Here is a simple way to lower CPU temperature and adjust under-voltage settings for an i9-9900KS.
Here is a simple way to lower CPU temperature and adjust under-voltage settings for an i9-9900KS.
So I'm a little bit obsessed with things that don't need to be, and my gaming computer gets super hot on its own settings in the BIOS. The temperature sometimes hits way up to 94 degrees Celsius when it's just sitting there. That is totally not cool at all, so I started trying to learn how to undervolt the CPU but yeah, you can easily find an OC guide or results that show higher temperatures than using undervolting. In my tests with an i9-9900k compared to a 9900ks, the difference wasn't huge, but it was there. So I decided to try overclocking instead, but just lowering the voltage without messing up the clock speed or frequency, which means keeping it at stock settings for 5ghz. So I started my test with the settings I used (you can see the screenshots below): *note that I live in a really hot place where the day temps are around 31 to 34 degrees Celsius. *Checked on the stock BIOS settings, the default voltage was 1.404v which is probably too high @1.3v = No BSOD when doing Cinebench r20 and during heavy games like Battlefield V and Warzone. I didn't test it with P95 because that stress test would be way too much work for me. There was no AIDA64 test done yet. @1.28v = No BSOD while gaming, but the temperature mostly stayed lower than 77c sometimes climbed up to reach 81-82c, Cinebench R20 temp was around 90 degrees if I'm not mistaken. There was no AIDA64 test done yet. @1.25v = BSOD after gaming for a few hours. Cinebench R20 finished its check. Aida64 stability test ran for one hour and fifteen minutes then it crashed at the highest temp of 95 degrees with no throttling slowing down the CPU. @1.26v = No crash yet, feels like I haven't done an AIDA64 test because gaming on a 5-7 hour run without anything happened. The temperature is also quite good under 65c only sometimes spikes to 71-72 degrees then goes back down under 65c. Idle temp was five degrees higher than my stock BIOS settings, which was around 42-44 degrees Celsius. If the BIOS settings rely on liquid cooling temps, if the CPU feels like it needs a temp of thirty-five degrees when it's running liquid usually the processor sits at three-six or three-seven degrees. Screenshots(Album) current BIOS settings: https://imgur.com/a/aMs4Mg4 View: https://imgur.com/a/aMs4Mg4 My main goal is to lower my CPU temp so here is the question, Are my BIOS settings okay? If I want to pass an AIDA64 stability test what changes do I need to make? I read on some posts that I have to change a specific value for VCCIO and VCCSA but I don't have any experience with this or so. Please tell me how to fix this if there are any bad settings or values please let me know I will change them right away. Full Specs: i9-9900KS Z390 Aorus Master | BIOS version F11c (Latest) G.Skill Trident Z 32GB(8GBx4) 3200Mhz CL14 (8GBx2 Trident Z RGB + 8GBx2 Trident Z NEO) B-Die NZXT Kraken x72 mod with Noctua NF A12x25 RTX 2070 Super EVGA G+ 1000w Lian Li PC-O11 Dynamic top mount radiator as exhaust others(side and bottom) as intake.
HVNtime says you've done well so far, but Phaaze88 is right; Prime95 isn't too much when used properly. Running P95 v29.8 with tiny calculations and all AVX features turned off makes it a steady 100% workload that follows Intel's official rules for checking how hot things get under normal conditions. According to the spec sheets, TDP (how much heat the chip generates) and thermal limits are tested without AVX enabled. If you want, Prime95 v26.6 also lacks AVX, so its tiny calculation workload is exactly the same as P95 v29.8 when AVX is turned off. When OCCT's first CPU test runs for a small data set and no AVX features are used, it becomes a steady 97% workload.
