fix demo video / screencast

[ThomasWaldmann]

https://chaos.social/@sindastra/107792763024290251 see thread

if we want to show compression:

• use zstd compression (do not use lz4, which is default anyway)
• do not use incompressible data, like jpeg images

[ReethuVinta]

• What are the expectations for the video? Do I need to make the replicate entire 6 min video( changing zstd compression) or just demonstrate zstd instead of lz4(only 1:18 part )?
• Should the terminal also resemble the one as seen in the video?

[ThomasWaldmann]

There is currently some work on automating asciinema videos, see PRs.

[ThomasWaldmann]

The automation has been merged into master (and 1.2-maint), see PR #6197.

The docs of master branch currently show the borg 1.1 demo video, this must be replaced by an automated build of a borg 2.0 demo video before going rc.

[rugk]

BTW a cool project I stumbled upon is https://github.com/k9withabone/autocast, which allows you to automate asciinema's. I created a new issue #8040 to track the idea of maybe using it.

Edit: Oh sorry I see it was automated in a different way hmm…