Benchmark #57

Download
osu!stable + ffmpeg - lavd pre-vdeadline vs lavd-metrics

Submitted 4 months ago by .dnaim

Specifications
Label OS GPU CPU RAM OS specific
osu!stable-lavd-metrics Fedora Linux 40 (KDE Plasma) NVIDIA GeForce RTX 3050 Laptop GPU/PCIe/SSE2 AMD Ryzen 5 6600HS Creator Edition 14 GB 6.9.9-cn1.0.fc40.x86_64 powersave
osu!stable-lavd-pre-vdeadline Fedora Linux 40 (KDE Plasma) NVIDIA GeForce RTX 3050 Laptop GPU/PCIe/SSE2 AMD Ryzen 5 6600HS Creator Edition 14 GB 6.9.9-cn1.0.fc40.x86_64 powersave
# Top runs: * **Highest FPS**: `osu!stable-lavd-pre-vdeadline` with approximately 20% higher average FPS compared to `osu!stable-lavd-metrics`. * **Smoothest FPS**: `osu!stable-lavd-pre-vdeadline` with about 30% lower standard deviation and variance in FPS than `osu!stable-lavd-metrics`. * **Best overall**: `osu!stable-lavd-pre-vdeadline` due to significantly higher average FPS and lower FPS variance. # Issues: * `osu!stable-lavd-metrics` shows low minimum FPS values, approximately less than 50% of `osu!stable-lavd-pre-vdeadline`, indicating significant frame drops. * FPS variance in `osu!stable-lavd-metrics` is nearly double that of `osu!stable-lavd-pre-vdeadline`, suggesting instability in frame delivery. # Summary The benchmark compares two configurations of the software associated with FFmpeg: `lavd pre-vdeadline` and `lavd-metrics`. Both tests were conducted using the same hardware and Linux OS environment. The `lavd pre-vdeadline` configuration significantly outperforms `lavd-metrics` in average FPS by roughly 20%, and also demonstrates more stable frame delivery with around 30% lower FPS variance. These results suggest that the `lavd pre-vdeadline` adjustment could lead to better overall performance and a smoother user experience.