Re: [linux-audio-dev] newest audio server for Linux (yep, yet another)

On Mon, 3 Feb 2003, Paul Davis wrote: I'm the lead developer on the MAS project. Thanks for checking it out! Here's my position: we didn't limit the architecture in any way that'd restrict it from being used pro-audio applications; however, the necessary stuff isn't exposed in the API now. It's all in the core, and if it isn't, it will be. I'm pretty confident we didn't screw this part up. Also, I hate the stream abstraction for audio. We're going to provide a callback-style "pull" API to compliment the standard push API, and add support for shared memory, and support for multiple threads in the server. We planned for all this. It just, you know... takes a while. That said, I think JACK is very cool. You guys had a different need -- to make multi-client pro-audio possible on Linux, and I think you've done it. We're coming at the audio (and time) problem from a different direction. Our goal is to serve the users who want audio to follow their remote X session and who want to make more than one sound at the same time, and to do it in a way that's higher quality, lower latency, and easier than what was possible before. And somehow, to do it such that the basic architecture, without all the "other stuff(TM)", can support high-end and pro applications. So, yeah, we have all kinds of plug-ins that do actual work and mess with the sound, like sample rate conversion and CODECs. But, my feeling is, if you don't need 'em, don't use 'em. Ultimately, if you know you've got N clients all at 96/24, you can ditch the sample rate converter. Anyway... we're working on documentation and more releases, and the actual "soundserver" API. And I'm looking forward to seeing how MAS and JACK can interoperate. Mike Andrews Lead Developer -- MAS Project