Monday, November 1st 2021

Intel Core i9-12900K 36% Faster Than Stock in Maximum Turbo Power Mode

The recently announced Intel Core i9-12900K is set to launch on November 4th alongside the i7-12700K, and i5-12600K which is the date when we will see reviews for the processors released. We have already seen numerous leaks regarding the processors performance leading up to the announcement and we have now received some new leaked performance figures for the processors when operating in Maximum Turbo Power (MTP). The MTP is defined by Intel as the maximum sustained power dissipation of a processor compared to TDP which is the base power draw. The performance difference between these two power modes has been revealed from Cinebench R20 multi-threaded results posted by Wofstame the Gaming Desktop Product Planning Manager for Lenovo China.

The Intel Core i9-12900K scores 7492 points when running at its TDP of 125 W and 10180 points or 36% faster when operating at the MTP of 241 W. This performance difference is less notable for the other processors with the Core i7-12700K seeing a 30% improvement between its 125 W and 190 W power modes while the Core i5-12600K sees a 10% improvement from the 125 W TDP to 150 W MTP. Intel appears to be extracting the maximum performance from their Core i9-12900K with diminishing returns from the increased power budget compared to the other processors.
Source: @9550pro
Add your own comment

120 Comments on Intel Core i9-12900K 36% Faster Than Stock in Maximum Turbo Power Mode

#76
Mussels
Freshwater Moderator
It's because people cant tell the difference between
A. Intel CPU's produce a lot of heat
B. Intel CPU's run hot

Reality is more like:
1. Intel requires big heatsinks/coolers
2. AMD requires heatsinks with good contact/baseplates
Posted on Reply
#77
AusWolf
MusselsIt's because people cant tell the difference between
A. Intel CPU's produce a lot of heat
B. Intel CPU's run hot

Reality is more like:
1. Intel requires big heatsinks/coolers
2. AMD requires heatsinks with good contact/baseplates
Even that's not true. ;)
I'm running my i7-11700 with a tiny be quiet! Shadow Rock LP at 75 °C max in Cinebench R23 all-core with a 125 W PL1. The R5 3600 couldn't keep this temperature at stock (88 W) with the same cooler.

In my experience it would be more like:
1. Intel requires tweaking to deliver performance that suits your cooling (unless it's a 65 W part kept at stock),
2. AMD requires good contact with the cooler and good case airflow, or a lowered power target.
Posted on Reply
#78
B-Real
Dyatlov AWow, better AMD halves the prices :nutkick:
LOL XDD
Posted on Reply
#79
RandallFlagg
And 12600K at 115W.
For reference a 5800X gets about 15k multi CB 23 and about 6.1K multi CB 20

i.e. 5800X is slower and will draw more power than the 115W of the 12600K :

Posted on Reply
#80
rares495
The Quim ReaperAs someone who much prefers traditional CPU coolers over AIO's, you have to ask if there is any current tower cooler out there that can cope with a 12900K at full power..The (supposedly) best of the best, the Noctua DH15 has a TDP limit of 250w, so at 241w, its going to be right on the edge of thermal throttling territory, in 100% CPU load productivity work like rendering and encoding, at least Gaming will be no problem, I suppose, as that never pushes any CPU to all core 100% load for anything other than brief periods for stuff like shader compilation.
The D15 can't cool 241W. No way. The 250W rating is a joke. Even some AIOs would struggle with that.
Posted on Reply
#81
HenrySomeone
RandallFlaggAnd 12600K at 115W.
For reference a 5800X gets about 15k multi CB 23 and about 6.1K multi CB 20

i.e. 5800X is slower and will draw more power than the 115W of the 12600K :
Oh boy, it's even better than what was shown previously - the insufferable redsters are in for a world of hurt, this is gonna be good! :cool:
Posted on Reply
#82
RandallFlagg
HenrySomeoneOh boy, it's even better than what was shown previously - the insufferable redsters are in for a world of hurt, this is gonna be good! :cool:
Ominous.. :laugh:

"Intel Core i9-12900K and i5-12600K lead ominous Alder Lake assault on UserBenchmark's charts as highest Ryzen chip languishes in 17th position"

www.notebookcheck.net/Intel-Core-i9-12900K-and-i5-12600K-lead-ominous-Alder-Lake-assault-on-UserBenchmark-s-charts-as-highest-Ryzen-chip-languishes-in-17th-position.576855.0.html
Posted on Reply
#83
AusWolf
RandallFlaggAnd 12600K at 115W.
For reference a 5800X gets about 15k multi CB 23 and about 6.1K multi CB 20

i.e. 5800X is slower and will draw more power than the 115W of the 12600K :

If this is true, and then AMD launches Zen 3+ with 3D cache, there will be some sweet competition happening! :cool: Unless both Intel and AMD jump on the chip shortage bandwagon to artificially inflate prices (I mean, the 12900K appears to be expensive enough already, but still).
Posted on Reply
#84
Mussels
Freshwater Moderator
AusWolfEven that's not true. ;)
I'm running my i7-11700 with a tiny be quiet! Shadow Rock LP at 75 °C max in Cinebench R23 all-core with a 125 W PL1. The R5 3600 couldn't keep this temperature at stock (88 W) with the same cooler.

In my experience it would be more like:
1. Intel requires tweaking to deliver performance that suits your cooling (unless it's a 65 W part kept at stock),
2. AMD requires good contact with the cooler and good case airflow, or a lowered power target.
Your 11700 is a 65W chip, and the most power efficient intel in recent years. It's literally the one example.


Meanwhile, a 3600 or 5600x runs on the wraith stealth
Why does such a tiny ass cooler work so well for them? The mounting screws, higher pressure helps them a lot (and i dont get why the bigger stock coolers dont use them, too)
RandallFlaggAnd 12600K at 115W.
For reference a 5800X gets about 15k multi CB 23 and about 6.1K multi CB 20

i.e. 5800X is slower and will draw more power than the 115W of the 12600K :

Edited: I see how it shows 115W, but it also shows PL1 and PL2 at 202W
I dont quite trust that it's not running close to either of those
Posted on Reply
#85
RandallFlagg
MusselsEdited: I see how it shows 115W, but it also shows PL1 and PL2 at 202W
I dont quite trust that it's not running close to either of those
I don't think he would be able to keep max package 56C during the run if it were over 200W. Just because PL1 and PL2 are set to 202W doesn't mean that it used that, and this is "just" a 12600K with the max frequency limits intact. See next image.



Same guy, 12700K, 158W and 67C. My AIO will keep my 10850K below that temp at 158W. This score smashes the 5900X, and draws less power too.

So there is a pattern developing..

Posted on Reply
#86
Mussels
Freshwater Moderator
I hope its true.

We just have a long history of intel leaks being faked before reviews.
Posted on Reply
#87
ViperXTR
RandallFlaggAnd 12600K at 115W.
For reference a 5800X gets about 15k multi CB 23 and about 6.1K multi CB 20

i.e. 5800X is slower and will draw more power than the 115W of the 12600K :

Nice, and yea it draws less power than my 5800X
Posted on Reply
#88
AusWolf
MusselsYour 11700 is a 65W chip, and the most power efficient intel in recent years. It's literally the one example.


Meanwhile, a 3600 or 5600x runs on the wraith stealth
Why does such a tiny ass cooler work so well for them? The mounting screws, higher pressure helps them a lot (and i dont get why the bigger stock coolers dont use them, too)
It's not that. When I had the 3600, the Wraith Stealth didn't work out for me at all. That's why I bought the Shadow Rock LP, which basically kept it at the top of its operational temperature range, but nothing more... at stock! Which means a power limit of 88 W. I'm using the same cooler with the 11700 with the power limit cranked up to 125 W (way above stock), and it keeps it at 75 °C max. That's 37 Watts more than the 3600 ate, and it runs cooler with the same cooling setup. It's the same temperature my R3 3100 runs at with the Wraith Stealth (of course the Shadow Rock LP is better, but still) eating around 50-55 Watts. Heat density is a big factor in how hot a chip runs - even bigger than its power consumption in my experience.
Posted on Reply
#89
RandallFlagg
AusWolfIt's not that. When I had the 3600, the Wraith Stealth didn't work out for me at all. That's why I bought the Shadow Rock LP, which basically kept it at the top of its operational temperature range, but nothing more... at stock! Which means a power limit of 88 W. I'm using the same cooler with the 11700 with the power limit cranked up to 125 W (way above stock), and it keeps it at 75 °C max. That's 37 Watts more than the 3600 ate, and it runs cooler with the same cooling setup. It's the same temperature my R3 3100 runs at with the Wraith Stealth (of course the Shadow Rock LP is better, but still) eating around 50-55 Watts. Heat density is a big factor in how hot a chip runs - even bigger than its power consumption in my experience.
Gen 10 doesn't need there to be a big temperature difference between the chip and the heat spreader to x-fer a lot of heat. Ryzen does. It's a materials engineering thing, there is a value called a heat transfer coefficient which when multiplied by the temp difference gives you a heat flux value aka thermal power transferred per unit area. The thermal power transfer per unit area on a Gen 10 is better than Gen 9 and any Zen. It's like having a factory delid.
Posted on Reply
#90
Mussels
Freshwater Moderator
ViperXTRNice, and yea it draws less power than my 5800X
We also have to be aware HWinfo may not fully support the chips yet, or there more be more than one source of power to read.

