News Core i9-14900KF Beats Ryzen 9 7950X3D, Core i9-13900KS in Geekbench 6

[Admin]

Intel's Core i9-14900KF (Raptor Lake Refresh) processor has surfaced multiple times in the Geekbench benchmark database.

Core i9-14900KF Beats Ryzen 9 7950X3D, Core i9-13900KS in Geekbench 6 : Read more

 

[JamesJones44]

12% boost in single core and 5% boost in multi core. Seems close to the 3% claims from OEMs and Intel on multi core. The single core score seems higher than I would expect (12% over last gen).

 

[hotaru251]

does geekbench use 3d cache?

 

[peachpuff]

Is it upto 350w now?

 

[TerryLaze]

Is it upto 350w now?

It already was at that when you overclocked.
(Anything above the official 253W is overclock)
12900k already was at 330W 13900k already was at 407w.

 

[NeoMorpheus]

Might be me, but do we really need to continue bothering with "brand x its faster than brand Z" for everything?

If anything, I would prefer to compare with the model that it replaces, like prices, new features, power consumption, items needed in case the person wants to upgrade etc and then after that, we can start with the fanboi triggers.

 

[Brian D Smith]

I may have misunderstood, but I thought refresh for the laptop versions were to be announced in October. Did I dream that? Meteor Lake

 

[funguseater]

Another, "New Generation beats old Generation" article, yaaaay? Isn't it expected to beat last gen components? Wouldn't the real story be if it COULDN'T beat last gen solutions?

 

[SyCoREAPER]

Another, "New Generation beats old Generation" article, yaaaay? Isn't it expected to beat last gen components? Wouldn't the real story be if it COULDN'T beat last gen solutions?

Must be a slow day lol.

Simply changing the title would have made it more relevant.

"i9-14900KF is 'x'% faster than the 13900KF, 'x'% faster than Ryzen 9 7950X3D"

Or something like that.

 

[DaveLTX]

Is it upto 350w now?

In PL4/Extreme boost mode 13900k went up to 350W
14900K? Likely 300w PL2 🤣

 

[DaveLTX]

does geekbench use 3d cache?

Not at all. Instruction footprints are small or out into dram.

 

[gg83]

does geekbench use 3d cache?

I was gunna ask the same. It seems it hates v-cache, 😆 😂

 

[gg83]

Might be me, but do we really need to continue bothering with "brand x its faster than brand Z" for everything?

If anything, I would prefer to compare with the model that it replaces, like prices, new features, power consumption, items needed in case the person wants to upgrade etc and then after that, we can start with the fanboi triggers.

Right. Even if brand x isn't better than the other, the marketing would have to find a way to make x look better. I think 1 benchmark is never accurate for overall performance.

 

[salgado18]

Another, "New Generation beats old Generation" article, yaaaay? Isn't it expected to beat last gen components? Wouldn't the real story be if it COULDN'T beat last gen solutions?

Yes, but this one gives numbers. The possibility of a 12% increase in single core performance might make someone hold on an upgrade to when it launches. It may be useless to you (and me, I'm not in the market for a cpu), but it has some numbers, so it is useful to someone.

 

[TJ Hooker]

Not at all. Instruction footprints are small or out into dram.

Soruce? What are you referring to by "instruction footprint"?

 

[pointa2b]

Lets see how Zen 5 compares against it.

 

[bit_user]

does geekbench use 3d cache?

This boils down to at least 3 different questions, the way I see it:

1. Does Geekbench 6 stand to gain from Ryzen 7000X3D VCache, if present?
2. When running the single-threaded benchmark, on a 7950X3D, does the thread get assigned to the CCD with the 3D VCache?
3. Depending on #2, is that because the benchmark is reliably assigned that CCD (or the other), or did they explicitly select one or the other?

I think the answers should be clear enough, if you compare the single-threaded scores of 7950X, 7950X3D, and 7800X3D.

 

[ilukey77]

Lets see how Zen 5 compares against it.

yeah ive got an idea the 8950x is going to render the 14900k a waste of sand thats not to mention if they get the x3d stacking right what the 8950x3d will do to the 14900k in gaming as well !!

The only good cpu in the 14th line up seems to be the 14700k and only for the boost in cores and being somewhat a last hurrah for the LGA1700 platform !!

