Intel Core Ultra 9 285K vs. Intel Core i9 14900K

Just a few weeks in the past, we in contrast Intel’s new Core Ultra 9 285K in opposition to AMD’s Ryzen 7 9800X3D in 45 video games to see how the 2 in contrast throughout a large vary of titles. The finish outcome was a one-sided defeat, with the 9800X3D delivering 24% extra efficiency on common. This isn’t ideally suited for Intel and definitely not an incredible end result for these hoping to purchase considered one of their new Arrow Lake CPUs for gaming.

From that comparability, it was clear that spending round $630 on the Ultra 9 285K for gaming is a foul thought, not less than for now. Since then, we have additionally in contrast the 9800X3D with the Core i9-14900K in 45 video games. While the i9 put up a stronger struggle, the Zen 5-based 3D V-Cache half was nonetheless, on common, 18% quicker.

But all this testing obtained us questioning: how do the 285K and 14900K evaluate throughout such a big number of video games?

Although we have already established that this comparability technically is not related (as players should not be both a Core i9 or Core Ultra 9 processor), we’re extra fascinated with simply seeing how they evaluate.

There are two most important issues we need to uncover right here. First, does the 6% margin favoring the 14900K over the 285K, seen in our day-one evaluate, translate to greater benchmarks with 45 video games? Second, we might like to determine baseline gaming efficiency for the 285K throughout a broad vary of video games, which can be utilized to trace the progress of any updates.

Although we have already accomplished that with the 9800X3D evaluate, Intel initially claimed parity between the 285K and 14900K. They’re pretty evenly matched general, although even within the day-one evaluate, we noticed some giant fluctuations in efficiency in both course. It might be fascinating to see how they evaluate throughout a a lot bigger vary of video games.

Test Setup

For testing, all CPU gaming benchmarks might be performed at 1080p utilizing the GeForce RTX 4090. If you’d prefer to be taught why that is one of the simplest ways to guage CPU efficiency for video games in the present day and sooner or later, take a look at our explainer for extra particulars. The 14900K has been examined utilizing the intense profile with DDR5-7200 reminiscence, whereas the 285K has been paired with DDR5-8200 reminiscence, particularly the brand new CUDIMM selection.

There was numerous confusion surrounding this matter, with many individuals not realizing that the DDR5-8200 reminiscence we paired with the Arrow Lake CPUs in our day-one critiques was CUDIMM reminiscence. Some claimed that had we used CUDIMM reminiscence, these new Intel CPUs would carry out a lot better, however this merely is not the case. Technically talking, in the event you run CUDIMM and UDIMM reminiscence on the identical frequency and timings, the ensuing efficiency might be an identical.

The benefit of CUDIMM reminiscence is that it might probably extra simply obtain larger frequencies whereas sustaining stability. So, whereas CUDIMM reminiscence ought to permit larger speeds, if each reminiscence sorts can obtain the identical frequency and have the identical timings, the efficiency would be the identical. In any case, we’re utilizing CUDIMM reminiscence and have accomplished so ever since we first examined Arrow Lake.

Now, let’s dive into the information…

Benchmarks

The Last of Us, Cyberpunk, Hogwarts Legacy, ACC, Spider-Man

As seen in our day-one evaluate, the 285K performs surprisingly nicely in The Last of Us Part 1, the place it is 17% quicker than the 14900K when evaluating the typical body price. Unfortunately, that result’s an outlier, as we discover it is 8% slower in Cyberpunk, 4% slower in Hogwarts Legacy, and eight% slower in ACC. We do see a small efficiency enchancment in Spider-Man, with a 7% uplift on this instance.

Baldur’s Gate 3, Homeworld 3, APTR, Flight Simulator, Starfield

Next, we’ve Baldur’s Gate 3, the place the 285K is 6% slower. It’s not a large margin, however any efficiency regression with a brand new technology is regarding. It’s additionally 3% slower in Homeworld 3, but it surely’s the 17% efficiency loss in A Plague Tale: Requiem that stands out, with 1% lows down by 32%. Thankfully, efficiency stays about the identical in Microsoft Flight Simulator and Starfield. While not nice information, we’re not less than not seeing any regression in these examples.

Horizon Forbidden West, Horizon Zero Dawn, Watch Dogs, Far Cry 6, T&L

Moving on to Horizon Forbidden West, the 285K was 7% quicker than the 14900K, a uncommon uplift. Unfortunately, it was 10% slower in Horizon Zero Dawn. It was additionally 12% slower in Watch Dogs: Legion, 12% slower in Far Cry 6, and three% slower in Throne and Liberty.

Hitman 3, Callisto Protocol, SoTR, Halo, Warhammer 3

This subsequent batch of video games exhibits largely aggressive outcomes, that means the 285K is ready to sustain with the previous-generation mannequin. The 14900K was solely barely quicker in Hitman 3, but it surely was 14% quicker in The Callisto Protocol. We then have a tie in Shadow of the Tomb Raider, a small 7% win for the 14900K in Halo Infinite, and comparable leads to Warhammer III.

Black Ops 6, Borderlands 3, Riftbreaker, Remnant 2, SWJS

Next, in Call of Duty: Black Ops 6, the 14900K and 285K had been neck and neck, reaching 296 fps. The 285K was simply 4% slower in Borderlands 3 and 13% slower in The Riftbreaker. Then we’ve Remnant II, the place efficiency was almost an identical, adopted by Star Wars Jedi: Survivor, the place the 285K was 5% slower.

