There was a talk on it this year, which you can watch here: https://fosdem.org/2020/schedule/event/videobox/ https://fosdem.org/2020/schedule/event/videobox/ T

There was a talk on it this year, which you can watch here: https://fosdem.org/2020/schedule/event/videobox/ https://fosdem.org/2020/schedule/event/videobox/ TLDR: Current setup is some Frankenstein monster of various black boxes (scaler, encoder, splitter etc) that uses a Banana Pi with annoying restrictions. They are developing a bespoke solution that should be far smaller, cheaper and without the annoying restrictions. I attended and spoke at FOSDEM for the first time this year (you can see my talk here https://fosdem.org/2020/schedule/event/riscv_lowrisc/ https://fosdem.org/2020/schedule/event/riscv_lowrisc/) and was very impressed at the video setup. I got a link the same day to edit the video. You get a simple interface (watch video, adjust end points, choose audio channel) and after confirming everything looks good it gets transcoded and uploaded in short order (I reviewed my talk around 8pm the day I gave it, think it was available from 8:30).