flags or concerns.

In order to have the lowest possible idle temperatures, "idle" must actually be idle. There are many variables that "a lot of people on various forums" don't bother to explain. Numbers get flung around like gorilla poo in a cage. "A lot of people on various forums" tend to compare apples to oranges, so all you end up with is thermal fruit salad in a blender.

The definition of "idle" is minimum activity at 1% CPU Utilization in Windows Task Manager, which includes the following conditions that are normal and expected when conducting an idle test:

Default power saving features enabled in BIOS.
Windows Power Options set to "Balanced".
Minimum Processor State at default value of 5%.
Computer in an undisturbed "hands off" state.
No programs or screensaver running, and off line.
No Dropbox or Folding or SETI or "tray-trash" running in the background.
Case Covers removed with pump and all fans at 100% RPM.

These conditions minimize as many variables as possible across all hardware platforms. With High end cooling, "idle" may be as little as 3°C above ambient.

Concerning Core voltage for 14 nanometer processors, 1.4 volts is considered by Intel's Engineers to be a prudent maximum value, with 1.425 being the limit. See page 7, paragraphs 7, 8 & 9: Inside Intel's Secret Overclocking Lab

Here's the Maximum Recommended Vcore per Microarchitecture from 14 to 65 nanometers since 2006:

Here's how the Degradation Curves correspond to Maximum Recommended Vcore for 22 nanometer 3rd and 4th Generation, which differs from 14 nanometer 5th through 9th Generation:

Regarding Prime95, while it's true that Prime95 can be an "unrealistic" workload, here again, what "a lot of people on various forums" are either unaware or don't bother to explain is that this only applies when AVX test selections are enabled.

Here’s a comparison of utilities grouped as thermal and stability tests according to % of TDP, averaged across six processor Generations at stock settings rounded to the nearest 5%:

Although these tests range from 70% to 130% TDP workload, Windows Task Manager interprets every test as 100% CPU Utilization, which is processor resource activity, not actual workload. Note that Prime95 with AVX is a brutal 130% workload, while Prime95 AVX disabled is an ideal 100% workload.

The definition of "full load" is a steady-state 100% TDP workload. Conducting a thermal test using Prime95 Small FFTs with all AVX test selections disabled is the correct test. It conforms to Intel's datasheets as a valid steady-state 100% workload.

As per Intel’s Datasheets, TDP and Thermal Specifications are validated “without AVX”. See page 87, Section 5.1.1 - Thermal Considerations, 1st paragraph, 2nd sentence: 8th and 9th Generation Intel® Core™ Processor Families Datasheet, Volume 1

Here's the nominal operating range for Core temperature:

Core temperatures above 85°C are not recommended.

Core temperatures below 80°C are ideal.

Core temperatures increase and decrease with Ambient temperature, so during the summer months your Core temperatures, will of course, be higher.

We have a Guide that covers this topic. It's a "Sticky" located at the top of the CPUs Forum where you posted your Thread. If you look up there you can't miss it:

Intel Temperature Guide
​

Most of the above information is straight from our Guide. We encourage our Members to check the Stickies as they frequently contain the information you need. This can save you time searching for answers from questionable sources who are often "somewhat less than well informed".

Once again, welcome aboard!

CT

Hello! Thanks for taking the time to read through!

So I’ve read through the stickied post, thanks for directing me to that. I don’t use forums an awful lot so it’s hard to navigate sometimes.

My main takeaway from your reply here, and the info in the sticky, is that my machine is running ‘safely’ but with the caveat that it isn’t a true reflection of my case and/or coolers performance without first establishing a) the ambient temperature in my room, and b) the true minimum idle (the specific settings you outline here and in the sticky).

If I repeat the testing following your advice and post updated results here, would you care to take a look at them again?

Thanks again!

 

[CompuTronix]

Absolutely!