War Thunder, Skull and Bones, Returnal, Ratchet & Clank, Dying Light 2

The 285K does not fare as nicely in War Thunder, coming in 19% slower than the 14900K, but it surely was 6% quicker in Skull and Bones. There was comparable efficiency in Returnal and Ratchet & Clank: Rift Apart, adopted by a 6% efficiency loss in Dying Light 2.

Forza Horizon 5, Forza Motorsport, Gears 5, Ghost of Tsushima, Hunt

In this subsequent batch of outcomes, the 285K was slower than the 14900K in all 5 examples. It was simply 4% slower in Forza Horizon 5, however 17% slower in Forza Motorsport and 13% slower in Gears 5. Ghost of Tsushima had blended outcomes: the 285K was 2% slower when trying on the common body price, however 9% quicker for the 1% lows. Unfortunately, the robust 1% low efficiency did not maintain up in Hunt: Showdown, the place the typical body price dropped by 3%, and the 1% lows had been hit with a large 22% drop.

World War Z, F1 24, Rainbow Six Siege, Counter-Strike 2, Fortnite

Our subsequent batch of outcomes exhibits extra of the identical: the 285K was 6% slower in World War Z, 5% slower in F1 24, 6% slower in Rainbow Six Siege, and had comparable efficiency in Counter-Strike 2. Then, we’ve Fortnite, which sadly nonetheless does not work on any of the brand new Arrow Lake CPUs. The sport freezes and locks all the system because of a compatibility subject with Easy Anti-Cheat. Intel is conscious of the difficulty, however a repair has not been issued but.

Assassin’s Creed x2, Space Marine 2, SW Outlaws, Dragon Age: The Veilguard

Finally, within the final 5 video games examined, the 285K is, at greatest, barely slower than the 14900K. For instance, it was 5% slower in Assassin’s Creed Mirage, simply 1% slower in Valhalla (although the 1% lows had been poor), and the most effective case outcome was in Space Marine 2, the place efficiency was about the identical. Then we noticed a 6% loss in Star Wars Outlaws and as much as a 15% efficiency loss in Dragon Age.

45 Game Average

In our day-one evaluate, the 14900K was 6% quicker on common, although the typical was calculated a bit in another way utilizing the geometric imply. The level is, the 14900K continues to be quicker, 5% throughout the 45-game pattern, with a number of vital outcomes and only a few examples the place the 285K is definitely quicker.

The greatest outcome for the 285K was in The Last of Us Part 1, however exterior of that, there is not a lot we are able to level to. Meanwhile, the 14900K was quicker by double digits in 9 of the video games examined, and 20% or extra in three of them.

What We Learned

So, there you’ve gotten it – affirmation of what we found when first testing the Core Ultra 285K. It’s round 5% slower than the 14900K for gaming, however at instances, it may be a lot slower than that. In phrases of general gaming efficiency, the 285K sits between the 12900K and 14900K, which is an honest outcome given how little energy it makes use of as compared.

However, a efficiency regression from the earlier technology is not one thing shoppers prefer to see, even when there are large energy financial savings on provide. In this particular instance, the ability effectivity enhancements are particularly related, contemplating AMD is providing significantly higher gaming efficiency whereas utilizing even much less energy.

Across this identical vary of video games, the 9800X3D is eighteen% quicker than the 14900K and 24% quicker than the 285K, which places Intel in a good place proper now. My hope is that Intel can ship on their promise of improved Arrow Lake efficiency, which is scheduled for supply this month. We’ll have to attend and see what comes of that.

If they’ll not less than match the 14900K’s gaming efficiency, that can assist, as productiveness efficiency and energy consumption are strong. However, for elements just like the 285K to turn into viable choices, Intel wants to handle each the gaming efficiency and compatibility points. They additionally want to scale back the worth – $630 is just too a lot for what the Core Ultra 9 affords.

That mentioned, we’re unsure Intel plans to promote many of those processors anyway. They do not seem to have made lots of them, and resupplies have been nearly non-existent, not less than within the U.S., which is odd provided that demand is extraordinarily low. In Australia, it is truly very straightforward to buy a 285K, as most retailers have inventory. I’m not stunned, although, contemplating the asking worth is $1,100 AUD, whereas the 9950X, for instance, prices $970 – fairly a bit cheaper.

We’d prefer to reiterate that the 285K was benchmarked utilizing CUDIMM reminiscence, so we’re testing Arrow Lake in its optimum out-of-the-box configuration. Speaking of CUDIMM reminiscence, it isn’t a silver bullet for Arrow Lake – it does not radically enhance efficiency over common reminiscence. In reality, as we famous earlier, given the identical frequency and timings, it does not enhance efficiency in any respect. That’s in all probability an excellent factor, because it’s each laborious to search out and really costly.

Most DDR5-8000 kits price north of $300, which is absurd. However, we have seen an 8200 equipment with comparable speeds to the equipment we use for testing retail for simply shy of $200, which continues to be costly, however a lot better than $300. For reference, the reminiscence we use to check the 9800X3D prices $110 for a 32GB equipment. If you are going to purchase an Arrow Lake CPU, our recommendation is to choose up a DDR5-7200 CL34 equipment for round $120. Performance is almost an identical to these premium 8200 CUDIMM kits for a fraction of the worth.

That’s going to do it for this evaluate. We do plan to revisit this testing as soon as Intel rolls out their efficiency repair.

Shopping Shortcuts: