Saturday, February 8th 2025

Intel Core Ultra 9 275HX Defeats Core i9-14900HX In Cinebench R23 Multi-Core, While Losing Out in Single-Core

The Intel Core Ultra 9 275HX was unveiled by Big Blue at CES 2025, packing 8 Lion Cove performance cores and 16 Skymont efficiency cores for a total of 24 cores and 24 threads, thanks to Intel doing away with Hyper-Threading support. As such, we already had quite low expectations for multicore performance from Arrow Lake-HX, although a recently listed Cinebench R23 run appears to confirm the contrary. However, considering the nature of leaks, be sure to accept this information with a fair amount of skepticism until more information comes our way.

To be honest, this result is quite a doozy, for the aforementioned reason. Like its predecessor, the Core Ultra 9 275HX will also boast the same number of performance and efficiency cores, without Hyper-Threading support. But somehow, the result indicates that the 275HX handily outperforms the Core i9-14900HX in multicore performance by ~11%, despite losing out in single-core performance by the skin of its teeth. Of course, there are plenty of desktop-replacement laptops that can pull off higher multicore scores, although it would be best to stick to median figures considering that the details of the tested system are not known. Clearly, something is going on here, that I am unable to wrap my head around. It is possible that Intel has massively improved multicore scaling, but that is not what we witnessed with a recent Passmark result, where the Core Ultra 7 255H portrayed healthy improvements over its predecessor in single-core, but not quite so in multicore.
Sources: @94G8LA, Notebookcheck
Add your own comment

65 Comments on Intel Core Ultra 9 275HX Defeats Core i9-14900HX In Cinebench R23 Multi-Core, While Losing Out in Single-Core

#51
ScaLibBDP
TumbleGeorgeOkay, but this is just a consumer processor. Yes, it's not from the Ultra low power consumption series. And yes, I know that workstations are also assembled with consumer processors. But still, when you say HPC, I somehow understand it as supercomputers with many orders of magnitude more productive than a laptop.
It works in a different way.

Many HPC applications designed and developed on regular computers, that is consumer ( ! ) desktops or notebooks. It is too expensive to use big supercomputers for such R&D tasks.

If you look at a list of HPC applications of Exascale Computing Project of the Department of Energy ( for example at Argonne Leadership Computing Facility or Oak Ridge Leadership Computing Facility websites ) you will see more than 50 applications ported to Frontier or Aurora exascale supercomputers from the regular smaller scale versions.

For example, QMC Pack or Gromacs or NAMD applications will be working on your gaming PC however you won't be able to push these applications for really big problems, like let's do a simulation with 1,000,000 atoms or molecules.

The High Performance Computing ( HPC ) covers a lot of different cases, small, medium, big and very-very big in size.

PS: A well known NAMD applcation ( NAnoscale Molecular Dynamics / versions 2.09 and 2.14 ) is installed on three computers around me, on a Dell Mobile Workstation, on an ASUS TUF FX505DU Gaming Notebook, and a desktop with Intel Pentium 4 processor to verify some compatibility issues.
Posted on Reply
#52
R0H1T
Most laptops will throttle like an overheating hummer on any decent sized render, except probably MacBooks. This is why it's a pretty bad IMO on Windows even now.
kondaminI remember people rendering stuff on desktops that make this generation of laptops seem like a server farm
We'll need to see how Strix Halo works as an actual viable workstation, among Windows machines!
Posted on Reply
#53
A&P211
I like soup and my toilet smells funny.

*news to wanted to say, good night
Posted on Reply
#54
kondamin
R0H1TMost laptops will throttle like an overheating hummer on any decent sized render, except probably MacBooks. This is why it's a pretty bad IMO on Windows even now.


We'll need to see how Strix Halo works as an actual viable workstation, among Windows machines
Fairly certain that's an other one of those paper amd products that practically don't make in to peoples hands
Posted on Reply
#55
Tek-Check
GGforeverI still believe it's best to compare a fresh benchmark against the previous generation’s median value, simply because we don't know the exact configuration in question. These leaks should always be taken with a grain of salt, and our performance "leak" articles consistently include similar disclaimers.
True that. And thank you for editing the article, which is now more accurate on the face value. Much appreciated. Taking median number this early gives us comfort of trying to position new CPUs, but it does not bring us closer to truth. We also do not know the power envelope. I's debatable whether median or average is more meaningful here. Both are heavily dependent on the nature of sample. We know that samples are vastly diverse.

