Question Seeking advice with i9-10900K overclock

0Artur0

Distinguished
Aug 7, 2009
73
0
18,530
0
Just bought a new computer and I'm trying to overclock i9-10900K but I stumbled upon a strange issue I just can't figure it out on my own. I can get a stable 5.1GHz OC without any issues. The temps when stress testing (prime95, OCCT, Cinebench, etc) are in the mid-60s! That's really low, right? The issue is that there's no way I can get 5.2GHz, it immediately gives me BSOD after leaving BIOS, every time, even if I up the voltages to 1.400. I'm really confused because it appears I still have a lot of room temperatures-wise but I don't think it would be ok to go above 1.400 with voltages or am I wrong? Can I or should I even? I didn't try higher numbers because if I won't use it daily with those voltages, what's the point. What confuses me even further is that with default settings turbo mode can go to 5.3GHz without issues. I remember when I was OCing my old CPU (4790k) a few years ago only the temps were the limiting factor. Maybe I'm changing the wrong settings in BIOS, I watched pretty much all 10900K OC Youtube videos and I followed their advice. Any ideas what's the issue here and can I go higher than 5.1GHz since I still have like 10-20°C headroom with temps?

Motherboard: ASUS ROG STRIX Z490-G GAMING (WI-FI)
CPU: Intel Core i9-10900K
Cooler: SILVERSTONE Permafrost 240mm ARGB AIO WHITE water cooling (SST-PF240W-ARGB)

BIOS settings that I changed:

Ai Overclock Tuner: XMP I
ASUS MultiCore Enhancement: Enabled - Remove All limits
AVX Instruction Core Ratio Negative Offset: User Specify
AVX Instruction Core Ratio Negative Offset Value: 0
CPU Core Ratio: Sync All Cores
ALL-Core Ratio Limit: 51
Ring Down Bin: Enabled
Min. CPU Cache Ratio: 48
Max CPU Cache Ratio: 48
CPU Core/Cache Voltage: Manual Mode
CPU Core Voltage Override: 1.370
CPU Load-line Calibration: Leve 4:Recommended for OC

When trying 5.2GHz I changed the ratio to 52 and upped the voltages up to 1.400. I also tried Ring Down Bin on Auto, it didn't make any difference.
 

TravisPNW

Upstanding
Aug 26, 2020
359
131
360
0
Is this a prebuilt? You said just bought a new computer. I'm assuming no but had to ask. What kind of power supply are you running?

60C temps are better than what I get under stress testing which seems a bit odd given you say you have a 240mm cooler. That is a lot smaller than what I would run with this CPU. I run a 360mm cooler and have mine at 5.2ghz all core at 1.34v and it's been fine for the last 2 months since the build. Temps are 80-85C under P95/Cinebench etc... stress tests so I called it good there. Everyday usage temps are much lower.

All CPUs aren't created equal but you should be getting better if temps are indeed that low... I do run an AVX offset of -2 because I use Handbrake a lot.

As far as voltages, everything I read said to shoot for 1.35v which is another reason I stopped at 5.2ghz. Going 1.4 and up isn't something I'd recommend. (I don't claim to be an expert I just did a lot of research prior to my build and during the OC process)
 

TravisPNW

Upstanding
Aug 26, 2020
359
131
360
0
They didn't specify exactly what they did, so there's not that much for us to go on.
"prime95, OCCT, Cinebench, etc", what does that mean?
Running those stress tests/benchmarks I would assume... those are the programs that pushed my temps to the limit of my cooler anyway. Everyday use doesn't even get close to 80C outside of AVX stuff...

Like you said though... silicon lottery. Nothing guaranteed. I've seen people complaining about not getting beyond 5.1 with a 10900k but it is what it is.
 

Phaaze88

Polypheme
Ambassador
@TravisPNW
I meant -
Prime 95: Small FFT, All 3 AVX options off is harder to run than the likes of Blend mode.
It also comparatively close to what a real world AVX load could do. Funny how the actual AVX options are way too hard on this app.

