Benchmark #379

Download
Bpflands in Rocket League

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

Submitted 2 months ago by yubysowhat

Specifications
Label OS GPU CPU RAM OS specific
BORE CachyOS NVIDIA GeForce RTX 3060 Ti AMD Ryzen 7 5700X 8-Core Processor 17 GB 6.11.0-rc5-2-cachyos-rc performance
Bpfland-next-a49f22c CachyOS NVIDIA GeForce RTX 3060 Ti AMD Ryzen 7 5700X 8-Core Processor 17 GB 6.11.0-rc5-2-cachyos-rc performance
Bpfland-next-a49f22c -m 0xff CachyOS NVIDIA GeForce RTX 3060 Ti AMD Ryzen 7 5700X 8-Core Processor 17 GB 6.11.0-rc5-2-cachyos-rc performance
Bpfland-next-e935cb2 CachyOS NVIDIA GeForce RTX 3060 Ti AMD Ryzen 7 5700X 8-Core Processor 17 GB 6.11.0-rc5-2-cachyos-rc performance
# Top runs: * **Highest FPS**: `BORE` with the mean FPS being significantly higher than other runs. * **Smoothest FPS**: `Bpfland-next-a49f22c -m 0xff` with the lowest standard deviation and variance in FPS. * **Best overall**: `BORE` due to its highest FPS with a slightly higher but still acceptable standard deviation. # Issues: * The run with `Bpfland-next-e935cb2` has noticeably lower average FPS compared to other runs, with performance degradation around 5% lower in mean FPS. * The run with `Bpfland-next-a49f22c` also shows lower FPS by about 4.5% and higher frame time variance, which indicates less stable performance compared to top runs. # Summary In comparing the different `bpfland` scheduler versions in Rocket League, it is evident that the baseline configuration (`BORE`) provides the highest and most consistent FPS, offering the best overall performance. The configurations with experimental changes, including specific versions of the `bpflands` and additional kernel parameters like `-m 0xff`, achieved slightly lower FPS and varied frame time stability. Thus, while experimental schedulers are promising, the baseline remains the most effective for stable, high-performance gameplay in this benchmark scenario.