AMD Strix Halo in the test: the Ryzen Ai Max + 395 with Radeon 8060s is so fast

Image: Dave2d

As expected yesterday, the embargo fell on AMD‘s tests on that day, the “strix halo”. However, the media area, which AMD was concerned with modeling, was exceptionally weak with a clear US focus. Techconseil wasn’t even informed in advance, but still provides an overview.

Summary :

AMD Strix Halo Im Asus Rog Flow Z13 (2025)

Only four test reports worldwide (DAVE2D, Hardware Canucks, Notebookcheck, The Phawx) were published today at 3 p.m., a fifth “review” was published on the Chinese platform Bilibili and comes from Asus itself in the person of Tony Yu, Head of Brand ROG in China.

All tests are based on the Strix-Halo-top AMD Ryzen AI Max+ 395 model with 16 Zen-5 cores and 40 Cu Strong RDNA 3.5 IGPU called “Radeon 8060s” in the new Asus Z13 (2025) gaming stream.

AMD Ryzen Ai Max (Pro) 300 (Strix Halo) to a Kerne architecture model /
Threads Takt
(Base / turbo) Grafik L2 + L3-cache
(MB) TDP
(Watt)npu
(Tops) AMD Ryzen Ai Max 300 (Strix Halo) Ryzen Ai Max + (Pro) 395 Z5 16/32 3.0-5.1 GHZ RADEON 8060S
rDNA 3.5, 40 CUS, 2.9 GHz 80 45–120 50 Ryzen Ai Max (Pro) 390 Z5 12/24 3.2-5.0 GHz Radeon 8050S
rDNA 3.5, 32 CUS, 2.8 GHz 76 45–120 50 Ryzen Ai Max (Pro) 385 Z5 8/16 3.6-5.0 GHz Radeon 8050S
rDNA 3.5, 32 CUS, 2.8 GHz 40 45–120 50 Ryzen Ai Max Pro 380 Z5 6/12 3.6-4.9 GHz Radeon 8040S
rDNA 3.5, 16 CUS, 2.8 GHz 22 45–120 50

First reference of the Ryzen Ai Max + (Pro) 395

According to Dave2D, the Radeon 8060 with 40 CUS (Strix Point, like Radeon 890m to date, offers a maximum of 16 CU) in games the performance level of a GeForce RTX 4060 laptop GPU (test) with 100 watt TDP, which in the in the 100 watt TDP, which in the in the in the in the in the Test of TechTip 17% faster GEFORCE RTX 4070 laptop GPU is not realized.

With which TDP the APU achieves this result is not yet clear: Dave2d speaks of a maximum of 60 watts, HXL mentions up to 120 watts for the sum of CPU and GPU. Techconseil would have liked to create clarity directly on this point, but without a model, this was not possible.

Screenshot (82) (Bild: Dave2d)

The CPU performance is at a very high level in Cinebench: with 1,654 points in Cinebench 2024 (multi-core), the CPU beats the Ryzen AI HX 370 “strix point” with four Zen-5 and eight Zen-5c cores by 50%.

Screenshot (79) (Bild: Dave2d)

Performance scaling for this benchmark works almost linearly through a wide ribbon at TDP levels.

Huge I/O die with IGPU, more compact CCD

AMD’s Halo Strix “Monster-APU” includes four different variants in the Ryzen Ai Max (+) 300 series. The not monolithic (one matte) APU for the first time is composed of a large i/o-die with the 40-cu-igpu and two CCDS together.

Screenshot (83) (Bild: Dave2d)

The two Zen 5 CCDS, each with 8 cores, are smaller than the known Zen 5 CCDs from the Granite Ridge series (Ryzen 9000). This is mainly due to the 42% smaller connection with the Infinity fabric, which has apparently been implemented in the strix-halo-CCD with TSV. There is no confirmation of AMD yet.

Strix Halo vs. Strix Point, Krackan Point, and Hawk Point Refresh

The service available on this basis enables a completely new class of continuous compact solutions with 16 ZEN-5-CPU cores and 40 CUS of the RDNA-3.5 graphics solution.

In AI applications, AMD even put a GeForce RTX 4090 in the shade, although this is mainly due to the fact that Strix Halo can use 96 GB in a system with 128 GB of RAM as graphics memory, while the 4090 with 24 GB is not enough storage for the large language model used.

The following table shows the technical specifications of AMD mobile APUs for 2025 in comparison.

