AvxSynth is a Linux port of the AviSynth toolkit. The objective of the porting effort was to bring the power of AviSynth into the Linux world. In particular, we are interested in AviSynth as a frame server front-end to the encode step of our media pipeline. An overview of the AvxSynth architecture can be found here.

The AvxSynth port was based on the AviSynth 2.5.8 code base, and we are calling this version 4.0, so as to differentiate from the dead AviSynth 3.0 project.

To build and run AvxSynth, follow these instructions.

You can read about the porting effort here.

You can see the status of the built-in function porting here.

A summary of the testing status can be found here and issues are tracked here.

The AvxSynth toolkit includes a frame server application that delivers decoded frames through stdout. This can be piped to x264. You can see how to use AvxSynth with x264 here.

Click here to see a short video of AvxSynth in action. This video demonstrates the ffmpeg source, ffmpeg scaling, ShowSMPTE, and AvxSubtitle, a new filter that does caption burn-in.

The AvxSynth error handling model is discussed here.

Doom9 discussion is here.

If you would like to contribute to AvxSynth, we have need for developers, testers, and build-install experts. More information can be found here.