Simplified build of 220.video-processing

[mcopik]

This benchmark relies on a static build of ffmpeg, which we source from an external website. Instead of pulling the ffmpeg binary at every single build, which adds a lot of pressure on that website in CI builds and can lead to errors when pulling data (e.g., rate limits), we could put a pinned and reproducible variant of ffmpeg in our benchmarks-data instead.

We should a new resources category in benchmarks-data, and change init.sh script to copy that binary during the initial build.

EDIT: we should also revise the command in the function to make sure that we explicitly select codecs and are reproducible.

[lolamaspoli]

Hi! I'd like to work on this. Plan: pin a static ffmpeg build in benchmarks-data under a new resources category, and update init.sh to copy it locally instead of pulling from the external site during build. Let me know if I can take this on.

[mcopik]

@lolamaspoli Thanks for the interest! I checked the issue closely, and I think we would have to do it slightly differently: ffmpeg builds are GPL-licensed due to the x264 codec that we use, so we cannot distribute them directly. We should instead add a "fetch" functionality that runs on the first execution of SeBS, and fetches the extra data for benchmarks - same path, just not in the directory. Then add a hidden mark that data is fetched, so this is not retried in the future. This would also be helpful because we will have to fetch extra data in the future (files not fitting into repo limits).

[lolamaspoli]

@lolamaspoli Thanks for the interest! I checked the issue closely, and I think we would have to do it slightly differently: ffmpeg builds are GPL-licensed due to the x264 codec that we use, so we cannot distribute them directly. We should instead add a "fetch" functionality that runs on the first execution of SeBS, and fetches the extra data for benchmarks - same path, just not in the directory. Then add a hidden mark that data is fetched, so this is not retried in the future. This would also be helpful because we will have to fetch extra data in the future (files not fitting into repo limits).

Got it, thanks for clarifying the licensing issue! I'll implement it as a fetch step that runs on SeBS's first execution — pulling ffmpeg the first time and dropping a hidden marker file so it's not re-fetched. I'll get started on this.