ROCm GPU Compute Performance With AMD Ryzen AI MAX+ "Strix Halo"
Written by Michael Larabel in Graphics Cards on 27 May 2025 at 05:00 PM EDT. Page 4 of 6. 23 Comments.
The Intel Lunar Lake OpenCL kernel latency is much lower than on the AMD GPUs with ROCm, thanks to Intel's brand new Compute Runtime release. As shown in the Intel CR benchmarks last week, there is a massive speed-up on Lunar Lake thanks to the Ultra Low Latency Scheduling (ULLS) being enabled.
AMD Strix Halo delivers great GPU compute performance atop the ROCm open-source compute stack.
The Strix Halo Radeon 8060S performance also did well for GPU-accelerated workloads like the Darktable RAW photography software.