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)
Add your own comment

31 Comments on Core Ultra 9 290K Plus "ARL-S Refresh" CPU Geekbenched - Test Shows ~9% Uplift Over 285K

#26
thestryker6
AntiX3DGlazerThe latency issues have nothing to do with the node and everything to do with the memory subsystem. I'm not really sure where you got the idea that 20A (a mildly competitive node vs. N3) has anything to do with the memory subsystem's issues.
You apparently think that memory is the only latency issue with ARL. Core to core and L3 latencies were all higher than prior generations and those are completely on the compute tile. Now it is certainly possible that Intel dropped the ball with the design entirely, but it cannot be discounted that they changed to a different node which had only been used for a mobile part and didn't have a ton of turn around time.
Posted on Reply
#27
AntiX3DGlazer
thestryker6You apparently think that memory is the only latency issue with ARL. Core to core and L3 latencies were all higher than prior generations and those are completely on the compute tile. Now it is certainly possible that Intel dropped the ball with the design entirely, but it cannot be discounted that they changed to a different node which had only been used for a mobile part and didn't have a ton of turn around time.
It's true. My speculation is that Intel saw the fiasco caused by Raptor Lake's degredeation issues, and decided not to clock everything to 11 for once. That includes the ring, D2D, and the cores themselves. the fact that the 20A product was cancelled doesn't tell us anything, because they developed two arrow lake dies: one on 20A and one on N3B, as a backup. In the end, they went with the backup and were still able to launch a product on time that was at least marginally competitive.
Posted on Reply
#28
igor_kavinski
AntiX3DGlazerthe fact that the 20A product was cancelled doesn't tell us anything
Disagree. It tells us that Pat Gelsinger LIED through his teeth about 20A and faced with dismal yields, decided to save face and cancel it. Everything he did was simply to fool the Intel board into keeping him afloat until he got his $100 million payday. Good thing they saw through his deception and fired him when they did.
Posted on Reply
#30
kondamin
igor_kavinskiDisagree. It tells us that Pat Gelsinger LIED through his teeth about 20A and faced with dismal yields, decided to save face and cancel it. Everything he did was simply to fool the Intel board into keeping him afloat until he got his $100 million payday. Good thing they saw through his deception and fired him when they did.
18A is just a refined 20A just like intel 3 is intel 4 with a couple of tweaks

Pat was set aside because enticed by promised government billions that never crystallised tried to build too many fabs at the same time and went into debt hard.
Never mind that he inherited a mess.

They replaced an engineer that put intel back on track technologically with a pencil pusher that is bleeding the corpse dry.
Posted on Reply
#31
XeonsNOpterons
kondaminThey replaced an engineer that put intel back on track technologically with a pencil pusher that is bleeding the corpse dry.
Yeah, unfortunate, but now maybe Intel can get back on track to being entirely on par with AMD in every regard. I am really hopeful for 18A and 14A, I think Diamond Rapids will really be the main reference point this year for 18A's true performance. Panther Lake is great and all, but it is small scale, especially compared to Diamond Rapids.
Posted on Reply
Add your own comment
Nov 17th, 2026 15:08 CST change timezone

New Forum Posts

Popular Reviews

TPU on YouTube

Controversial News Posts