Is there a second power package for the E cores just off screen, too?
Posted on Reply
#91
AusWolf
RandallFlaggGen 10 doesn't need there to be a big temperature difference between the chip and the heat spreader to x-fer a lot of heat. Ryzen does. It's a materials engineering thing, there is a value called a heat transfer coefficient which when multiplied by the temp difference gives you a heat flux value aka thermal power transferred per unit area. The thermal power transfer per unit area on a Gen 10 is better than Gen 9 and any Zen. It's like having a factory delid.
If that's the case, then this is something that AMD still has to learn. It's also one more reason not to jump to conclusions on how hot the 12900K is going to run based purely on its peak power consumption.
Posted on Reply
#93
Why_Me
ViperXTRhere comes more leaks
gofile.io/d/sVNYVP
[MEDIA=imgur]a/r9lOl2L[/MEDIA]


AMD Ryzen 5900X (161 FPS) $524 USD

Intel Core i5 12600K (158 FPS) $289 USD


Posted on Reply
#94
owen10578
Guwapo77AMD is anything but comfortable at the top... I must say, they have been doing some very respectable IPC gains over the years. From what I see, they aren't asleep at the wheel either.
In terms of innovating? No they definitely are still going hard at it. But in terms of pricing? Yes they are. Look at the 5600X. Its a 6-core still at $300 in 2021.
Posted on Reply
#95
Arcdar
RandallFlaggGen 10 doesn't need there to be a big temperature difference between the chip and the heat spreader to x-fer a lot of heat. Ryzen does. It's a materials engineering thing, there is a value called a heat transfer coefficient which when multiplied by the temp difference gives you a heat flux value aka thermal power transferred per unit area. The thermal power transfer per unit area on a Gen 10 is better than Gen 9 and any Zen. It's like having a factory delid.
Thanks for one of the few really on-point comments here beside all the "but it runs hot" or "but my cooler runs better on intel now then on amd?" comments from people who don't understand heat-dissipation, heat or energy transportation through materials, heat-gaps and bridges or aerodynamics (but saw xyz proven in a vid somewhere or a leak-picture which HAS TO BE true).

I mean I get it. Everyone wants to see comparisons between absolute equal systems - but that'll never happen. Even with the same cooler, the same PSU, the same GPU, the exact same case-fan-speeds and room-temperature you'll not get a 100% comparison. Why? The same cooler performs differently on any cpu. No matter if you're talkin 8th/9th/12th gen Intel or Intel vs AMD. They are designed for a specific purpose for a specific cpu or a "will fit all" scenario not taking into account where/how the heat-areas are and the general requirements to get the heat from zone a/b/c away but a general "just slap it on" mentality.

That's like saying "the big air intake-compressor solutions of the 80s just flatly screwed ontop of the v8's are better than custom injector/turbocharged designes with single-valve configurations because they fit more engines than those stupid ultra-specific designs that work only for one engine and only one generation of that engine even though the next engine has the same dimentions and the valves/cylinders at the same place!!!".



Every cooler manufacturer decides which brand to focus on the most and accepts that with this he'll not have a 100% perfect solution for the other brand - but one that is "working". Also they want to cover as many generations as possible which prevents them from fine-tuning the heat-dissipation to certain areas (especially as the heatspreaders on top of the cpu are also not always 100% even and you'd need to lap those too, to get the best surface contact out of your combination of cooler/CPU --- or, in the best case of course delid and have a direct-contact). But also - no one could pay for a design like that (which takes care of different heat-zones and requirements and specific preasure-fit to certain areas of the die with even a 99% perfect fit).


But we'll always hear "but my cooler (which was designed with gen xyz of intel/amd in mind) works better on this cpu with this TPD than with THIS one with a different layout! This means this generation is better/worse!


The only thing we should take from those comparisons is how well they do with which load and how much energy is needed for it - and, if you want: cooler xyz works well with this generation (or better/worse than with intel/amd in comparison) and you're good if you have that one (but doesn't automatically mean that any cooling solution rated at xyzTDP will be bad for this chip or the chip in general doesn't work with any cooling solution below this TDP or whatever. Just - this one rated for this doesn't do as well with chip X while he did great with chip Y).
Posted on Reply
#96
AusWolf
ArcdarThanks for one of the few really on-point comments here beside all the "but it runs hot" or "but my cooler runs better on intel now then on amd?" comments from people who don't understand heat-dissipation, heat or energy transportation through materials, heat-gaps and bridges or aerodynamics (but saw xyz proven in a vid somewhere or a leak-picture which HAS TO BE true).