Halo-Series: AMD Ryzen Ai Max 300 (Strix Halo)
Max. 16 Cornwall 5, 40 Cu rDNA 3.5, 50 Topspremium-Series: AMD Ryzen AI 9,300 (strix point)
Max. 4/8 CORNWALL 5 / SE CORN 5C, 16 CU rDNA 3.5, 55 TOPSADVANDED-SERIE:
Max. 4/4 CORNWALL 5 / SE CORN 5C, 8 CU rDNA 3.5, 50 Topseinstiies-Serie: AMD Ryzen 200
Max. 8 cores Zen 4, 12 Cu RDNA 3, 16 Tops Ryzen Ai Max 300 (Strix Halo), Ryzen Ai 9 300 (Strix Point), Ryzen AI 7/5 300 (Krackan Point) and Ryzen 200 (Hawk Point) in the architecture cores comparison model /
Threads Takt
(Base / turbo) Grafik L2 + L3-cache
(MB) TDP
(Watt)npu
(Tops) AMD Ryzen Ai Max 300 (Strix Halo) Ryzen Ai Max + (Pro) 395 Z5 16/32 3.0-5.1 GHZ RADEON 8060S
rDNA 3.5, 40 CUS, 2.9 GHz 80 45–120 50 Ryzen Ai Max (Pro) 390 Z5 12/24 3.2-5.0 GHz Radeon 8050S
rDNA 3.5, 32 CUS, 2.8 GHz 76 45–120 50 Ryzen Ai Max (Pro) 385 Z5 8/16 3.6-5.0 GHz Radeon 8050S
rDNA 3.5, 32 CUS, 2.8 GHz 36 45–120 50 Ryzen Ai Max Pro 380 Z5 6/12 3.6-4.9 GHz Radeon 8040S
rDNA 3.5, 16 CUS, 2.8 GHz 22 45–120 50 AMD Ryzen Ai 9 300 (dot strix) Ryzen Ai 9 HX 375 4 × Z5 + 8 × Z5C 12/24 2.0–5.1 GHz Radeon 890M
rDNA 3.5, 16 CUS, 2.9 GHz 36 15-54 55 Ryzen Ai 9 Hx 370 4 × Z5 + 8 × Z5C 12/24 2.0-5.1 GHz Radeon 890M
rDNA 3.5, 16 CUS, 2.9 GHz 36 15-54 50 Ryzen Ai 9 365 4 × Z5 + 6 × Z5C 10/20 2.0-5.0 GHz Radeon 880m
rDNA 3.5, 12 CU, 2.9 GHz 34 15-54 50 AMD Ryzen Ai 7 and 5 300 (Krackan dot) Ryzen Ai 7 (Pro) 350 4 × Z5 + 4 × Z5C 8/16 2.0-5.0 GHZ Radeon 860m
rDNA 3.5, 8 CUS, 3.0 GHz 24 15-54 50 Ryzen Ai 5 (Pro) 340 3 × Z5 + 3 × Z5C 6/12 2.0-4.8 GHz Radeon 840M
rDNA 3.5, 4 CUS, 2.9 GHz 22 15-54 50 AMD Ryzen 200/8000 Mobile (Hawk Point) Ryzen 9 270 Zen 4 8/16 4.0-5.2 GHz Radeon 780m
rDNA 3, 12 CU, 2.8 GHz 24 35-54 16 Ryzen 7 260 Zen 4 8/16 3.8-5.1 GHz Radeon 780m
rDNA 3, 12 CU, 2.7 GHz 24 35-54 16 Ryzen 7 (Pro) 250 Zen 4 8/16 3.3-5.1 GHz Radeon 780m
rDNA 3, 12 CU, 2.7 GHz 24 15-54 16 Ryzen 5 240 Zen 4 6/12 4.3-5.0 GHz Radeon 760m
rDNA 3.8 CUS, 2.6 GHz 22 35-54 16 Ryzen 5 230 (Pro) Zen4 6/12 3.5-4.9 GHz Radeon 760m
rDNA 3.8 CUS, 2.6 GHz 22 15-30 16 Ryzen 5 220 (Pro) 2 × Zen 4 + 4 × Zen 4C 6/12 3.2-4.9 GHz Radeon 740M
rDNA 3, 4 CU, 2.8 GHz 22 15-30 – Ryzen 3 210 1 × Zen 4 + 3 × Zen 4C 4/8 3.0-4.7 GHz Radeon 740m
rDNA 3, 4 CU, 2.5 GHz 12 15-30-topics: AMD processors AMD Strix Halo APU Ryzen