Introduction

Intel Core Ultra 7 270K Plus leads a new 2-part refresh of the company's Core Ultra Series 2 Arrow Lake family of desktop processors. These aren't just speed-bumps over existing SKUs, but come with increased CPU core counts, more L3 cache, minor increases in clock speeds, faster uncore frequencies, and a brand new software-side update that makes games more aware of Intel's contemporary x86 architecture, extracting more performance. The other SKU from today's release is the Core Ultra 5 250K Plus, which we've reviewed here.

Intel's Core Ultra Series 2 Arrow Lake generation received mixed reviews despite the company significantly increasing the IPC of its E-cores over the previous generation, and massively gaining on the energy efficiency front, thanks to the new TSMC 3 nm node for the Compute tile, a node more advanced than even the TSMC 4 nm that AMD uses for its Ryzen 9000 Zen 5 series. On its own, the new Lion Cove P-core is touted to offer an IPC increase over the previous Raptor Lake core, but it lacks support for Hyper-Threading, and since the CPU core counts haven't increased over the previous generation Raptor Lake, the overall thread count of the processor has reduced. With its 2026 refresh of Arrow Lake, Intel attempts to address this shortfall in thread counts by increasing core counts for the two new SKUs over the chips they are intended to replace.



The Core Ultra 7 270K Plus replaces the current Core Ultra 7 265K, it launches at an aggressive price of $300, which is a whole $100 cheaper than what the 265K launched at. While the 265K is carved out of the Arrow Lake-S silicon with a core-configuration of 8 performance cores and 12 efficiency cores (8P+12E) with 30 MB of L3 cache, the new 270K Plus maxes out the silicon, enabling the entire 8P+16E cores present, along with the entire 36 MB slab of L3 cache. All that sets the 270K Plus apart from the flagship Core Ultra 9 285K, on paper, is a slightly lower maximum P-core boost frequency of 5.50 GHz vs. 5.70 GHz, resulting from a lack of Thermal Velocity Boost algorithm. But wait, there's more. The 270K Plus and 250K Plus come with new Intel Binary Optimization technology that all previous Arrow Lake chips, including the 285K, lack support for, at least currently.

Intel Binary Optimization Technology is a new software feature-addition by Intel for the 270K Plus and 250K Plus, which builds on top of Intel Application Performance Optimization (APO). It provides a user front-end for APO, letting software change the way machine code from a game's binaries is executed by the processor, to better utilize the various innovations Intel made for its microarchitecture. This feature is an opt-in by users, and Intel releases game-specific optimization through regular software updates under its new Intel Platform Performance Package (IPPP) releases. Much like monthly graphics driver updates by GPU manufacturers, IPPP will keep the drivers of the various processor- and chipset-integrated devices up-to-date, and bring in new or updated profiles for Binary Optimization.

The underlying Intel Arrow Lake microarchitecture remains unchanged from Intel's 2024 debut. Intel's first desktop processor to implement a disaggregated tile-based chip design realizing the IDM 2.0 vision of former CEO Pat Gelsinger, Arrow Lake combines a Compute tile based on TSMC 3 nm foundry node. This contains the CPU complex consisting of up to 8 Lion Cove P-cores, and up to 16 Skymont E-cores, sharing 36 MB of L3 cache. The SoC tile, built on the TSMC 6 nm node, contains the processor's integrated memory controllers, a 13 TOPS NPU, and a PCI-Express Gen 5 root complex, which it wires out with a little assistance from a breakout I/O tile built on the same node. The Graphics tile, built on TSMC 5 nm, contains the processor's Xe-LPG iGPU with up to 4 Xe cores.

As stated earlier, the Core Ultra 7 270K Plus maxes out the Arrow Lake-S silicon, enabling all 8 P-cores, and all 16 E-cores, along with the entire 36 MB L3 cache. The P-cores boost up to 5.50 GHz, while the E-cores go up to 4.70 GHz. The chip lacks the Thermal Velocity Boost feature that the flagship 285K offers, but gives you classic Turbo Boost 2.0, and Turbo Boost Max 3.0, for maximum boost frequency bins across a set number of UEFI CPPC preferred cores. The chip comes with 125 W processor base power, and 250 W maximum turbo power. Intel has increased the die-to-die frequencies of the fabric connecting the Compute and SoC tiles, by 900 MHz over what the 265K comes with. Intel already offers a die-to-die frequency increase that's covered by warranty, under the "Core 200S Boost mode" feature that it released as a UEFI BIOS setting through motherboard vendors and OEMs last year.

Perhaps the biggest "feature" of the new 270K Plus is its price, with Intel pricing it at just $300. The 265K launched at $400, before price-cuts brought it down to roughly $340, less for the KF variant. The aggressive launch price is Intel doing its bit to offset brutal memory and SSD prices that are putting gamers off from building or upgrading their rigs. With this price, Intel also places AMD on notice, the 270K Plus squares off against a plethora of Ryzen 9000-series SKUs, including the Ryzen 7 9800X3D, Ryzen 9 9900X, and Ryzen 7 9700X, which it hopes to comprehensively outclass.