Benchmark #170

Download
Rocket League powersave governor

Submitted 3 months ago by yubysowhat

Specifications
Label OS GPU CPU RAM OS specific
ASDF powersave CachyOS NVIDIA GeForce RTX 3060 Ti AMD Ryzen 7 5700X 8-Core Processor 17 GB 6.10.3-1-cachyos-lto powersave
BORE powersave CachyOS NVIDIA GeForce RTX 3060 Ti AMD Ryzen 7 5700X 8-Core Processor 17 GB 6.10.3-1-cachyos-lto powersave
Bpfland powersave CachyOS NVIDIA GeForce RTX 3060 Ti AMD Ryzen 7 5700X 8-Core Processor 17 GB 6.10.3-1-cachyos-lto powersave
LAVD powersave CachyOS NVIDIA GeForce RTX 3060 Ti AMD Ryzen 7 5700X 8-Core Processor 17 GB 6.10.3-1-cachyos-lto powersave
Rusty powersave CachyOS NVIDIA GeForce RTX 3060 Ti AMD Ryzen 7 5700X 8-Core Processor 17 GB 6.10.3-1-cachyos-lto powersave
## Top runs - **Highest FPS**: `BORE powersave` with an average FPS approximately 3% higher than the others. - **Smoothest FPS**: `LAVD powersave` with a standard deviation of FPS approximately 11% lower than the next smoothest run. - **Best overall**: `BORE powersave` due to its significantly higher FPS with a standard deviation only slightly higher (approximately 5%) than the smoothest run. ## Issues - **LAVD powersave**: This run is significantly worse than the others, with an average FPS around 5% lower and inconsistency shown by a lower FPS value (with the lowest FPS roughly 60% lower). ## Summary All runs used the same hardware and software, with the same Linux kernel, and scheduler, ensuring consistency in the benchmarking environment. `BORE powersave` stands out with the highest FPS, while `LAVD powersave` has the smoothest FPS but an unacceptable performance drop compared to other runs. `LAVD powersave` also has a stability issue, with an average FPS roughly 5% lower than other runs, suggesting a problem in performance settings potentially linked to the Linux scheduler in this specific scenario. Other runs (`ASDF powersave`, `Bpfland powersave`, `Rusty powersave`) demonstrate similar performance patterns, making them reliable but not remarkable in this benchmark.