This update is as always long overdue, with mainly some subtle fixes, but it has been available on vdo.ninja/alpha/ in a stable form for a while now.
- MediaMTX SFU support is fairly polished now, so if you want to deploy your own Meshcast-like service, it's easy to do now.
- issues with no OPUS audio, flickering frame rates now and then, and other whep/whip related things addressed.
- Support for Meshcast v2 servers has been added, which is the primary reason for the urgency of update. (Meshcast v1 servers kept freezing up)
- Half the meshcast servers are already v2, and I'll be retiring v1 servers over the next little while.
- meshcast v2 servers have more reliable WHIP/WHEP support, and in general will allow me to offer more media input options into VDO.Ninja
- Quite a few WHIP / WHEP improvements were made in general, including automatic reconnect if a stream fails
- WHEP streams that haven't been publish yet will continuously try to be loaded, every few seconds, rather than just give up and fail
- Chunked video streaming mode supports dynamic buffers, so you can delay VDO.Ninja video output by minutes, with sync accurate down to a few frames.
- this is being used in combo with Ninbot + Sonobus for joint musician play-togethers, which an evolving project that is pushing chunked-mode development forward.
- The director's room has had some audio-channel features added, like a PGM mode.
- Another channel option is to control up to 7.1 channels of recording in OBS, allowing you to record 6 different guests in a group room to 6 different audio channels, which should make post processing audio for interviews or podcasts quite a bit easier
- The VDO.Ninja chat pop out has been redone, as recently mentioned
- Improved error messaging and fixed translations here and there, helping users with common issues to get better support messages
- Picture in Picture full window was fixed, to address a recent chrome change
- &voiceislation supported; this is an advanced noise filtering option in Chromium-based browsers
- Timecode support, enhanced MIDI, and standalone pages for minor tool testing, like TTS, added.
This list of changes is not exhaustive, nor is all the documentation on our Gitbook updated yet. Currently for more details of every new feature added though, see the Discord's updates channel at https://discord.vdo.ninja. It contains the feature descriptions for everything that has been added, but there is a lot to dig thru there.