P95 isn't too much of a burden. It's used to check how well things get hot when you turn off fast math features like AVX and have a small FFT. You need to make sure error reporting shows up in the settings menu. Cinebench R20 has an "infinite loop" that checks if voltage stays steady. 1) Use an AVX offset of 2 or 3. Some apps and stress tests use AVX, and running at the same speed as normal SSE means they eat more power and need a higher voltage to stay cool than when using just SSE. 2) I don't know Aida64 well enough to answer that one. 3) They can help you get memory overclocked, like to 4000mhz or so. It's not needed if you're only running at 3200mhz. 4) Just the AVX offset, which is what I know about from this point.
My i9 9900KS is getting hot. It's cool at around 29 degrees, but when I type this message the heat spikes up to 33 degrees. My room temperature is about 24 degrees right now. While typing that, I ran IntelBurnTest v2.54 eleven times. The hottest cores reached 84 degrees and the coolest went down to 74 degrees during those tests. I have a Noctua NH-U14S air cooler and my case has three 200mm fans plus one 140mm fan and one 120mm fan. I use HWiNFO64 to watch temps, voltages, etc. The voltage on the VR OUT pin was 1.342 volts when it was idle and went down to about 1.287 to 1.318 volts while BurnTest ran. My Vcore is higher than yours. I set my offset to zero but I locked all cores at 4.8GHz because in games the difference between 5GHz and 4.8GHz isn't really noticeable, so I want them to last longer even though this chip is pretty weak for an i9 9900KS and it was hard to make it run stable. Getting it to work has been a real struggle since you probably have similar hardware. My motherboard uses the same F11c BIOS as yours and I'm using G.Skill RAM with 32GB at 3600MHz. I tried a trick from another person but can't remember who or where they posted it. Maybe I'll take some screen shots of my BIOS settings if you want to see them.
I just ran another 11 tests on my computer using the software called Intel Burn Test, set to 5GHz with a voltage of 1.35 volts in my BIOS settings. The highest temp reached was 85 degrees Celsius and the lowest was 77 degrees Celsius. These are the peak temps, not the normal ones. During the test, I saw that the power going from the motherboard stayed between 1.283 volts and 1.342 volts while the software ran. You can check those voltages by looking at a picture of my board here or another one here. You'll also need to install two other programs called Intel Burn Test and HWiNFO64 so you can see how my temps compare to what others show up on their websites.
HVNtime , It looks like you've done a good job so far, however, Phaaze88 is correct; Prime95 is not overkill when used correctly. P95 v29.8 Small FFTs with all AVX test selections disabled is a steady-state 100% workload which conforms to Intel's Datasheets for valid thermal testing. As per the Datasheets, TDP and Thermal Specifications are validated “ without AVX ” . If you prefer, Prime95 v26.6 doesn't have AVX, but its Small FFTs is the same workload as v29.8 Small FFTs without AVX. If OCCT's first CPU test, called " OCCT ", is configured for Small Data Set and No AVX , then it's a steady-state 97% workload that's nearly identical to Prime95's Small FFTs without AVX. A steady-state workload is key for thermal testing so the PSU, VRMs, CPU, socket, motherboard and cooler can thermally stabilize for steady Core temperatures. This is how Intel eliminates software variables to develop thermal specifications. 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. That's a lot of variables. 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. “Stress” tests vary widely and can be characterized into two categories; stability tests which are fluctuating workloads, and thermal tests which are steady workloads. 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 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. As you can see at the top of the scale, if P95 is configured for AVX, then it's a brutal 130% workload . Without an AVX offset to reduce power consumption, Core temperatures skyrocket. If you only game and don't anticipate ever using your rig for heavy CPU workloads such as rendering and transcoding which use AVX, then there's no need to concern yourself with configuring an AVX offset. Regardless, keep in mind that the AVX code that's used in recent games is much less demanding, and won't approach or exceed Prime95 Small FFTs without AVX. For a better perspective, use HWiNFO "Sensors Only" to observe Vcore, Package Power (Watts) and Package temperature (hottest Core) during testing. Numbers, even when logged, don't show the big picture, so right click on those parameters to see the graphs. Remember that frequency, Vcore and workload drive power consumption, which in turn drives Core temperatures. CT