> having hosts have to be prepared to insert encoding converters, > channel converters, channel multiplexers, channel splitters, > float/double converters, PCM/PWM converters: all i can say as a host > author is ... what a nightmare! i'd rather stick with LADSPA, as > not-even-really-adequate as it is. This is making it sound scary, but it's a bit overgrown. Channel-Bundle converters will be needed, if we semantically join mono streams together. They are almost free if implemented at the host level (SDK code), and can be a GMPI plugin, with cost depending on buffer management. Encoding converters - bad idea. Assuming we support encodings, it is either a hint, or a hard rule. There is no generic way to transform any encoding to any other. Channel mux and demux converters are needed if the host allows things like T-splitting a signal. No way around that. Datatype conversion. Uggh. Yep. PCM-PWM conversion - I don't recall ANYONE making any suggestion that non-PCM streams are on the horizon. I'm not saying that three (well, two if you make channel-bundle and channel-mux be the same) adaptors are a good way to start GMPI. Just that you're overstating it a bit. > GMPI doesn't have to be all things to all people in all situations. no > existing plugin API. GMPI has to be demonstrably better than existing > plugin APIs, which it could do by simply being cross-platform and > non-proprietary. throwing in some fixes for some problems with > existing plugin APIs can only help that. This is an excellent point that we have sometimes lost sight of. To which I'd add "Good will always lose to Good Enough". If Float32 is really Good Enough, maybe it is good enough. ---------------------------------------------------------------------- Generalized Music Plugin Interface (GMPI) public discussion list Participation in this list is contingent upon your abiding by the following rules: Please stay on topic. You are responsible for your own words. Please respect your fellow subscribers. Please do not redistribute anyone else's words without their permission. Archive: //www.freelists.org/archives/gmpi Email gmpi-request@xxxxxxxxxxxxx w/ subject "unsubscribe" to unsubscribe