I am happy to announce that after a quiet patch of CasparCG, there is now a v2.4.0 RC1 release.
It’s been a while since v2.3 was released, and various improvements have been worked on in the time since.
The highlights of this release is an updated CEF, better workflows for videowalls and a bunch of smaller improvements as well as bug fixes.
To go with this, is an updated CasparCG Client, and some work has also been done on the media-scanner last year.
If you encounter any issues in these, let us know on GitHub.
There is a known bug in the version of media-scanner included in the downloads, it is recommended to download and use the latest version of media-scanner with the v2.4.0 RC1 release of server
You can find the full changelogs and downloads for each in the GitHub releases.
It is now using CEF 117. 2.3.3 was on CEF 71, and the builds for a while were using CEF 95.
I don’t know what major changes CEF brings, other than CEF 71 is from 2018, so is missing or lacking in a lot of ways compared to modern browsers. Support for newer javascript standards, better WebGL and other technologies too. Somewhere online will probably be able to give a proper rundown of the changes (you might need to search for changes in Chromium, CEF is a wrapper around that)
I have not yet had time to investigate the detail, but some early tests suggest the audio is behaving differently in server 2.4 to previous versions.
I have a simple file that has 4 stereo track sets that report as 4 stream IDs in ffprobe. Think 4 languages. When played on server 2.1.12_NRK in 8ch mode and server 2.3.3 these reproduce the layout used when the video edit was made:
Audio Stream 1 - channels 1 and 2
Audio Stream 2 - channels 3 and 4
Audio Stream 3 - channels 5 and 6
Audio Stream 4 - channels 7 and 8.
Played in server 2.4.0_RC1 the audio outputs are:
Audio Stream 1 - channels 1 and 2
Audio Stream 2 - channel 3 only
Audio Stream 3 - channels 4 and 5
Audio Stream 4 - channels 6 and 7
I repeat that these are early tests, so I have not had time to create test content material that can examine if stream 2 channel 2 is being output elsewhere - possibly with LFE low pass filtering.
Andy
Further investigation showed correct CasparCG opertion. See later posting.
Yestarday I tested and can confirm that this server realase has full 16ch passthru output 1-16 in order. I use synchronizer before and everything work perfectly.
I’m not tested mapping, but as far as I know it is not possible,maybe I’m wrong.
I have now created some new audio channel identification tests, and I am pleased to report that Caspar server V2.4 is correctly assigning 16 channels of audio in all test I have run.
What I discovered is the video editor I used to create the packaged video and audio was not fully honouring the instructions to output all audio tracks as discrete encodes. This placed metadata in the file that CasparCG read and configured the audio outputs as a 7.1 encode.
Im trying to setup the new version and whichever client I use I cant see my resources on the media folder. My steps of the whole setup are the same I did to the previous version of the server. Media-scanner is up and running. I cant think anything else I could try. win10. Any idea?
Forum post Scanner won't add media notes thare are known problems with the scanner release distributed with server RC1. If you have not yet updated the scanner to the latest release available from the github distribution, try the update to see if that helps.
After installing the new version of the scanner delete the _media folder in the server folder before restarting the new version of the scanner. Observe the scanner window as it starts and check it shows a list of media files scanned.
You can also check if the scanner is seeing the media folder by using the http endpoint access. On the server machine use a browser with address line:
http://127.0.0.1:8000/CLS
which returns the list of audio, stills and movies stored in the scanner database.
(hopefully something about animation and requestAnimationFrame being much better??)