OCCT: I've never used this one, and felt I never needed to. Following Computronix's Intel Temp Guide, this one appears to apply unnecessarily harder loads than Prime, so yeah... just gonna leave that one alone.

Cinebench R15: light and out of date.
Cinebench R20: although it was recently replaced by R23, I think it's still somewhat relevant.

5.1ghz 10900K: Boo-hoo?
This thing's already a beast out of the box.

The following is just my opinion, but: Up to 5.3ghz single, 4.9ghz multi... why the heck would I bother overclocking it? People forget, or are oblivious that not everything we do on the PC takes advantage of multi thread performance.
5.1ghz OC... so I give up 200mhz single for 200mhz multi... again, not everything we do takes advantage of the extra multi thread performance.
With Thermal Velocity Boost on this cpu, it's even more pointless. Large cooler + TVB + Intel Performance Maximizer(optional) = Win.
I look forward to more and more cpus having their boost clocks tied to cooling - today's gpus already do it.
It's so much simpler. Also, it's a smack in the face to folks doing 'sacrificial overclocks' - the single/multi tradeoff I mentioned earlier.
 
Reactions: TravisPNW

TravisPNW

Upstanding
Aug 26, 2020
359
131
360
0
5.1ghz 10900K: Boo-hoo?
This thing's already a beast out of the box.

The following is just my opinion, but: Up to 5.3ghz single, 4.9ghz multi... why the heck would I bother overclocking it? People forget, or are oblivious that not everything we do on the PC takes advantage of multi thread performance.
5.1ghz OC... so I give up 200mhz single for 200mhz multi... again, not everything we do takes advantage of the extra multi thread performance.
With Thermal Velocity Boost on this cpu, it's even more pointless. Large cooler + TVB + Intel Performance Maximizer(optional) = Win.
I look forward to more and more cpus having their boost clocks tied to cooling - today's gpus already do it.
It's so much simpler. Also, it's a smack in the face to folks doing 'sacrificial overclocks' - the single/multi tradeoff I mentioned earlier.
I agree about P95 and Cinebench... and Handbrake does crank up the heat which is why I run the -2 offset for all my encoding.

As for being a beast... I agree on that too. Totally. As for the overclocking and why the heck would you do it... it's funny you mention that because I'm considering just going back to stock for all those reasons that you mentioned. I read something a few weeks ago that basically said the same thing (maybe it was your post, I don't remember) but yeah, I'm starting to feel the need to dial back the all core overclock to stock speed because quite frankly I just don't see much of a need for it.

Been away on business for 2 weeks and my PC has been in sleep mode... just got back last night and plan to make some changes this coming week during some time off. I swear you are reading my mind because I'm thinking the same thing... with 4.9 multi and 5.3ghz single... why do I care about 5.2 all core?

I really don't. The benchmarks are done... and I already said I was a set it and forget it type guy a while ago when it comes to that... LOL did it once... no desire to do it again.

Cheers.
 
Reactions: Phaaze88

0Artur0

Distinguished
Aug 7, 2009
73
0
18,530
0
Ok, guys, I'm a noob. I was doing Large data set tests. I switched to small and temps went into 90s and pretty quickly it just BSOD'ed. With Small data set I can't even go to more than 5.0GHz or it just crashes. I guess I have a lot to learn. Sorry for the confusion, as I said, a noob.
 

Phaaze88

Polypheme
Ambassador
It's all good.
The current cooler isn't adequate for testing all core overclocks on that cpu - stock settings it's fine though.
1.40v is also too much Vcore. Dial it back down around 1.30v, or so.

