Intel's rumoured Razor Lake-AX processor will reportedly not be unlocked for overclocking, according to new details shared by leaker Jaykihn. The chip is set to feature Nova Lake CPU architecture on a large APU platform rather than a conventional socketed desktop processor.

Jaykihn (via VideoCardz) claims that Razor Lake-AX relies on Coyote Cove P-Cores and Arctic Wolf E-Cores. This architecture retains more Nova Lake technology than its name implies, separating it from Razor Lake-S/H, which should use a different P-Core and E-Core configuration.

Previous leaks linked Razor Lake-AX to a BGA4326 package featuring on-package memory alongside 16 or 32 Xe3p graphics cores. These physical specs indicate that AX is a large APU platform rather than a standard desktop CPU.

Official branding isn't clear yet, with questions remaining over whether it falls under the Core 400 or 500 series. Intel hasn't officially announced Razor Lake-AX, but a release is expected to occur in late 2027 or early 2028.

Discuss on our Facebook page, HERE.

KitGuru says: Do you think Intel can compete against AMD's Halo series APUs with this platform?

Performance metrics allegedly belonging to AMD's Ryzen AI Max+ Pro 495 APU have appeared on the Geekbench 7 database. The Gorgon Halo processor was apparently tested inside an HP ZBook Ultra G3a, a 16-inch mobile workstation configured with 96GB of RAM.

In the Geekbench 7 listing spotted by Gray (via VideoCardz), the APU achieved a score of 2593 points in the single-core test and 26816 points in the multi-core test. Clock monitoring logged the chip running at a boost frequency of 5207MHz, which matches AMD's official specifications listing a 5.2GHz boost clock.

The listing reveals a 16-core and 32-thread configuration running at a 3.1GHz base clock, mirroring the core layout of the previous Max+ Pro 395. This follows an earlier PassMark appearance in May, where the processor was benchmarked inside a similar HP machine equipped with 192GB of RAM. Integrated graphics duties rely on a Radeon 8065S based on the RDNA3.5 architecture.

Official AMD details confirm the Gorgon Halo architecture features up to 16 cores and 32 threads, 5.2GHz boost clocks, 55TOPS of performance from its XDNA2 NPU, and 40TOPS from RDNA3.5 CUs hitting up to a 3GHz boost.

KitGuru says: The benchmark results align closely with what we've been hearing about the Ryzen AI Max 400 series, pointing to a clock-boosted refresh of the current Halo APU generation rather than an entirely new series.

AMD's Zen 6 Low Power core will be a mixture of various Zen core architectures but will be able to match or exceed Zen 5 cores. AMD Is Making A Third Zen Core Starting Zen 6 Which Aims To Offer Low Power Characteristics, But With Similar Performance Output Starting with Zen 6, AMD will include three core types: a classic core, a dense core, and a low-power core. The company has already introduced Zen 6 classic and Zen 6C dense cores for its EPYC 9006 family, but the third core is yet to see any product announcements. It looks like […]

Read full article at https://wccftech.com/amd-zen-6-low-power-core-to-match-or-exceed-zen-5-collage-of-zen-architectures/

The company has come a long way since its acquisition of ATI Technologies in 2006. This transformed it from a CPU maker to a full-fledged computing giant. Twenty Years Later, the ATI Acquisition Remains AMD's Most Important Milestone, Which Helped in Driving Innovation from Radeon Graphics to AI Accelerators On July 24, 2006, AMD announced one of the biggest acquisitions in semiconductor history by acquiring ATI Technologies in a deal worth approximately $5.4 billion. At that time, it was seen as a big gamble that would put AMD directly against giants like Intel and NVIDIA. Exactly two decades later, it […]

Read full article at https://wccftech.com/20-years-since-amds-5-4-billion-ati-acquisition/

AMD's next-generation Zen 6 CPUs will include a brand-new core type, which will aim to bring low-power mode to several platforms. After Zen Classic and Zen Dense, AMD Is Working On A Third Core Type For Zen 6 With Low-Power In Mind AMD's current Zen 5 lineup includes two major core types based on the same ISA (Instruction Set Architecture). These include the Zen 5 and Zen 5C variants. The Zen 5 is the standard variant, and the Zen 5C focuses on higher core densities but with optimized clocks/power characteristics. We have seen the inclusion of these two Zen core […]

Read full article at https://wccftech.com/amd-zen-6-new-low-power-core-beyond-zen-6-and-zen-6c/

A benchmark entry for AMD's Ryzen AI 5 435G has appeared in the Geekbench database, marking the first public performance data for the upcoming Ryzen AI 400 desktop series.

