No, what is significant about Strix Halo is its 256bit width RAM.
I don't really see that as significant. History has shown, that people within the AI/LLM field are willing to pay near limitless amounts of cash to get what they want.
So, based on that, why is 256-bit bus significant when there is hardware out there with 512-bit bus?
Know that Strix Halo has a RDNA3.5 iGPU and as such does not support hardware-based, machine learning (ML), DLSS-like, upscaling. Only RDNA4 does that and it's only available in dedicated desktop GPU cards.
QuoteThe Ryzen AI Max+ 395 APU packs the impressive Radeon 8060S iGPU with 40 CUs based on RDNA 3.5, allowing performance that is in the same league as an Nvidia GeForce RTX 4070 Laptop GPU.
3dmark.com/search:
Time Spy:
4060 Laptop: Average score: 10506
Radeon 8060S iGPU: Average score: 10022
4070 Laptop: Average score: 12532
Steel Nomad (Vulkan):
4060 Laptop: Average score: 2432
Radeon 8060S iGPU: Average score: 2628
4070 Laptop: Average score: 3060