Silicon Lottery - the company - gets away with even lower Vcore, but they're using settings I'm not familiar with. It's something new that came with 10th gen.
@TravisPNW
Are you familiar with these 2 columns I marked in red?
Comet LakeAll Core SSE FrequencyAll Core AVX2 FrequencyPer Core FrequencyAll Core Die Sense VcorePower Limit% Capable
10900K4.80GHz4.70GHz6C+100MHz
3C+200MHz
1.130V210W100%
10900K4.90GHz4.80GHz6C+100MHz
3C+200MHz
1.150V220WTop 99%
10900K5.00GHz4.90GHz6C+100MHz
3C+200MHz
1.170V230WTop 68%
10900K5.10GHz5.00GHz6C+100MHz
3C+200MHz
1.190V250WTop 21%
10900K5.20GHz5.10GHz6C+100MHz
3C+200MHz
1.210V270WTop 1%
 

TravisPNW

Upstanding
Aug 26, 2020
359
131
360
0
Silicon Lottery - the company - gets away with even lower Vcore, but they're using settings I'm not familiar with. It's something new that came with 10th gen.
@TravisPNW
Are you familiar with these 2 columns I marked in red?
Actually I'm not... feel free to clue me in with any info you think is needed. My OOTB initial setup had the vcore running high like 1.4+ (for stability I assume?) so when I go back to stock settings I'm definitely not going to have the mobo on auto settings.
 

Phaaze88

Polypheme
Ambassador
Actually I'm not... feel free to clue me in with any info you think is needed. My OOTB initial setup had the vcore running high like 1.4+ (for stability I assume?) so when I go back to stock settings I'm definitely not going to have the mobo on auto settings.
That should've been single core boost performance - that's always high. Under heavy load, it shouldn't be as high, but if it is, there's a problem in the bios.
Yeah, there was a problem with that some months ago: https://www.youtube(dot)com/watch?v=qQ_AETO7Fn4
Basically board vendor shenanigans, end user gets screwed.


I'm trying to look up those things I highlighted, because I don't really understand them.
 
Reactions: TravisPNW

Phaaze88

Polypheme
Ambassador
Oh for crying out loud... Die Sense is Vcore? I'm dumb. Why can't these vendors agree on one bloody name? I did find this:
"Maximus boards read die sense as vcore (Default) if it's selected in the BIOS as die sense. Non-maximus boards do Super I/O only."

Per Core Frequency is pretty self explanatory though.

Geez, then no one needs to be cranking 1.30v + through their 10900K OCs...
 
Reactions: TravisPNW

TravisPNW

Upstanding
Aug 26, 2020
359
131
360
0
Yeah, there was a problem with that some months ago: https://www.youtube(dot)com/watch?v=qQ_AETO7Fn4
Basically board vendor shenanigans, end user gets screwed.

I'm trying to look up those things I highlighted, because I don't really understand them.
Yeah I actually have that as one of my liked videos viewed as part of my research prior to the build. It was also one of the reasons my build didn't stay on auto mobo settings very long.

Upon further review of that chart you posted... looks like OP's chip is the top 21% (5.1) and mine is the top 1% (5.2) although it says 1.21v and I'm at 1.34v.

Geez, then no one needs to be cranking 1.30v + through their 10900K OCs...
I run the MSI Unify board and all the reviews said it was amazing all around especially for OCing... LOL
 

CompuTronix

Intel Master
Moderator
... i9-10900K ... 240mm AIO ... temps when stress testing (prime95, OCCT, Cinebench, etc) are in the mid-60s ... no way I can get 5.2GHz ... even if I up the voltages to 1.400 ... don't think it would be ok to go above 1.400 with voltages or am I wrong ... my old CPU (4790k) a few years ago only the temps were the limiting factor ...
Respectfully, not quite so. Concerning Vcore, one size does not fit all. Overclocking any processor is always limited by two factors; voltage and temperature. Each Microarchitecture has a “Maximum Recommended Vcore”. For example, it’s important to point out that 22 nanometer 3rd and 4th Generation processors, such as your 4790K, will not tolerate the higher Core voltages of other Microarchitectures.

