Microsoft's latest attempt to optimize rendering workloads has delivered impressive performance gains, with Intel's Battlemage GPUs seeing a significant boost.
Microsoft Manages to Optimize Complex Rendering With SER, Leading to Improved Rendering Performance
As rendering pipelines become increasingly complex each day, the traditional approach to such workloads has begun to create a performance bottleneck, and Microsoft's SER addresses an important one. For those unaware, SER stands for Shader Execution Reordering, and it is a feature in Shader Model 6.9, meaning we will see SER become a standard with the upcoming driver code. I won't make SER too complex to understand, but before we dive into the technicals, let's see the potential performance improvements Microsoft witnessed.
Microsoft tested SER in a sample test, where they witnessed a 90% frame rate increase with Intel's 'B-series' GPUs, and a 40% bump with NVIDIA's GeForce RTX 4090. Interestingly, SER isn't an entirely new implementation, as NVIDIA has already integrated it with Path Tracing optimizations, resulting in decent gains across rendering workloads. If you are curious about whether you could see these performance bumps, it is important to note that Microsoft's testing doesn't entirely reflect an actual gaming workload, so the difference with it might be lower.
Shader Execution Reordering focuses on improving ray tracing efficiency, especially when a single ray hits multiple objects, each requiring a different shader. Usually, when different shaders needed to be processed, every thread in a warp would need to wait for the others, leading to massive idle time. With SER, all objects hit by the ray are stored 'somewhere' and then reordered based on spatial locations and shader similarities. This allows execution in a much more 'coherent' manner.
The combination of HitObject and SER is particularly powerful and enables reordering for execution and data coherence using information in the HitObject and additional hints supplied by the user. The result is further improved coherence potential for hit/miss processing.
- Microsoft
If you are thinking about how to benefit from SER, well, you don't need to worry at all, since the feature is integrated with Microsoft's Shader Model 6.9, which requires AgilitySDK 1.619. We aren't aware of any hardware limitations for now, but considering Microsoft's demo, it is likely that Ada Lovelace and above, along with Intel's Battlemage GPUs, will support SER.
About the author: Muhammad Zuhair is a hardware and technology reporter for Wccftech, specializing in the semiconductor industry and the complex interplay between technology, manufacturing, and geopolitics. His coverage focuses on the corporate strategies and technological roadmaps of industry giants like TSMC, NVIDIA, Samsung, and Intel. Zuhair's expertise lies in deconstructing complex topics such as fabrication nodes (e.g., 2nm process), the economic impact of policies like the CHIPS Act, and the strategic development of AI infrastructure from NVIDIA, AMD and Intel.
Follow Wccftech on Google to get more of our news coverage in your feeds.
