Intel Core Ultra 9 285K

[oxrufiioxo]

der8auer got 8800 MT working without going to gear 4. Some interesting performance results there when compared against lowest common denominator testing at 6000 MT, same RAM kit between rigs. Zen 4/5 have to run out of sync to get 8000 MT, due to that, generally offers no gaming benefit.

View attachment 368764

View attachment 368766

Definitely seems like some strange behaviour, depending on game. These 1% lows are crazy low, considering P cores have more dedicated cache, buffering memory. I suspect it's early BIOS/software shenanigans. Looking forward to W1z's memory scaling articles, and later testing with more mature platform software and Win updates.

View attachment 368765

It's still unfortunate that after intel watched amd release a seemingly half baked cpu in need of both windows and bios updates they decided to do the same thing on a more expensive platform that also loses in gaming overall vs it's outgoing one....

 

[dgianstefani]

It's still unfortunate that after intel watched amd release a seemingly half baked cpu in need of both windows and bios updates they decided to do the same thing on a more expensive platform that also loses in gaming overall vs it's outgoing one....

I'm curious to see if it does lose in gaming when tuned and with mature software, compared to tuned Raptor Lake, which topped out at around 8/8200 MT memory (if you're lucky, and have a two DIMM $500+ board), or even stock vs stock in a few months. Looks like there's potential there, even in this state of release.

The Z890 platform does seem better than X870E X670E, IMO.

 

[btk2k2]

While I get that it also hampers Zen 4/5 performance somtimes by a little somtimes by a lot. Although the patched version of 23H2 isn't horrible I guess.

There are other reviews that tested amd on 24H2 and Arrowlake on 23H2 so it isn't an issue in the grand scheme of things but using somthing that benefits one architecture but not the other still isn't ideal.

I get it though w1z was put in a difficult position and wanted as much parity with platform as possible dealing with these half baked cpus.

That is poor form to use 23h2 for AMD given 24h2 is available now.

 

[LittleBro]

May I ask where are all those dudes that yelled at Zen 5 regarding how bad it was at launch, unfinished, unpolished, delivering only few digits generational performance uplift while also lowering power draw? Also, where are those who screamed that AMD is responsible for the disastrous performance in Windows, meaning they should have worked with Microsoft more and solve scheduler issues before launch?

Or is it different, because this is Intel?

Anyway, another Intel claims not delivered. I'm not surprised. Intel is one claim producing machine, that's all.

So, power draw is lower but so is the performance. It's like the same or slightly lower perf we already had but with lowered consumption. Gaming performance is pathetic, Intel's own two generations old CPU beats this shiny super piece of silicon. BSODing, poor performance on newest OS version. Even with node advantage Intel failed to surpass Zen 5's efficiency. Not good.

This is bad for us consumers. This year is exceptionally terrible. But hey, let's take another look at this situation - Panther Lake and Zen 6 are just around the corner! Another series of claims are now to be made about future products and god only knows whether they will get delivered. I don't care anymore.

Next: Intel becomes best chip producer in 2025 with their amazing Intel 18A process. We shall see. Or not.

 

[oxrufiioxo]

I'm curious to see if it does lose in gaming when tuned and with mature software, compared to tuned Raptor Lake, or even stock vs stock in a few months. Looks like there's potential there, even in this state of release.

The problem is we can only go by what was actually released

People in both camps are starting to sound the same though hoping microcode/bios/windows updates will save the day but it would need a massive improvement to beat it's real gaming competition the 9000X3D serues which even if only 3-5% better than 7000X3D will be almost a generation in front gaming performance wise. 2 going by this generation.

That is poor form to use 23h2 for AMD given 24h2 is available now.

I don't like it either but I get it and fell be probably chose the best middle ground it's not like zen5/4 is broken on patched 23H2 it just loses a couple percent and that really wouldn't change the overall conclusion on Arrow Lakes gaming performance that in its current state kinda sucks.

 

[AnotherReader]

I'm curious to see if it does lose in gaming when tuned and with mature software, compared to tuned Raptor Lake, which topped out at around 8/8200 MT memory (if you're lucky, and have a two DIMM $500+ board), or even stock vs stock in a few months. Looks like there's potential there, even in this state of release.

The Z890 platform does seem better than X870E X670E, IMO.