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


Each Microarchitecture also has a "Degradation Curve". 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 10th Generation:


Degradation Curves are relative to the term “Vt (Voltage threshold) Shift” which is expressed in millivolts (mv). Users can not monitor Vt Shift. Vt Shift basically represents the potential for permanent loss of normal transistor performance.

There's more detailed explanations in the Intel CPU Temperature Guide 2021, Section 8 - Overclocking and Voltage. I suggest that you check out the entire guide, especially Section 11 - Thermal Test Basics.

They didn't specify exactly what they did, so there's not that much for us to go on.
"prime95, OCCT, Cinebench, etc", what does that mean?
... Prime 95: Small FFT, All 3 AVX options off ... comparatively close to what a real world AVX load could do. ...
OCCT: I've never used this one ... just gonna leave that one alone ... Cinebench ... R23, I think it's still somewhat relevant ...
I agree about P95 and Cinebench... and Handbrake does crank up the heat which is why I run the -2 offset for all my encoding ...
@0Artur0
We need more info from you in regards to the settings you used in those stress tests.
... was doing Large data set tests. I switched to small and temps went into 90s and pretty quickly it just BSOD'ed ...
Here's the nominal operating range for Core temperature:

Core temperatures above 85°C are not recommended.

Core temperatures below 80°C are ideal.


... The current cooler isn't adequate for testing all core overclocks on that cpu ... 1.40v is also too much Vcore.
As Phaaze88 has pointed out, a 240mm AIO is inadequate for cooling a 10 Core 20 Thread high-end "K" processor which can consume over 250 Watts, especially when overclocked at high Core voltage. If you intend to overclock, the 10900K demands a 360mm AIO or a custom loop in order to keep it cool. Silicon Lottery can professionally delid your processor, which can drop temperatures by 5 to 12°C. They can also "bin" your processor so you'll know its overclocking capability and the settings needed.

Guys,

Voltage and temperature numbers get flung around forums like gorilla poo in a cage. For example, users will often say "I ran AIDA64" ... yes ??? ... and ... ??? what exactly did you run ??? 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 temperature possibilities. When ambient (room) temperature isn't mentioned, and load test conditions aren't defined, the Core temperatures you see on various websites and forums can be highly misleading. Therefore, several points need to be clarified and emphasized.

0Artur0, in order to provide any meaningful apples-to-apples comparisons, as Phaaze88 alluded to in his 1st and 3rd posts, it's important to be VERY specific, otherwise, all we have is apples-to-oranges thermal fruit salad in a blender. When discussing thermal performance, there are 3 major variables; environment, hardware and software. By taking a methodical approach, variables in environment and software can be accounted for, which then leaves differences in hardware to sort out. This reduces the major variables to their lowest common denominators, so test results make sense, are repeatable and easier to compare.

There's been no mention of ambient (room) temperature (environment), for which the International Standard for "normal" is 22°C or 72°F. Ambient can be a HUGE variable. Users write into our forums who live anywhere from the Arctic Circle to the Equator, so air temperature at the computer's intake might be anywhere from 10°C (50°F) to 40°C (104°F). If you don't say, and we don't ask, then we're blundering blindly forward based on an unknown major variable.

In your 2nd post you provided very limited specifics. It's always critical to define your exact software load conditions. Most users don't realize how much “stress” tests vary, which 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-state 100% workload with steady Core temperatures. As per Intel’s Datasheets, TDP and Thermal Specifications are validated “without AVX.

With respect to the %TDP scale shown below, when heavy "real-world" AVX workloads are at "peak" load, such as video transcoding apps like HandBrake (which are fluctuating workloads), the workload will typically approach, but not exceed P95 Small FFTs without AVX. The CineBench R23 CPU Render Test shown below is a good example of a utility which replicates real-world AVX transcoding workloads. Prime95 Small FFTs (all AVX test selections enabled) is nearly a 130% workload, which is unrealistically higher than real-world AVX workloads. This is why an AVX "Offset" is used to keep Core temperatures in check, just as TravisPNW pointed out.

