Benchmark #552

Download
EEVDF vs SCX schedulers in Rocket League kernel 6.11

Rocket League first two minutes of match replay no additional workload Ryzen 5700X RTX 3060Ti EPP powersave / balance_performance

Submitted 1 month ago by yubysowhat

Specifications
Label OS GPU CPU RAM OS specific
Bpfland Kernel 6.11 CachyOS NVIDIA GeForce RTX 3060 Ti AMD Ryzen 7 5700X 8-Core Processor 17 GB 6.11.2-2611EEVDF powersave
EEVDF Kernel 6.11 CachyOS NVIDIA GeForce RTX 3060 Ti AMD Ryzen 7 5700X 8-Core Processor 17 GB 6.11.2-2611EEVDF powersave
LAVD Kernel 6.11 CachyOS NVIDIA GeForce RTX 3060 Ti AMD Ryzen 7 5700X 8-Core Processor 17 GB 6.11.2-2611EEVDF powersave
Rusty Kernel 6.11 CachyOS NVIDIA GeForce RTX 3060 Ti AMD Ryzen 7 5700X 8-Core Processor 17 GB 6.11.2-2611EEVDF powersave
# Top runs: * **Highest FPS**: `Rusty Kernel 6.11` achieves the highest average FPS, making it the most efficient in terms of frame rate among the runs. * **Smoothest FPS**: `Rusty Kernel 6.11` provides the smoothest experience with the lowest FPS variance and standard deviation, indicating consistently stable performance. * **Best overall**: `Rusty Kernel 6.11` offers the best overall performance due to its combination of high FPS and smooth frame delivery, optimizing both speed and stability. # Issues: * The configuration using the `LAVD Kernel 6.11` provides significantly lower FPS (over 10% less compared to others) alongside a higher standard deviation, indicating both reduced performance and less smooth gameplay. This could potentially be a concern if maximum performance is a priority. # Summary This benchmark assesses the performance impact of different schedulers, specifically comparing `EEVDF` and `sched_ext` variants like `Rusty` Kernel 6.11, in the context of a Rocket League match replay. Testing on a consistent platform with Ryzen 5700X and RTX 3060 Ti under the 6.11 kernel in `powersave` mode, it is observed that `Rusty` Kernel 6.11 stands out due to its efficient frame rate and stability. On the other hand, `LAVD` Kernel 6.11 encounters a significant dip in performance, suggesting its scheduler implementation may not be as optimized for this use case relative to its counterparts. Additionally, while `Bpfland` and `EEVDF` implementations perform well, they don't quite match the high frame consistency and peak FPS achieved by the `Rusty` configuration.