AMD Strix Point APUs With RDNA 3+ iGPU Should Match RX 6400 With 12 CUs, RTX 3050 With 16 CUs

AMD's Strix Point APUs featuring the RDNA 3+ iGPU should offer comparable performance to entry-level discrete GPUs as indicated in new performance rumors.

AMD's Strix Point APUs are going to be a full-on overhaul, bringing brand new core IPs on all fronts. The CPU side will be upgraded to Zen 4 cores, the GPU side will be upgraded to RDNA 3+ cores and the NPU side will be upgraded to the latest XDNA 2 "Ryzen AI" architecture. Besides that, Strix APUs will come in two configurations, a standard monolithic and a premium chiplet design. AMD will be designating the monolithic variants as Strix Point 1 and the chiplet variants as Strix Point 2. Both of these will feature the same architectures but come in vastly different configurations.

You can tell by its name that the chiplet design will be the one to look out for as it is designed for enthusiast PCs & while it will take the performance of APUs to the next level, the standard monolithic design will also be offered a good improvement in terms of graphics performance.

It was recently mentioned by @Xinoasassin, an insider with good knowledge of AMD and Intel architectures, that the RDNA 3+ iGPUs featured on the Strix Point APUs will feature performance comparable to entry-level discrete GPU solutions. Xino stated that the 12 Compute Unit config is rated at around 3150 points in 3DMark Time Spy at a constrained TDP of 22-24W & we can expect close to 4000 points in the same benchmark with the full configuration.

AMD Ryzen 8050 Strix Point Mono Expected Features:

For comparison, the AMD Radeon 780M (RDNA 3) iGPU offers around 2300-2400 points in the Time Spy benchmark at the same TDP and averages around 2800 points at peak TDP on laptops. The same chip also averages around 3300-3500 points using the full unconstrained desktop TDPs. At the same TDP, that's a performance uplift of around 35% which is great for what's mostly going to be the same architecture with some tuning and faster clock speeds. The AMD Radeon RX 6400 & NVIDIA GeForce RTX 2050 discrete notebook GPUs also average around 3500 points in 3DMark Time Spy and both GPUs run at a higher TDP.

The full 16 Compute Unit configuration of the AMD RDNA 3+ iGPU featured on the high-end Strix Point monolithic dies should be able to match the GeForce RTX 3050 GPU (35-50W) which is going to be great for the mainstream gaming segment. AMD's RDNA 3 iGPUs have already shown very strong performance leadership in the integrated space, competing with the likes of Intel's Arc Xe-LPG architecture and sitting ahead of them in the majority of gaming titles. They are getting good driver support and features such as Ray Tracing, FSR upscaling, and FSR Frame Generation work great. We can see further improvements and tech innovations supported by the next-gen RDNA 3+ graphics architecture.

It was recently reported that AMD's RDNA 3+ graphics architecture will last in Ryzen APUs till 2027 so it's going to be around for a while. We may see optimizations down the road like we got to see with Vega iGPUs. Now we also reported on how the AI PC craze might have nerfed some of these chips as they were anticipated to include other technologies such as a larger cache that would have helped boost performance for both CPUs and GPUs. Having bigger iGPUs also comes with the cost of bandwidth scaling.

GPUs require higher bandwidth and there's only so much that you can do. A likely solution is to implement an on-die cache such as Infinity Cache used by discrete GPUs but that's unlikely for Strix Point. The possible route is to ship these premium laptops with faster LPDDR5x memory solutions which will prove a major benefit, plus there are also talks about the LPDDR6 in the industry which can further eliminate some bandwidth bottlenecks from these chips. Intel's Lunar Lake CPUs are going with the on-package LPDDR5x memory route but there's still no onboard GPU cache seen in the latest block diagram so it looks like we still have some time before we see more innovative chip-packaging technologies brought down to SOCs to help elevate graphics performance on the said chips.

These chips will also mark a big upgrade for Mini PCs and handheld gaming PCs which have taken the gaming industry by storm.