Friday, August 21st 2026

Intel "Nova Lake-S" 28-Core SKU With bLLC to Draw Nearly 300 W at PL2

Intel's upcoming "Nova Lake-S" desktop processors are nearing launch, and the rumor mill is intensifying as we receive more information about clock speeds, core configurations, and power consumption. According to a post on the Weibo forums, Intel's 28-core SKU with a big last-level cache (bLLC) integrated into the package will reach up to 296 W of power consumption at the Power Level 2 (PL2) setting. This is the maximum power draw the CPU can achieve during short bursts of turbo clocking, reaching its peak. Compared to the current top-end "Arrow Lake-S" 24-core Core Ultra 9 285K, which has a PL2 setting of 250 W, the additional four LPE-Cores on the latest "Nova Lake-S" will increase that number by about 18.4%.

Previously, we learned that "Nova Lake-S" will be very power-hungry, especially in the top-end SKUs. For example, Intel's top-end, overclocking-focused, dual compute die "Nova Lake-S" CPU with 52 cores will feature a PL2 mode with a power level of 474 W. This is an unprecedented number for an Intel consumer desktop platform, typically expected from a high-end desktop (HEDT) design. However, since this is a 52-core design that can be overclocked, it represents the highest CPU PL2 setting that Intel will be shipping with "Nova Lake-S." For single compute die SKUs, the PL2 limit appears to be 296 W.
Finally, this could be the previously leaked Intel Core Ultra 9 4950K, identified as a 28-core "Nova Lake-S" desktop chip. It features 8 "Coyote Cove" P-Cores, 16 "Arctic Wolf" E-Cores, and 4 LPE-Cores. Intel is testing many SKUs for "Nova Lake-S," and a four-digit naming scheme could be necessary due to the numerous SKUs and configurations, so the new naming scheme might be on the way.
Source: HXL on X
Add your own comment

125 Comments on Intel "Nova Lake-S" 28-Core SKU With bLLC to Draw Nearly 300 W at PL2

#26
BoggledBeagle
Dr. DroAccording to your table, it is clearly at 200 W. Which, mind you, produces similar scores to Raptor Lake at 320 W. ARL is crazy power efficient compared to its predecessor
It is not clear to me at all. You can see the energy saving to performance loss ratio ( = benefit of limiting it ) dropping the more you limit it, but that is a smooth curve, there is no clear point to choose on it.

Somebody would have to define, that when a CPU has the maximal power limit established, with which it can survive long enough, it should be sold with the power limit corresponding to the value of the ratio say 4 for a mainstram version and 2 for the ECO version?
Posted on Reply
#28
Dr. Dro
BoggledBeagleIt is not clear to me at all. You can see the energy saving to performance loss ratio ( = benefit of limiting it ) dropping the more you limit it, but that is a smooth curve, there is no clear point to choose on it.

Somebody would have to define, that when a CPU has the maximal power limit, that it can survive long enough, it should be sold with the power limit corresponding to the value of the ratio say 4 for a mainstram version and 2 for the ECO version?
I might suck at maths, but that's a 13% power reduction for a 2% performance loss. That is clearly the point where the voltage/frequency curve begins to yield poorer results.

The CPU will run for a very long time in its rated, safe specification, that isn't even up for debate IMHO
Posted on Reply
#29
JustBenching
BoggledBeagleIt is not clear to me at all. You can see the energy saving to performance loss ratio ( = benefit of limiting it ) dropping the more you limit it, but that is a smooth curve, there is no clear point to choose on it.

Somebody would have to define, that when a CPU has the maximal power limit, that it can survive long enough, it should be sold with the power limit corresponding to the value of the ratio say 4 for a mainstram version and 2 for the ECO version?
35w is where the intel efficiency is maximized. What I mean by that is the maximum ratio of performance to power. For amd dual CCD chips it's around ~50-60 watts.
Posted on Reply
#30
bug
Darmok N JaladNeeding 2 times the energy to get 20% more performance very much suggests that there is a sweetspot. If there was no ideal, then 2x the power would get you 2x the performance, which clearly isn't happening--it's not even close. Many machines are designed to operate under a range of conditions, but they are most efficient under specific conditions. Once you start moving past those conditions, either the power-in becomes no longer worth the performance-out, and/or the machine simply fails. In the case of a 270K, Intel designed it past the ideal but stopped somewhere acceptably short of failure. AMD does the same, as does NVIDIA, as in all their designs, you can drastically reduce power demands while not surrendering an equal proportion of performance. In many cases, a lot more power only gets them 5%, but that can be the difference between winning and losing a benchmark, which is what helps determine pricing. A losing product typically sells for considerably less than the winning product. In saner times, at least.
There's always a sweet spot. Of the many equation I was taught in uni that governed transistors' functions, not one was linear ;)
Posted on Reply
#31
yfn_ratchet
Yeesh. Part of why Alder (and to a leser extent, Raptor) Lake was so lukewarm to me was its nuclear power draw and thermal output. It was damn cheap near the end of LGA1700's life cycle, probably still is, but the thermals put me off. My room is not perfectly ventilated and heat gets trapped in here. The relatively meager ~250W my entire system puts out gets my room notably toasty, regardless of my efforts.

