Paul Davis wrote: >steinberg; CoreAudio comes *with* OS X, and so they have no >distribution issues to worry about - its part of the same overall >license as the OS. i presume that means that implementing a version of >the API for a different OS is completely legal. header files? not >sure.
that's the way i read the 'apsl' too. >i don't consider AudioUnits a good candidate for LADPA, but that >doesn't mean i don't like the AudioUnits API. instead, it was a prime >motivator behind JACK. i think that VST is actually much better >overall, but it has some rough edges that might, or might not, be >fixed in 3.0. VST is a lot, lot, lot more tested in the real world. VST is also very, very 'tainted' as linus would put it; personally i've decided to not even look at it. (ot: in a recent longish mail conversation i had with info@steinberg, they refused to give the spec of the '.arr'-file format that is more than ten years old. so much for steinborg's interpretation of 'live and let live' :). i believe that with patience and prudence 'lad' may be capable of producing an interface definition capable of connecting all our audio/MIDI apps and libraries in style, without deriving from CA or VST. i think that Jack and its quick adoption shows the potential is here, but i also think that o-o-p is a big performance burden in most cases (otoh i seem to have a faible for aged hardware ;). >the AudioUnit API still shows way too many signs of "Apple-speak" for >me. i can't really pin down what i mean by this, but whenever i browse >the API specs, i just have this pascal-y, motorola-ish feeling in my >gut :) true. i like to see this as idealism shining through. otoh i think that apple's policy of 'hiding gory detail from users/developers' sent OS < X down the road to hell, and (again ot:) i think that if they do not come to their senses in time, OSX will travel down that road, too. >plus there's that little problem with AudioUnits not being capable of >capture as of a few months back. somebody on the VST plugins list >claimed it was something to do with a kernel thread design issue - >quite deep, if true. still, the API seems generic enough to hide whatever quirks may be lurking beneath the shiny java spec. tim
