I7 7700K Low Temps underload

[sillypubg]

Fairly new build running stock 7700k and nhu12s but i noticed even through a cpuz stress test and fire strike and intels ipdt my cpu only every gets to 60c usually stays well below that is this normal i feel like this cpu should be hotter with this cooler

 

[mdd1963]

According to task manager, is it ramping up to 4.5 GHz during a Defender scan, for example? Are you in Balanced power plan within Windows? 60c under load would be 4c less than mine, maybe you got a great lottery draw! Try Prime 95/ small FFTs, 8 threads...(mines rises to 71-73c , but temps go no higher after 12 -15 min; maybe yours would max at 69, which would be awesome for a 7700k!

 

[mdd1963]

My CPUz stress shows 64c , just as a point if reference for what I consider 'good/normal'...60c is VERY good, IMO...

 

[lumineZ]

Pretty good for a stock 7700k with no delid.

 

[mdd1963]

It's awesome, in fact...best i've heard of, stock, with that cooler....without delidding. Lotto winner!

 

[CompuTronix]

sillypubg,

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

Keep in mind that all computer temperatures increase and decrease with Ambient temperature. Normal or "Standard" Ambient temperature is 22°C or 72°F. People write into our Forums from all over the world. They run PC's in summer months at room temperatures of 40°C (104°F) without A/C, or winter months in room temperatures of 10°C (50°F) with very little heat.

If people don't offer their Ambient temperature and we don't ask, then it's easy to see how temperature questions can quickly reach the wrong conclusions based on incomplete information and false assumptions. Ambient temperature can be a huge variable that must be taken into consideration, and should always be included with system specs. Without knowing Ambient temperature, you can't compare apples to apples.

Q: What is your ambient (room) temperature?

Also, CPU-Z's "Stress CPU" test is not the correct test to use for determining your processor / cooler thermal performance. Although this test (as well as numerous others) will indicate 100% CPU Utilization (busy), it's not equal to 100% workload, which means 100% Thermal Design Power (TDP) at stock settings. CPU-Z's stress test is instead only about 78% workload, which gives you a false sense of security, since it also gives you lower Core temperatures.

Intel tests their processors at 100% TDP, which is Power (Watts) that's dissipated as heat. Since your 7700K is 91 Watts TDP, in order to determine a valid thermal baseline, the goal is to use a test which will run your processor as close as possible to 100% TDP, or 91 Watts. Additionally, an ideal thermal test also needs to be a steady workload, rather than a fluctuating workload. Although CPU-Z's stress test is a steady workload, most other stress tests are not, some of which can spike as high as 120% workload.

The only other test which is both a 100% TDP workload and is a steady workload is the Small FFT's test in Prime95 Version 26.6 - http://www.mersenneforum.org/showthread.php?t=15504

Download and run Small FFT's for just 10 minutes to determine your computer's valid CPU thermal baseline performance.

• Prime95 v26.6 - http://www.mersenneforum.org/showthread.php?t=15504

Do not use versions later than 26.6 on 2nd through 7th Generation i3, i5 or i7 CPU's, which all have AVX (Advanced Vector Extension) Instruction Sets. Later versions run AVX code on the CPU's Floating Point Unit (FPU) which causes unrealistic temperatures up to 20°C higher. The FPU test in the utility AIDA64 shows similar results.

If you run AVX apps such as for rendering or transcoding, you may need to reduce Vcore and Core speed and / or upgrade your cooling so that Core temperatures don’t reach 85°C. Recent motherboards address the AVX problem by providing offset adjustments in BIOS. An offset of -2 or -3 (200 or 300 MHz) is usually sufficient. Asus RealBench runs a realistic AVX workload typically within +/- a few % of TDP, and is an excellent utility for testing overall system stability, whether you're overclocked or not.

• Asus RealBench - http://rog.asus.com/rog-pro/realbench-v2-leaderboard/

Expect your Core temperatures to be considerably higher at 100% TDP than Core temperatures using CPU-Z's 78% TDP stress test. Core voltage and the resulting Power consumption may vary between motherboard manufacturers, BIOS versions and Auto / Default settings.

Download and run Core Temp to see your CPU's TDP and Power consumption - http://www.alcpu.com/CoreTemp

Here's the operating range for Core temperature:

Core temperatures above 85°C aren't recommended.

CPU workloads and the resulting Core temperatures can vary greatly between games. Idle temperatures below 25°C are generally due to Ambient temperatures below 22°C.