I remember when both AMD and Intel tried their hand at reducing power profiles on their CPUs some years ago and people lost their minds. "What do you mean my CPU doesn't suck down as much as it posssibly can OOTB for an extra 1-5%?!" I exaggerate, but that's how it read to me. Some people appreciate having a reasonably power-light CPU whose performance metrics are well advertised at that power level. Like me. I like my 65W CPUs.
Posted on Reply
#32
Darmok N Jalad
bugThere's always a sweet spot. Of the many equation I was taught in uni that governed transistors' functions, not one was linear ;)
Yeah, my 5800XT can be set at a few different Eco modes, Even as low as 45W, it will still boost to the same max single core speed as the out-of-box setting (which is 150MHz higher than the rated boost clock). What it won't do is operate at the same all-core speed. All core load is where these CPUs really get rowdy. Each core can certainly handle the voltage and frequency on its own, but when combined, they make for one hot slag of power-dense silicon. Nova Lake S doesn't need all this power to reach its single core peak performance. These high PL2s are for multicore wins.
Posted on Reply
#33
Sol_Badguy
JustBenchingOkay bro

You should learn to read. :slap:
I clearly said 9950X3D not 9950X3D2.
You can see where each is positioned in that chart, along with 270K and 285K.
Posted on Reply
#34
BoggledBeagle
JustBenching35w is where the intel efficiency is maximized. What I mean by that is the maximum ratio of performance to power. For amd dual CCD chips it's around ~50-60 watts.
I experienced only improving efficiency with decreasing power limit, and then the CPU breaks and fails to work properly with low enough limit. I found no optimal power limit for the highest efficiency.

So if 35W is a power limit with which the CPU is still fully functional and highly efficient, its total performance is very poor. I would never run 270K with 35W power limit, it makes no sense at all to limit a 24 core CPU so drastically.
Posted on Reply
#35
bug
yfn_ratchetYeesh. Part of why Alder (and to a leser extent, Raptor) Lake was so lukewarm to me was its nuclear power draw and thermal output. It was damn cheap near the end of LGA1700's life cycle, probably still is, but the thermals put me off. My room is not perfectly ventilated and heat gets trapped in here. The relatively meager ~250W my entire system puts out gets my room notably toasty, regardless of my efforts.

I remember when both AMD and Intel tried their hand at reducing power profiles on their CPUs some years ago and people lost their minds. "What do you mean my CPU doesn't suck down as much as it posssibly can OOTB for an extra 1-5%?!" I exaggerate, but that's how it read to me. Some people appreciate having a reasonably power-light CPU whose performance metrics are well advertised at that power level. Like me. I like my 65W CPUs.
You've cheated yourself out of some good CPUs. My 12600k has been running capped between 95 and 125W, no heat issues over here. With the cooler you see in my sig, nothing exotic.
Posted on Reply
#36
Hecate91
JustBenchingOkay bro

Ok bro.

That's not the 9950X3D, and that is peak all core power consumption, not power draw across a selection of applications.
And no, setting power limits aren't a fix for this either, given how memory makers got sued for not printing JEDEC speeds on the box, the average user is too dumb to know how to tinker around in the BIOS to get power consumption under control. Desktop CPU's should have a limit of 200W. With a 300W+ CPU, rest of the system power draw, a GPU, monitor, audio setup and it will nearly max out a 1500W North American power outlet.
Posted on Reply
#37
zus
Hecate91That's not the 9950X3D, and that is peak all core power consumption, not power draw across a selection of applications.
Looking at average results, Arrow Lake and Zen 5 seem to be pretty close in power draw. 285K, 270K+, 9950X, 9950X3D are pretty close in performance and power consumption, while the X3D2 draws a little more power, and performs a bit better. But Intel won't have the node advantage this time, so I don't know how things will play out for Nova Lake vs Zen 6.
Posted on Reply
#38
kondamin
Nice, i hope they get to keep well below it stock and give access to it for those that are willing to risk their system and spend money on more exotic cooling solutions.
Posted on Reply
#39
Denver
Personally, I want my CPUs and GPUs to run at under 120W and 300W, respectively, while gaming.

No one in their right mind should support this trend toward products that are getting hotter and hotter and are major power guzzlers.
Posted on Reply
#40
truerock
For my use of a desktop PC, CPU power requirement is irrelevant.
DenverPersonally, I want my CPUs and GPUs to run at under 120W and 300W, respectively, while gaming.

