Hi all, I want to do some screen recording on my linux desktop. And like a normal-functioning member of society, I decided to do it the hardest way and learn ffmpeg CLI to do it. Why? well, something about using underlying tools and customizing their usage excites me.
I have already started doing this, and I am finding I have to do a lot of trial and error to get things right. Before I dive deeper, I want to ask: Am I limiting myself in doing this? Is there anything I could be missing out on taking this route, or something that ffmpeg could not do on its own that a dedicated solution can?
What will I use this for exactly? well, things like recording a video game as I play it (which I suppose will require hardware acceleration to be of viable quality), or recording a tutorial (requiring voice input from mic), things like that.
I guess no Wayland support, which is pretty big
Actually there’s some support for Wayland. Tried it last year, but had to do serious rummage trying to make it work. For starters I can recall it spitted a video but the command throwed an error at the end and could not understand why. Also i seem to recall the video stopped way after finishing the command.
Even had to recompile all of ffmeg to add support for wayland recording (though Gentoo makes this really easy). One thing for certain is that Gentoo’s ffmpeg stable version is fairly behind from upstream’s so that could have had a hand on it too.
what do screen recording softwares for wayland use under the hood then?
They use XDG Desktop Portal, which makes video and audio capture system level instead of application level to enhance security. When a program wants to record your screen you get a pop up to choose which screen/window to share. After that the app can only see that window or screen.