Intel needs to get their finger out of their arse and follow AMD ( personally i think it a big ask now )
By sorting their power and efficiency adding X3d and giving customers a longer socket life span!!

Otherwise they will get gobbled up by AMD spare me the Intel fan boys BS that intel has anything going for it now !!
( love the wait till 15th and 16th they will be beasts ) is laughable at best AMD will counter with a longer socket life and x3d rendering Intel pointless again )

we need Intel to be competitive again but with them losing to AMD across so many platforms hand held , server , console and slowly taking DIY !!

I dont see it happening unless BIG changes are made !!

they hold the crown for production gaming with the 13900k but gaming cpu's still out sells production cpu's..

Take myself no doubt the 13900k holds the crown but WHY would i bother with a 13900k when all i play is games the 7800x3d is just as good in games in most cases cheaper cooler and overall more efficient on a longer life platform ????

 

[Amdlova]

I only care about the 14700k

 

[bit_user]

Intel needs to ... follow AMD ( personally i think it a big ask now )
By sorting their power and efficiency adding X3d and giving customers a longer socket life span!!

Well, let's not forget that the original plan was for their Gen 14 desktop CPU to be Meteor Lake-S, built on the Intel 4 node. So, it's not like they don't know what needs to happen. There have been some apparent problems in implementing it, reminiscent of Broadwell or Ice Lake (i.e. their 10 nm fiasco).

 

[ilukey77]

Well, let's not forget that the original plan was for their Gen 14 desktop CPU to be Meteor Lake-S, built on the Intel 4 node. So, it's not like they don't know what needs to happen. There have been some apparent problems in implementing it, reminiscent of Broadwell or Ice Lake (i.e. their 10 nm fiasco).

which has been their biggest F up this 14th refresh has been silly they needed to release 15th ( well as 14th ) with a new socket give it a 3 or 4 year life span even if it doesnt compete great with 8000series it would bring people back to intel with the socket life and sell 15th ( as the 14th ) as a competitor then focus on 15th and 16th maybe even 17th on its socket !!

As it is now AM5 is slated till 2025 so zen 6 even when Intel try to leap frog 15th and want 2or 3 gens what incentive is there to bother with 16th when AM6 is around the corner ??

At that point i will have the last hurrah 9000x3d cpu and look at investing in AMD again with AM6 for 3 to 4 years !!
I see absolutely no incentive to switch to intel with no X3d and lesser socket life ..

Unless Intel can drop a gen or 2 crushing product ( which would defeat the purpose small incremental uplifts is the way these companies both AMD and Intel make money because im 100% sure AMD are making testing the 9000 series ( so i bet all these huge 8000 uplift stories are basically the 9000 series in development ) now and just incrementally uplifting the 7000 series for 8000 )

I see no point to switching !!

 

[hotaru251]

By sorting their power and efficiency adding X3d and giving customers a longer socket life span!!

wont happen.

x3d caps speeds.

Intel's claim in cpu world is their clockspeeds.

 

[TerryLaze]

wont happen.

x3d caps speeds.

Intel's claim in cpu world is their clockspeeds.

You do understand though that intel wouldn't be using actual x3d right!?
They have their own technologies like CPU/GPU max with Gbs of cache...I didn't hear any mention of those components losing any clocks.

Also intel had clock offset for AVX before, so it's not unprecedented for them to reduce clocks.

Also finally intel's claim are the accelerators and things like thread scheduler that makes working on CPUs smarter instead of having to brute force everything.

The reason they won't do any of that for desktop anytime soon is that they don't need to.
The only thing the competition has is a server CPU that does kinda reasonably for desktop work so they can only sell it for its efficiency in brute force.

 

[rtoaht]

It already was at that when you overclocked.
(Anything above the official 253W is overclock)
12900k already was at 330W 13900k already was at 407w.

Peak power usage on a stress stress is useless

 

[bit_user]

Peak power usage on a stress stress is useless

Perhaps, if it's a synthetic test, but Cinebench tests 3D rendering performance using Cinema 4D's production renderer. So, if you're someone who does a lot of CPU-based rendering or similar work, then it's definitely relevant.

What I don't know are things like CPU-Z scores and Coremark. Perhaps those don't correlate well with performance on real-world workloads. If they fail to model a real workload, then what they tell you about a CPU is less useful.