For more context, in future articles, it'd be very useful if you could include essential comparison with equivalent AMD CPUs, so that readers have richer picture. This allows for more context, as Intel's HX CPUs typically compete against AMD's Dragon Range and soon Fire Range. Seen from this point of view, 35K score of 275HX comes two years after 34K score of 7945HX, which is ~3% uplift, and most likely with more power, as Turbo Power of 275HX is at 160W in the spec. This would significantly shape your narrative in such articles.

www.notebookcheck.net/Intel-Raptor-Lake-HX-Refresh-analysis-Core-i9-14900HX-with-more-single-core-performance.790833.0.html


Ultimately, we will know once two very similar or the same laptops are compared.
Posted on Reply
#56
Raffles
lexluthermiesterUm...

Yup..

I think you might have misread..

However...

Because they aren't doing the math properly. This is basic 5th grade math people.
www.calculatorsoup.com/calculators/algebra/percentage-increase-calculator.php

To the find the percent increase, first subtract the initial value from the final value. Then take the difference and divide it by the initial value. Finally, multiply this number by 100 to convert the number to a percentage. This final result will represent the percent increase between the two values. So;

35481 - 31854 = 3627
3627 / 31854 = 0.1138632511
0.1138632511 * 100 = 11.38632511

The difference is 11.38% aprox.

EDIT:
All that said. 11.3% is a solid increase gen on gen.
Seems like a strangely convoluted way of doing 35481 / 31854 which results in 1.1138. That is easy enough for anyone to decipher it's an 11.38% increase.
Posted on Reply
#57
lexluthermiester
RafflesSeems like a strangely convoluted way of doing 35481 / 31854 which results in 1.1138. That is easy enough for anyone to decipher it's an 11.38% increase.
It's the proper way. The short-cut way will work, but in some situations will render an invalid result.

For everyone who reads this, no, I will not explain.
Posted on Reply
#58
tpuuser256
Tek-CheckWe don't know this. This 275HX sample could also be an upper range sample and its median could fall in future too, as different laptops with this chip will perform differently and will have different power budgets, design and cooling performance.

That's why we should never calculate percentages from one single sample with a median from another, as it gives us distorted, false picture, detached from specific devices.


I had several really good teachers who transformed my life. Sorry if you did not have one.
Good for you, but the way education is it's not expected to have good quality teaching. You got lucky I guess...
Posted on Reply
#59
Raffles
lexluthermiesterIt's the proper way. The short-cut way will work, but in some situations will render an invalid result.

For everyone who reads this, no, I will not explain.
Give me an example situation ;)
Posted on Reply
#60
Tek-Check
tpuuser256Good for you, but the way education is it's not expected to have good quality teaching. You got lucky I guess...
There is no luck if one lives in a country where society cares about education, for example in Finland, Singapore and several other countries.
Posted on Reply
#61
tpuuser256
Tek-CheckThere is no luck if one lives in a country where society cares about education, for example in Finland, Singapore and several other countries.
There is luck indeed, because you were there at the right moment ;)
Posted on Reply
#62
Tek-Check
tpuuser256There is luck indeed, because you were there at the right moment ;)
Well..., by the same token, you can push this reasoning further and say we are extremely lucky to have evolved and been born in this unprobable universe. ;)
Posted on Reply
#63
tpuuser256
Tek-CheckWell..., by the same token, you can push this reasoning further and say we are extremely lucky to have evolved and been born in this unprobable universe. ;)
Sure buddy
Posted on Reply
#64
Panch8
Is this good?


I just bought this msi vector 16 ai with 5080 and 64gb ram. Its mainly for cad use and ai rendering. Looks nice. That’s what the vendor sent but the laptop is arriving this Monday.
Posted on Reply
#65
unclewebb
ThrottleStop & RealTemp Author
Panch8Is this good?
Yes it is.

I have to overclock and go outside during the winter to get those kind of Cinebench results from my 14900HX.

A Core Ultra 9 275HX should be more efficient. That means less power consumption and less heat for the same performance.
The 275HX will perform better when power consumption is the same as a 14900HX.

Posted on Reply
Add your own comment
Nov 16th, 2026 09:41 CST change timezone

New Forum Posts

Popular Reviews

TPU on YouTube

Controversial News Posts