No one in their right mind should support this trend toward products that are getting hotter and hotter and are major power guzzlers.
With all due respect... who cares how much power a desktop PC uses? Why is it relevant?
Posted on Reply
#41
kondamin
truerockFor my use of a desktop PC, CPU power requirement is irrelevant.


With all due respect... who cares how much power a desktop PC uses? Why is it relevant?
Here running a computer like that 4hours a day 320day in a year would cost about 230Euro's if all your power comes from the grid.
Posted on Reply
#42
Denver
truerockFor my use of a desktop PC, CPU power requirement is irrelevant.


With all due respect... who cares how much power a desktop PC uses? Why is it relevant?
Anyone whose critical sense hasn't yet been removed via brainwashing.

The lower the hardware’s power consumption, the less heat it generates. This simplifies the cooling requirements, the case, and all peripheral components. Even the motherboard design can be streamlined if the top-tier SKU supported by a given chipset has a more suitable/tame TDP. Not to mention reduced component wear, since they are more efficient. Has everyone already forgotten about the connectors catching fire and the CPUs, both X3D and i9, burning out?

There is nothing to celebrate with components seeing their TDP rise with every new generation.
Posted on Reply
#43
BoggledBeagle
truerockFor my use of a desktop PC, CPU power requirement is irrelevant.


With all due respect... who cares how much power a desktop PC uses? Why is it relevant?
It is the same relevant as gas mileage in cars, people usually care about these things.
Posted on Reply
#44
Cliff Bungalow
truerockFor my use of a desktop PC, CPU power requirement is irrelevant.


With all due respect... who cares how much power a desktop PC uses? Why is it relevant?
Anyone who pays for electricity does.

Power consumption will also affect what kind of cooler and PSU you need., and how much you spend on them.
Posted on Reply
#45
JustBenching
Power draw is indeed important, but what power it uses out of the box isn't, since it takes 5 seconds to change that, whether within the bios or within windows itself. The name of the cpu is more important than the stock power draw, that's how useless that metric is.
Posted on Reply
#46
Tyler-98-W68
JustBenching35w is where the intel efficiency is maximized. What I mean by that is the maximum ratio of performance to power. For amd dual CCD chips it's around ~50-60 watts.
I didn't do extensive benchmarking, (mostly playing around with BeamNg) but at a stock PL of 250w vs 35w, yes it looks like it lost a lot of performance but it was still very good for 35w

This is a 285K
Posted on Reply
#47
Intelko
When Rentable Units?
When APX?
When Unified Core?
When full UMA where the memory is truly allocated in dynamic way like on Apple Silicon?

When will we finally have all of this?
Posted on Reply
#48
thestryker6
DenverPersonally, I want my CPUs and GPUs to run at under 120W and 300W, respectively, while gaming.

No one in their right mind should support this trend toward products that are getting hotter and hotter and are major power guzzlers.
My 270K Plus can pull 350W under an extreme load, ~280-290W under heavy load and games under 100W most of the time so I'm not sure what the point you think you're making is.

The only thing anyone should care about is hardware running efficiently at any given workload regardless of how much power it's consuming. If someone has a problem with absolute power consumption power limits exist for a reason or just don't buy top end hardware.
Posted on Reply
#49
Dr. Dro
Cliff BungalowAnyone who pays for electricity does.

Power consumption will also affect what kind of cooler and PSU you need., and how much you spend on them.
Indeed, hope you're ready to invest on a Ruby certified unit that has >91% efficiency at 5% load, otherwise, you might have to pay the European Union some additional carbon tax over having such grossly inefficient, wasteful 80+ Platinum unit wasting *gasp* 10% of all the power it consumes, even more if you haven't placed enough load on it. How dare you run any sort of electronics that require more power than the common LED bulb? This is why we have dynamic kWh pricing by the hour.
IntelkoWhen full UMA where the memory is truly allocated in dynamic way like on Apple Silicon?
Hopefully never. I like my computers upgradable, so no unified memory for me.
Posted on Reply
#50
Hecate91
thestryker6The only thing anyone should care about is hardware running efficiently at any given workload regardless of how much power it's consuming. If someone has a problem with absolute power consumption power limits exist for a reason or just don't buy top end hardware.
I care because electricity isn't getting any cheaper, anyone who pays their own power bill will usually care about a CPU chugging down 300W, a more power hungry CPU also requires additional cooling, a high end motherboard, and a larger case.
Also it seems like people forgot that even high end hardware used to not be extremely power hungry, there was OC room if you wanted your PC to double as a room heater.
Posted on Reply
Add your own comment
Nov 12th, 2026 05:54 PST change timezone

New Forum Posts

Popular Reviews

TPU on YouTube

Controversial News Posts