vladakv,
Respectfully, Core temperatures are the result of Power consumption (Watts) which is driven by Core voltage, Core frequency and software workload. In terms of workload, AVX, AVX2 and AVX-512 Instruction Sets can have a
HUGE impact on Power consumption. This is why "AVX Offsets" are provided in BIOS, which "downclocks" the processor to reduce Power consumption and maintain reasonable Core temperatures.
The correct way to run Prime95 is with
all AVX test selections disabled, which makes it a
perfect 100% workload. For those who do a little research rather than rely on hearsay, this is, and has been, common knowledge for many years.
If AVX test selections are
not disabled, then Prime95 is a brutal 130% workload which WILL overload your CPU. The Small FFTs test is a steady-state workload, which is perfect for checking thermal performance. A steady-state workload is key. No one can get an accurate indication of thermal performance by using software that runs a fluctuating workload that looks like a bad day on the stock market. Fluctuating workloads are better suited for "stability" testing, NOT "thermal" testing. The best way to visualize and understand the different characteristics between various software workloads is to see it in utilities that use graphs.
It's all covered in the
Intel CPU Temperature Guide 2021. The Guide is a Sticky at the top of the CPUs Forum, or you can simply use Google with words like Intel + CPU + temperature. See
Section 11 - Thermal Test Basics and
Section 12 - Thermal Test 100% Workload.
If you configure OCCT > CPU to run Small > Normal > Steady > SSE, then you get a steady-state workload that's very nearly identical to Prime95 configured for Small FFTs > (all AVX test selections disabled). Both tests are within a few percent Power consumption (Watts) of one another with Core temperatures that are also within a degree or two of one another. CineBench R20 & R23 Multi Core tests also have Power consumption and Core temperatures that are nearly identical to both P95 and OCCT tests as specified. Although the CineBench workload is steady-state, since it pauses between passes which allows hardware to briefly cool down, CineBench is better suited for testing
stability, while P95 and OCCT are better suited for testing
thermal performance.
If cooling is inadequate on any given rig, then Core temperatures will soar past 80°C on P95, OCCT and CineBench, which can adversely affect stability and longevity. However, if you don't run rendering / transcoding software and all you do is game, then CPU-Z > Bench > Stress CPU is a steady-state workload that's only about 80%, which is more consistent with the most CPU intensive gaming titles. Keep in mind that actual % TDP workload (Power consumption in Watts) is NOT the same as % CPU Utilization. When overclocked, workloads can easily exceed 100% TDP (either PL1 or PL2) while CPU Utilization is processor resource activity, which can only indicate a maximum of 100%.
“Stress” tests vary widely and can be characterized into two categories;
stability tests which are
fluctuating workloads, and
thermal tests which are
steady workloads. Prime95 Small FFTs
(AVX disabled) is ideally suited for testing thermal performance, because it conforms to Intel's Datasheets as a
steady 100% workload with
steady Core temperatures. No other non-proprietary utility can so closely replicate Intel's thermal test workload.
Utilities that don't
overload or
underload your processor will give you a valid thermal baseline. Here’s a comparison of utilities grouped as
thermal and
stability tests according to % of TDP, averaged across six processor Generations at stock Intel settings rounded to the nearest 5%:
Take a look:
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 %TDP workload.
Core temperatures respond directly to Power consumption (Watts), which is driven by workload. Prime95 Small FFTs
(AVX disabled) provides the correct workload for testing thermal performance. So when you're tweaking your overclock and Core voltage, Watts (Power consumption) is a much more critical indicator than CPU Utilization.
Intel Extreme Tuning Utility (XTU) is a fluctuating workload at about 80% TDP. AIDA64 has 4 CPU related stress test selections (CPU, FPU, Cache, Memory) which have
15 possible combinations that yield
15 different workloads and
15 different Core temperatures. The individual FPU test is about 115% TDP workload, the CPU/FPU combination is about 90%, all 4 tests combined is about 80% and the individual CPU test is only about 70%. All other AIDA64 test selections are fluctuating workloads which are suitable for
stability testing, but not for
thermal testing.
Milopingg,
On behalf of Tom's Moderator Team, welcome aboard!
To partially answer your question, it's best for the CPU to settle down to ~ 800 MHz (minimum) and ~ 0.700 volts (minimum) when idle, and to ramp up to maximum boost frequencies when workloads dictate, without exceeding 1.425 volts or 80°C, provided that BIOS settings are correct, Windows Power Options are "Balanced" with Minimum Processor State at its default value of 5%, and cooling is adequate. Remember that lower Core voltage = lower Power consumption (Watts) = lower Core temperature.
Although "Throttle" temperature (Tj Max) is 100°C, if your hottest Core is near this temperature, then your CPU is already too hot. The consensus among well informed and highly experienced reviewers, system builders and expert overclockers, is that it's prudent to observe a reasonable thermal margin below Throttle temperature for ultimate stability, performance and longevity. So regardless of environmental conditions, hardware configurations, software workloads or any other variables,
Core temperatures above 85°C are not recommended.
Here's the nominal operating range for Core temperature:
Core temperatures above 85°C are not recommended.
Core temperatures below 80°C are ideal.
Moreover, for Thermal Velocity Boost (TVB) to perform at its full potential, Core temperatures should not exceed 70°C.
Sorry to pop your bubble, but you've jumped into the deep end of the pool and you're in
WAY over your head. Clicking around in BIOS without knowing exactly what you're doing is a recipe for unknowingly damaging or destroying your brand-new hard-earned hardware. However, if you just want to change Core voltage settings to reduce Core temperatures, then we strongly suggest that you make changes in BIOS, rather than using a software utility such as Intel Extreme Tuning Utility or Intel Performance Maximizer. Whether you intend to eventually overclock or not, since Overclocking Guides cover the topic of Core voltage adjustments and stability testing, there are plenty of Overclocking Guides and YouTube videos available on the Internet to help you. Some are better than others, so don't rush into it; take your time, be patient and pay attention to detail. Google is your friend, but here's a good place to start:
• CPU Overclocking Guide and Tutorial for Beginners
• Intel CPU Temperature Guide 2021
Once again, welcome aboard!
CT
I'd be afraid If cpu spikes 1.4-1.5V on idle. Maybe it will be better to check it. In bios you could make constant voltage for this cpu. I did constant 1.28V.
Facebook
Twitter
Instagram