The Ryzen AI 5 435G is a 6-core, 12-thread APU designed for the AM5 socket. AMD specifications list the chip with a 2.0GHz base clock, a boost frequency up to 4.5GHz, and a 65W TDP. The unit features Radeon 840M integrated graphics based on the RDNA 3.5 architecture with 4 compute units, alongside an XDNA 2 NPU rated for up to 50 TOPS.

According to the Geekbench entry shared by @Olrak29_ (via VideoCardz), the processor achieved a single-core score of 2620 and a multi-core score of 10594. The log indicates the CPU reached a maximum frequency of 4094MHz during the test, which sits below the official 4.5GHz boost specification. The entry also confirms a hybrid-style core configuration featuring two CPU clusters: one with 2 cores and another with 4 cores. The processor was tested on a JGINYUE B850M Snow Dream WIFI motherboard equipped with 32GB of memory and running Windows 11.

Compared to public performance averages, the Ryzen AI 5 435G single-core score of 2620 slightly exceeds the Ryzen 5 8600G average of 2493. Multi-core performance is comparable to the 8600 G's 10857 points, though it trails the 8-core Ryzen 7 8700G, which averages 13478 points. AMD previously confirmed that Ryzen AI 400 series desktop systems are expected to be available from OEMs starting in Q2 2026.

Discuss on our Facebook page, HERE.

KitGuru says: Considering the maximum frequency was slightly below its specifications, we assume the chip will probably score higher once it hits retail.

AMD's roadmap for integrated graphics is becoming increasingly complex as the company appears to be looking to bridge the gap between its current offerings and the next-generation UDNA architecture. New commits to the LLVM compiler suggest that the previously spotted GFX1170 target, branded as RDNA 4m, is expanding, as two additional software IDs, GFX1171 and GFX1172, have surfaced.

As spotted by Phoronix in the LLVM compiler, these GFX117x targets are not “true” RDNA 4 parts, which reside in the GFX12 branch. Instead, they appear to be a custom evolution of RDNA 3.5. By backporting specific RDNA 4 modules into the GFX11 instruction set, AMD is creating what we might call RDNA 3.5+. This hybridisation adds support for INT8 and FP8 data types, which are essential for running the machine-learning-based upscaling and frame-generation features of FSR 4.

The decision to modify RDNA 3.5 rather than jump straight to a full RDNA 4 iGPU isn't clear, but one might guess it's due to limitations shrinking the full RDNA 4 architecture down for things like thin and light laptops, where a smaller power envelope is crucial. By updating RDNA 3.5 with WMMA (Wave Matrix Multiply-Accumulate) and SWMMAC instructions, AMD might be ensuring that its upcoming APUs can leverage FSR Redstone features.

While “Medusa Point” is expected to use an RDNA 4m iGPU, “Medusa Halo” is expected to leap straight to the RDNA 5/UDNA microarchitecture.

Discuss on our Facebook page, HERE.

KitGuru says: While RDNA 4m is mostly a rebrand of RDNA 3.5, the addition of native FP8 support could be enough to enable FSR 4. 

New LLVM patches confirm additional RDNA 4m variants alongside GFX1170 with similar instruction capabilities. AMD Quietly Expands Next-Gen iGPU Roadmap by Adding New RDNA 4m Graphics Targets, GFX1171 and GFX1172 Earlier compiler patches showed that RDNA 4m incorporates several RDNA 4-style instruction-set upgrades, such as support for FP8 and BF8 data formats, along with WMMA matrix instructions used in AI workloads. Now, AMD is further expanding its next-gen integrated GPU roadmap, as new Linux compiler patches unveil two additional RDNA 4m graphics targets. These are the GFX1171 and GFX1172. These targets are now joining the recently discovered GFX1170, confirming that […]

Read full article at https://wccftech.com/amd-adds-rdna-4m-gpu-targets-gfx1171-and-gfx1172-hinting-at-next-gen-zen-6-apu-graphics/