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. Core temperatures respond directly to Power consumption (Watts), which is driven by Vcore and workload. Prime95 Small FFTs (AVX disabled) provides the correct workload for testing thermal performance.

When running heavy utility tests or AVX apps, in addition to Core temperatures, always keep an eye on Package Power consumption (Watts). Download the "Portable" version of HWiNFO and run "Sensors Only". However, numbers alone can't reveal the big picture. The best way to visualize thermal performance is to observe how your hardware responds to software workloads on a graph. In addition to a few other select utilities, HWiNFO also has graphs. Just right-click for "Show Graph" on the parameters you want to see.

The "Charts" in SpeedFan span 13 minutes, and show how each test creates distinct thermal signatures.



Figure 12-1

Shown above from left to right: Small FFTs, Blend, Linpack and IntelBurn Test.​

Note the steady thermal signature of Small FFTs, which allows accurate measurements of Core temperatures. A steady 100% workload is key for thermal testing so the CPU, cooler, socket, motherboard and voltage regulator modules (VRM) can thermally stabilize.

Phaaze88, in OCCT 7.3.2, if the first test, called "CPU", is configured for Small Data Set, Steady Load, SSE Instruction Set, then it's very nearly identical to Prime95's Small FFTs without AVX. When CineBench R23 is configured for MultiCore, Test Duration 10 minutes or more, although it uses AVX and is a somewhat fluctuating workload and pauses between renders, if you observe CPU thermal behavior on any utility that can display a temperature graph, you'll see that (when configured as described) P95, OCCT and CineBench R23 all provide workloads within a degree or so from one another.

CT:sol:
 

TravisPNW

Upstanding
Aug 26, 2020
359
131
360
0
Here's the nominal operating range for Core temperature:

Core temperatures above 85°C are not recommended.

Core temperatures below 80°C are ideal.​

As Phaaze88 has pointed out, a 240mm AIO is inadequate for cooling a 10 Core 20 Thread high-end "K" processor which can consume over 250 Watts, especially when overclocked at high Core voltage. If you intend to overclock, the 10900K demands a 360mm AIO or a custom loop in order to keep it cool. Silicon Lottery can professionally delid your processor, which can drop temperatures by 5 to 12°C. They can also "bin" your processor so you'll know its overclocking capability and the settings needed.

Good stuff. I read a lot of your posts when I was researching my December build... and when I was overclocking.

"Core above 85C not recommended" is why I stopped my overclock at 5.2ghz all core... 80-85C temps during stress/stability testings with my 360mm AIO is where I was at... so I called it good. This was on Jan 8th and 6 weeks later the system hasn't given me any issues. As mentioned upthread I am considering just going back to stock speeds. The OC boost was great for the benchmarks but do I really need all core speeds like this given the features the 10900k already has? I'm thinking not.

Quite happy with the performance of the 10900k. I actually delid my previous CPU (7700k) back in 2017 but not seeing a need here. Even though the stress/stability temps are near the top the every day use temps are well within optimal range... 50-70C. As said I do a lot of Handbrake encoding and that did push my temps to 80C... so I just set a -2 AVX offset and now I encode around 70C with no real performance loss... we are talking 2-3 minutes difference in encoding time which I don't worry about.

Not worried about power either... I went with a 1000W unit because the CPU (and 3090 GPU) love the juice. I see so many people build premium systems and want to go cheap on a PSU... and that isn't how I roll.

Cheers.
 
Reactions: CompuTronix

Afro_ninja199

Prominent
Aug 10, 2019
307
34
740
6
Ok, guys, I'm a noob. I was doing Large data set tests. I switched to small and temps went into 90s and pretty quickly it just BSOD'ed. With Small data set I can't even go to more than 5.0GHz or it just crashes. I guess I have a lot to learn. Sorry for the confusion, as I said, a noob.

did you run small with disable avx 512 /avx2 / avx?
 

ASK THE COMMUNITY