If you'd like to learn more about processor temperatures, then read this "Sicky" near the top of our CPUs Forum: Intel Temperature Guide - http://www.tomshardware.com/forum/id-1800828/intel-temperature-guide.html

Once again, welcome aboard!

CT

 

[mdd1963]

Yet to see anything that will reveal/expose/get higher temps than Prime95/small FFTs....save for perhaps Prime with Blender at the same time?

Agree CPUz is not the 'ultimate stresser/heat tester', as it is a full 9-10C under what Prime does...; it is a reasonable 'heavy gaming equivalent' cpu load, at least, however...

 

[TJ Hooker]

cpuz stress test and fire strike and intels ipdt my cpu only every gets to 60c

I find that CPU-Z stress test doesn't actually thermally stress my CPU all that hard. Firestrike is primarily a GPU benchmark (and relatively short), so not a huge surprise your CPU isn't getting too hot. Never used IPDT, but doesn't sound like it's really meant for thermal/stress testing. As said above, try running Prime 95 v26.6, small FFTs test.

 

[mdd1963]

Using my own again as semi-normal point of reference, it idles at 33C, BF peaks at about 61C at 1080P gaming, Intel stress/CPU tester gets 63C, CPUz stress gets 64C, Prime 95/small FFTs bounces between 71-73C after 15 minutes, no more rises after that, all with 24C ambients, on NH-D15... (Oh, and Linux stress test package/cpu tests at about 54C...which is not really a stress test at all, however...in comparison)

 

[CompuTronix]

Yet to see anything that will reveal/expose/get higher temps than Prime95/small FFTs....save for perhaps Prime with Blender at the same time?

Agree CPUz is not the 'ultimate stresser/heat tester', as it is a full 9-10C under what Prime does...; it is a reasonable 'heavy gaming equivalent' cpu load, at least, however...

Respectfully, many of our Members and readers are not aware that there's a significant difference in load and the resulting temperatures between Prime95 v26.6 and later versions. By simply mentioning "Prime95", those who are not better informed naturally assume it's OK to download and run later versions, then are confused and astonished when they see their poorly TIM'd i7's quickly reach "Throttle" temperature at 100°C.

Just as it's important in these temperature Threads to include Ambient room temperature, it's also important to be very specific when discussing stress test utilities, such as using only Prime95 version 26.6 due to the AVX issue.

This helps to reduce the amount of misinformation and confusion on our Forums, and provides everyone with a better understanding of an already complex and confusing topic, which is largely due to Intel's Thermal Specifications, that for most folks, are about as clear as mud.

I've tested every utility available over the years on various Intel platforms with respect to % TDP as well as Core temperatures. P95 v26.6 Small FFT's is the only utility that runs a steady workload typically within +/- 3% of TDP, which is critical to replicating a valid thermal baseline that so closely matches Intel's test methods.

I'm continually testing, cross referencing and evaluating stress test utilities. Although there are several utilities better suited for stability testing, I've never found any for thermal testing as accurate as Prime95 v26.6 Small FFT's.

CT

 

[sillypubg]

According to task manager, is it ramping up to 4.5 GHz during a Defender scan, for example? Are you in Balanced power plan within Windows? 60c under load would be 4c less than mine, maybe you got a great lottery draw! Try Prime 95/ small FFTs, 8 threads...(mines rises to 71-73c , but temps go no higher after 12 -15 min; maybe yours would max at 69, which would be awesome for a 7700k!

so i am really new to pcs and have never overclocked b4 thanks for all the support btw i ran the prime 95 fft and after 10 mins or so the highest temp i got was 72 with an average of around 70 so i guess im pretty lucky as for my ambient temperature im not sure but im in the basement and it is fairly cold so that could have something to do with it

 

[mdd1963]

72C peak with that cooler is great...enjoy!

For CompTronix- (Are the Prime 26.6-induced temps typically higher or lower than that of later versions?)

 

[CompuTronix]

... For CompTronix- (Are the Prime 26.6-induced temps typically higher or lower than that of later versions?)

Core temperatures using version 26.6 are lower. Later versions are up to 20°C higher. From my first post:

" ... Do not use versions later than 26.6 on 2nd through 7th Generation i3, i5 or i7 CPU's, which all have AVX (Advanced Vector Extension) Instruction Sets. Later versions run AVX code on the CPU's Floating Point Unit (FPU) which cause unrealistic temperatures up to 20°C higher.

If there's anything about this topic that isn't quite clear, send me a PM and I'll be glad to help you, or you can click on the link in my signature and read the Guide. It's all in there.

CT