I mean I get it. Everyone wants to see comparisons between absolute equal systems - but that'll never happen. Even with the same cooler, the same PSU, the same GPU, the exact same case-fan-speeds and room-temperature you'll not get a 100% comparison. Why? The same cooler performs differently on any cpu. No matter if you're talkin 8th/9th/12th gen Intel or Intel vs AMD. They are designed for a specific purpose for a specific cpu or a "will fit all" scenario not taking into account where/how the heat-areas are and the general requirements to get the heat from zone a/b/c away but a general "just slap it on" mentality.
I can only add a few things:

1. The average user (or even the average PC enthusiast) doesn't care about physics. They just want a CPU that works in their system.

2. The vast majority of people love drawing general conclusions out of their own specific experiences. They don't have the perspective to see further, which is not necessarily a bad thing, just a fact.

3. Speaking of perspective, the average user doesn't have the money (or willingness) to try various different systems just to see how they work, so their only source of information is the online reviews. Unfortunately, journalism in general is trending towards emotional influencing and away from factual presentations. You'll sooner hear a youtuber say "oh Jesus Christ's mother, is this chip hot" or "AMD is super efficient" than discuss how you'd have to set up your power limits with different levels of cooling and airflow. Our average user has no choice but to give credit to these (often false, or at least one-sided) comments about specific use cases that are far from what he/she needs the CPU for to begin with.

I was lucky enough to have tried several different Zen 2 and 3 chips before I settled with my Core i7-11700 as the brain in my main rig for the next few years. The 5950X is a beast and I had no problem cooling it with a 240 mm AIO in a mid tower case. I just wanted to save some money and desktop space by going SFF (once again after the last couple years). In that situation, even a 3600 didn't work - hence the 11700, which I'm absolutely happy with. I also have a 3100 which I'm also happy with in my HTPC - being the coolest and least hungry modern AMD CPU. These are all excellent CPUs, but you need to know what you want to use them for, and in what kind of system with what kind of cooling setup. It also doesn't hurt to know your options in terms of power limit configurations and expected performance. Unfortunately, the media isn't filled with such information (unlike catchy titles and results taken from extreme scenarios).

In general, I don't think it's only the user's fault that they're misinformed. The media plays a huge role. Personally, I think it's awesome to see so such distinct, but very capable architectures from Intel and AMD. :)
Posted on Reply
#97
Guwapo77
owen10578In terms of innovating? No they definitely are still going hard at it. But in terms of pricing? Yes they are. Look at the 5600X. Its a 6-core still at $300 in 2021.
They will slash prices eventually. MLD brought up a point from a business perspective, that AMD might hold off on slashing prices until after the holidays. Because AMD could be questioning Intel's supply. AMD has the stock in place for the holidays. We shall see who wins this chess game this quarter.
Posted on Reply
#98
chrcoluk
RichardsElectricity is cheap... intel wants the performance crown
4 months of my electric bill buys a 12900k. We have a electricity crisis in our part of the world right now.
Posted on Reply
#99
Arcdar
AusWolfUnfortunately, the media isn't filled with such information (unlike catchy titles and results taken from extreme scenarios).

In general, I don't think it's only the user's fault that they're misinformed. The media plays a huge role. Personally, I think it's awesome to see so such distinct, but very capable architectures from Intel and AMD. :)
Totally agree with this - and those two sentences summarize the two biggest issues and why there is so much "revolt" / "emotional distress" in forums (besides the dunning-kruger effect ^^). One of the reasons I really love TPU as wizzard and the team are far away from this behavior and besides doing extremely detailed reviews keep it very neutral and "unbiased" (yes, some "leaked-postings" are the opposite of that, but I'm exclusively talking about the testing and how much work goes into trying to keep it balanced :) )
chrcoluk4 months of my electric bill buys a 12900k. We have a electricity crisis in our part of the world right now.
Electricity and heating go crazy right now in every part of the world, but this is insane ôO .... I mean - you're not joking?
Posted on Reply
#100
The Quim Reaper
chrcoluk4 months of my electric bill buys a 12900k. We have a electricity crisis in our part of the world right now.
ArcdarElectricity and heating go crazy right now in every part of the world, but this is insane ôO .... I mean - you're not joking?
I'm from the UK and he's talking nonsense. There is no 'crisis' just energy supply companies who low balled their prices to attract customers and then got caught out by Putin's games. These companies go under but their customers are just switched to another supplier with no disruptions.

..and if he's paying £2000 for electric ( I suspect he means gas supply as well) , then that is down to his wastefulness. My combined Gas & Electric bill is just £60 a month for comparison, with Electric accounting for about £40 of that.
Posted on Reply
Add your own comment