Thursday, January 15th 2026
Core Ultra 9 290K Plus "ARL-S Refresh" CPU Geekbenched - Test Shows ~9% Uplift Over 285K
Recent Geekbench Browser database entries have outed test scores and basic specifications for two of Intel's upcoming "Arrow Lake Refresh" flagships. On January 13, a not-yet-launched Core Ultra 9 290HX Plus mobile unit was tracked in a new Acer Predator Helios gaming laptop. A few hours earlier, Team Blue's Core Ultra 9 290K Plus desktop chip entered the fray onboard a test rig built around Gigabyte's top-level and world record-holding Z890 AORUS TACHYON ICE mainboard configured with 48 GB DDR5-8000 of system memory. This leak shows the future 24-core "Arrow Lake Refresh" range topper posting overall scores of 3456 single-core and 24610 multicore points in Geekbench 6.5 test scenarios. Naturally, news reports have quickly compared the Core Ultra 9 290K Plus to a nearby non-refreshed "Arrow Lake" part—Core Ultra 9 285K.
The newcomer managed to build upon the current flagship's officially ranked achievements (based upon multiple entries); doing 7% better in single-core stakes, and 9% in multicore tests. When compared to AMD's fanciest "Granite Ridge" desktop offering—Ryzen 9 9950X3D—the Core Ultra 9 290K Plus is 2% faster in terms of single-core results, and 11% above in multicore. Interestingly, the lesser Core Ultra 9 270K Plus model was Geekbenched in December—producing 3235 (single-core) and 21368 (multicore) tallies. The flagship "Arrow Lake-S Refresh" SKU is about 7% and 15% faster (respectively) than its (also) 24-core sibling. According to Geekbench diagnostic data and other specification leaks, the two (8P + 16E) siblings feature minor differences in core clock speeds, so it is possible that Team Blue engineers have tuned the more expensive option to a greater degree. Golden Pig Upgrade has proposed a March/April launch window for the Core Ultra 200 Plus desktop family, consisting of the two aforementioned models and an 18-core (6P + 12E) Core Ultra 5 250K Plus SKU.
Sources:
Geekbench Browser, VideoCardz, Guru3D, Wccftech (perf. comparison graphic)
The newcomer managed to build upon the current flagship's officially ranked achievements (based upon multiple entries); doing 7% better in single-core stakes, and 9% in multicore tests. When compared to AMD's fanciest "Granite Ridge" desktop offering—Ryzen 9 9950X3D—the Core Ultra 9 290K Plus is 2% faster in terms of single-core results, and 11% above in multicore. Interestingly, the lesser Core Ultra 9 270K Plus model was Geekbenched in December—producing 3235 (single-core) and 21368 (multicore) tallies. The flagship "Arrow Lake-S Refresh" SKU is about 7% and 15% faster (respectively) than its (also) 24-core sibling. According to Geekbench diagnostic data and other specification leaks, the two (8P + 16E) siblings feature minor differences in core clock speeds, so it is possible that Team Blue engineers have tuned the more expensive option to a greater degree. Golden Pig Upgrade has proposed a March/April launch window for the Core Ultra 200 Plus desktop family, consisting of the two aforementioned models and an 18-core (6P + 12E) Core Ultra 5 250K Plus SKU.




31 Comments on Core Ultra 9 290K Plus "ARL-S Refresh" CPU Geekbenched - Test Shows ~9% Uplift Over 285K
Maybe I'm blind but I see Arrow Lake beyond No. 30? That's not terrible?
I'm just sorely disappointed because Intel lagging in their game is bad for competition and progress. If they up their game, so does AMD and we get faster, better CPUs at a much quicker pace. Intel's decisions seem to be boneheaded executive decisions rather than engineering decisions. No one asked for the dumb E-core. Why can't they offer products with only P-cores? It's the same type of mad, stubborn behavior that got them shipping Pentium 4 for years.
In raw floating point for compute tasks, 285K has 225000 versus only 165000 for the 9950X through PassMark's suite of tests. Whatever the issue with some productivity tasks are, it is not the cores as they have a ton of power, Intel probably just failed to have these tasks properly optimized. I have Intel's Arrow Lake P-cores have around ~2330 FP/Ghz/core IPC and AMD's Zen 5 cores at ~1820 FP/GHz/core IPC based on PassMark. For reference, Skylake was around 1145 FP/GHz/core.
But then someone would say "oh but these CPUs wouldn't be targeted at the MT productivity crowd they would be targeted at something that scales past 8 P-Cores but isn't Blender, NAMD or whatever". Okay then what would that workload be, gaming??? I suppose some games would benefit from extra P-Cores but what % are those from the totality of games?
Also there is the power draw issue, such a CPU would suck on those EPS connectors like crazy in full load.
Comparing such a CPU with 12P or 14P vs 8P+16E for roughly similarly sized compute tiles, both 12P and 14P would draw more power in full load but lose in MT. They would fare better in gaming, sometimes due to the extra P-Cores and sometimes due to not having scheduling issues.
Then there is the cost issue, would such a CPU be cheaper to make? Correlating cost with performance and the result is not great. There has to be a balance, everything is a compromise, even if that sounds negative it's pretty much what a balanced product is, a sum of reasonable compromises, whenever there is a clear handicap then that product isn't balanced.
If we're talking just 8P for gaming, the problem is that it still wouldn't perform great, sure there won't be scheduling issues but the inter-tile latency would still be there and extra L3 cache wouldn't help, maybe a bit but it won't solve the issue like super speed RAM which was expensive before the fake crisis.
So, despite the lackluster performance in gaming, Intel isn't that stupid, sure Arrow Lake should've stayed in the oven for a couple more months, but realistically that wouldn't have changed the outcome, meaning AMD would've still dominated in gaming as such the reception of Arrow Lake might have been warmer but gamers would have still bought AMD.
Anyhow, it’s a skip able generation
That way they could've released a better polished product that would've still had the lackluster performance in gaming but they would've owned it at least, and that pretty much would've been the only problem with the gamer demographic, in other scenarios Arrow Lake would've been considered a worthy alternative.
Nowadays it is considered as a worthy non-gaming alternative by more informed consumers, but they are a minority. The disappointing launch has tainted the perception for many people and it took big discounts (for mobos as well) to change some of that perception.
Intel could've played the same cards a little better.
my 285k, not yet heavily tuned.
As for the 9995WX, I doubt that they are up to any shenanigans; it has low clocks compared to the 9950X. When its clocks approach that of the latter, it consumes a lot of power.