The gap in TPU's tests is too large to be overcome after Zen 5 X3D releases. Note that these tests don't use 24H2 where the Ryzens would gain even more.

 

[Bwaze]

"I just hope both teams have fun."

Because for us consumers this is a very underwhelming year.

 

[chrcoluk]

May I ask where are all those dudes that yelled at Zen 5 regarding how bad it was at launch, unfinished, unpolished, delivering few digits generational performance uplift while lowering power draw? Also, where are those who scrramed that AMD is responsible for the disastrous performance in Windows, meaning they should have worked with Microsoft more and solve scheduler issues before launch?

Or is it different, because this is Intel?

Anyway, another Intel claims not delivered. I'm not surprised. Intel is one claim producing machine, that's all.

So, power draw is lower but so is the performance. Gaming performance is pathetic, two generations old CPU beats this shiny super piece of silicon. BSODing, poor performance on newest OS version. Even with node advantage Intel failed to surpass Zen 5's efficiency. Not good.

This is bad for us consumers. This year is exceptionally terrible. But hey, let's take another look at this situation - Panther Lake and Zen 6 are just around the corner! Another series of claims are now to be made about future products and god only knows whether they will get delivered. I don't care anymore.

Next: Intel becomes best chip producer in 2025 with their amazing Intel 18A process. We shall see. Or not.

Still here, and I said exactly the same thing about scheduling for Intel on these chips as I did for AMD.

From what I can tell the response to this launch is largely negative, so I dont know why you are so upset if thats what you wanted.

From a selfish point of view I am glad neither set of chips is a big improvement, as it means my current setup stays relevant for longer, rapidly moving tech is only ok if you got the pockets to keep upgrading. I also especially dont want to be changing motherboards considering how much regression has happened in that market.

 

[dgianstefani]

The problem is we can only go by what was actually released

People in both camps are starting to sound the same though hoping microcode/bios/windows updates will save the day but it would need a massive improvement to beat it's really gaming competition the 9000X3D which even if only 3-5% better than 7000X3D will be almost a generation in front gaming performance wise. 2 going by this generation.

The thing is, as usual, the X3D chips are only good for gaming. They're slower than the non X3D chips as well as the Intel competition in anything that isn't gaming, which happens to be what the vast majority of people in the world use CPUs for (not gaming). They're also more expensive.

Compare, say, a mainstream segment $330 9700X against the $310 245K, you're essentially getting 8% more application performance per dollar with the 245K, a more modern platform, and generally it's more efficient, while having slightly slower gaming prowess (with 6000 MT and early firmware). The 7800X3D is both more expensive and 10% slower in applications, but 20% faster in gaming (when the 245K is tested with slow memory), for $490.

What I'm seeing with the $590 285K is a CPU that compares favorably against its more expensive competition ($649 9950X), 30% more efficient in ST

essentially the same in MT

25% less power in idle

...plus a better platform, but currently it's slightly slower in gaming despite being more efficient, when tested with memory 2000 MT slower than Intel's "sweetspot" 8000 MT.

 

[Oberon]

I dunno - many feel AMD is just one good IMC implementation away from boosting performance... that and maybe addressing any IF / interconnect limitations between CCDs.

AMD's IMC is quite good - better even than Intel's at running high speed memory (just look at the APUs); the problem is routing the chiplet interconnect over the package, which severely limits the bandwidth and speed. Hopefully Zen 6 will finally be the iteration that moves to 2.5D packaging for the CCDs and IOD, which would very likely remove that bottleneck.

 

[RandallFlagg]

Where'd all the people from the Zen 2 days who only cared about rendering/encoding speed and power efficiency go??

Tom's said it best - lateral move. Nothing here for gamers, mostly for people doing encoding and some niche scientific uses. People who *actually* care about that, prob aren't using a CPU though.

Looks like a 14700K will be my next upgrade. LGA 1700 FTW

 

[Dr. Dro]

What a bitterly disappointing result. Wow. Seems that it can't beat my i9-13900KS (early access 14900K lol) at anything that isn't "muh AI" or some other things I don't really use my gaming PC for, and even then it's quite a narrow margin. And to think I had some FOMO going on when I decided to get my Z790 Apex Encore.

The highlight here was really the CUDIMM preview. I can't wait for the Trident Z5 CK's to be available worldwide, hopefully sooner rather than later. The 9600 MT/s model got listed $389 on Newegg, so it's gonna be a bit expensive, but totally worth it. Just hope my CPU's IMC can take it, at least into the 8000's. 7600 was easy with my existing 6800 kit.

 

[dgianstefani]

The gap in TPU's tests is too large to be overcome after Zen 5 X3D releases. Note that these tests don't use 24H2 where the Ryzens would gain even more.

All the CPUs gain more in 24H2, not just Ryzen, besides ARL for some unknown reason, likely patched soon.

 

[Bwaze]

The thing is, as usual, the X3D chips are only good for gaming. They're slower than the non X3D chips as well as the Intel competition in anything that isn't gaming, which happens to be what the vast majority of people in the world use CPUs for (not gaming). They're also more expensive.

But ironically, gaming is one of the rare things where small differences in CPU performance even matter to buyers - for home users who don't really rely for CPU speed to do any time critical rendering, video or audio editing etc... Even though most of gamers don't own RTX 4090 and don't game at low resolutions where these differences even matter.

 

[chrcoluk]

Where'd all the people from the Zen 2 days who only cared about rendering/encoding speed and power efficiency go??

Tom's said it best - lateral move. Nothing here for gamers, mostly for people doing encoding and some niche scientific uses. People who *actually* care about that, prob aren't using a CPU though.

Looks like a 14700K will be my next upgrade. LGA 1700 FTW

I do think the Z690 (and some Z790) boards are better balanced on i/o vs Z890, and with Bartlett coming as well, it ended up getting 4 revisions of chips, with Bartlett potentially being a 12 p-core chip as well.

 

[Metroid]

Intel = step in the right direction, now I'm waiting for Intel 3dvcache and then we will have a better gaming competition x AMD. I like the idea that SMT/HT is dead for Intel desktop cpus, that was hacker's paradise, especially now they know how to take control using those virtual threads, more security for users = step in the right direction.

 

[matar]

My i9-12900KS @ 5.3GHz All P-Cores and 4.2Ghz All E-Cores Ring @4.2Ghz DDR4 3600MHZ CL16 ( Got it late May-2024 ) I am so happy that i skipped all 13-14- Because of the reported bugs with 13-14900k and now 15 gen after watching reviews so happy i got the 12900KS.

 

[MikeSnow]

Unless there are some patches coming up that improve performance, this is comically bad, slower than a 12900K in some games.

Cyberpunk 2077 RT 720p:

@W1zzard for the last couple of months the CPU reviews have shown some strange inconsistencies between the FPS and the relative performance, not sure where they come from:

If 192.8 FPS represents 100.0%, then 218.2 FPS would represent 113.2% for the 7800X3D. But the chart on the left shows 112.6% instead.

Same for 7950X3D, 206.9 FPS would be 107.3%. But the chart on the left shows 106.2% instead.

For the 14900K it's the other way around, for 203.8 FPS it should be 105.7%. But the chart on the left shows more, 106.7%.

The inconsistencies are significant enough that the CPU positions in the two charts are different, which makes no sense, unless the relative performance is not based on FPS, but something else.

Later edit: My initial comment was referencing 13900K incorrectly, I replaced it with 14900K.

 

[AnotherReader]

All the CPUs gain more in 24H2, not just Ryzen, besides ARL for some unknown reason, likely patched soon.

While that is correct, if I recall correctly, the Ryzens gained more than Intel did.

 

[Dr. Dro]

I'm curious to see if it does lose in gaming when tuned and with mature software, compared to tuned Raptor Lake, which topped out at around 8/8200 MT memory (if you're lucky, and have a two DIMM $500+ board), or even stock vs stock in a few months. Looks like there's potential there, even in this state of release.

The Z890 platform does seem better than X870E X670E, IMO.

CUDIMMs should be compatible with Raptor Lake as well, and should similarly raise the clock frequency ceiling. Probably won't go as high as Arrow, but I think chances are good. Do you know if anyone tested it, now that some reviewers have a kit in their hands?

I don't think any regular Z890 boards will be doing 8000-8200MT on standard memory (without a clock driver) other than the Z890 Apex and its usual counterparts, though

 

[oxrufiioxo]

Where'd all the people from the Zen 2 days who only cared about rendering/encoding speed and power efficiency go??

Tom's said it best - lateral move. Nothing here for gamers, mostly for people doing encoding and some niche scientific uses. People who *actually* care about that, prob aren't using a CPU though.

Looks like a 14700K will be my next upgrade. LGA 1700 FTW

Yeah, everything this cpu is good at can be done better on a gpu.....

This does feel like Zen2 though decent application performance but meh gaming uplifts.

The thing is, as usual, the X3D chips are only good for gaming. They're slower than the non X3D chips as well as the Intel competition in anything that isn't gaming, which happens to be what the vast majority of people in the world use CPUs for (not gaming). They're also more expensive.

Compare, say, a mainstream segment $330 9700X against the $310 245K, you're essentially getting 8% more application performance per dollar with the 245K, a more modern platform, and generally it's more efficient, while having slightly slower gaming prowess (with 6000 MT and early firmware). The 7800X3D is both more expensive and 10% slower in applications, but 20% faster in gaming (when the 245K is tested with slow memory), for $490.

What I'm seeing with the $590 285K is a CPU that compares favorably against its more expensive competition ($649 9950X), 30% more efficient in ST View attachment 368773

essentially the same in MT

View attachment 368772

25% less power in idle

View attachment 368774

...plus a better platform, but currently it's slightly slower in gaming despite being more efficient, when tested with memory 2000 MT slower than Intel's "sweetspot" 8000 MT.

View attachment 368775

It preforms like a 7950x in w1z applications and gaming benchmark aggregate I fail to see how that is impressive after 2 years.... we are talking a couple percent better if we don't cheery pick what each is good at.....

Same problem the 9950X had really.

 

[dgianstefani]

CUDIMMs should be compatible with Raptor Lake as well, and should similarly raise the clock frequency ceiling. Probably won't go as high as Arrow, but I think chances are good. Do you know if anyone tested it, now that some reviewers have a kit in their hands?

I don't think any regular Z890 boards will be doing 8000-8200MT on standard memory (without a clock driver) other than the Z890 Apex and its usual counterparts, though

They should be, but will those platforms get BIOS updates?

Dubious.

 

[oZ65]

In this steaming mess that Windows has become, one is now forced to see the results on Linux, and results on Linux have spoken.

Kudos to Intel for being a generation behind in performance, and probably two generations behind in consumption. Kudos indeed, it was not easy.

Intel Core Ultra 9 285K "Arrow Lake" Delivers Strong Linux Performance Review - Phoronix

www.phoronix.com

 

[W1zzard]

@W1zzard for the last couple of months the CPU reviews have shown some strange inconsistencies between the FPS and the relative performance, not sure where they come from:

Average FPS averages the FPS, so higher numbers have more weight. Relative performance normalizes each result to "100" for the tested product and then averages. Neither is "wrong". Can't normalize FPS or you lose the absolute value, not a fan of geomean for this chart because it doesn't solve the ordering problem and just makes things complicated for people to understand who barely know what "percent" means

 

[dgianstefani]

In this steaming mess that Windows has become, one is now forced to see the results on Linux, and results on Linux have spoken.

Kudos to Intel for being a generation behind in performance, and probably two generations behind in consumption. Kudos indeed, it was not easy.

Intel Core Ultra 9 285K "Arrow Lake" Delivers Strong Linux Performance Review - Phoronix

www.phoronix.com

View attachment 368777

The 24 physical cores with the Core Ultra 9 285K was enough to outperform the 32-thread Ryzen 9 9950X in many code compilation tasks, some MPI workloads, and more, but in areas leveraging AVX-512 like AI and many creator workloads, the AMD Ryzen 9 9900 series continued to dominate.
At pricing of $589~599 USD for the Core Ultra 9 285K, this top-end Arrow Lake desktop processor is around $100 less than the current pricing on the Ryzen 9 9950X. This is a compelling difference such as for a developer/build box and areas where the Core Ultra 9 285K was particularly very competitive but if running a lot of AVX-512 workloads and areas where Zen 5 was delivering striking wins, the Ryzen 9 9950X and the ~$429 Ryzen 9 9900X can deliver great value.
For Linux gamers the Core Ultra 9 285K doesn't look particularly compelling but for developers, (non-AVX-512) creator workloads, and many other multi-threaded productivity workloads there were nice generational improvements over Raptor Lake, really great performance-per-Watt, and competiveness to the AMD Ryzen 9900 series processors.

Conclusion of the review you took a chart from.