Skip to content

README: timings measured on an idle machine - #56

Merged
joeljose merged 1 commit into
mainfrom
docs/idle-benchmarks
Sep 29, 2026
Merged

joeljose merged 1 commit into
mainfrom
docs/idle-benchmarks

Conversation

@joeljose

Copy link
Copy Markdown
Owner

The earlier timing figures were taken without checking system load (another session was running containers part of the time), and before #40/#43. These numbers are re-measured on an idle machine (load < 0.5, no containers), as the median of 5 runs after a warm-up, and split by output format, since lossless FFV1 costs about 10 ms a frame.

face.mp4 .avi .mkv
CPU 4.5 s 7.4 s
GPU 1.5 s 4.6 s

Docs only.

Medians of 5 runs after a warm-up, with no other load: CPU 4.5 s (.avi)
/ 7.4 s (.mkv), GPU 1.5 s / 4.6 s on face.mp4. Lossless FFV1 output
costs about 10 ms a frame.
@joeljose
joeljose merged commit a9ce667 into main Sep 29, 2026
1 check passed
@joeljose
joeljose deleted the docs/idle-benchmarks branch September 29, 2026 07:46
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant