I love your enthusiasm and continual support of our projects audiojunkie!
But I also agree with Andreya! Our goal is definitely to have a fantastic instrument which supports loads of unique features. But the 'format' we are using to stream the internal state is, by design, closely tied to that internal state. You will certainly have files you can share which SCXT will load, but they would be pretty inscrutable to other synths and, most importantly, we don't want to have the burden of 'streaming format' maintenance beyond doing it at load time.
We instead have import from sfz, sf2, and the open bitwig/presonus multisample (which works today) and will almost definitely have a (fidelity reducing) export to those formats also (and maybe to decent also).
And I say fidelity reducing because, as Andrea points out, there are going to be some things SCXT does which isn't really a standard, it's a feature. I'm (right now - or would be right now if I wasn't typing on KVR instead![:)]( "Smile")
) working on the inline zone and group processors and, for instance, the weird custom pitch shifting / phasing / modulating thingy we have is certainly not something you'd expect in sfizz or decent or so on.
But anyway I do agree it is a cool and exciting project, and we have some other adjacent ideas for sample-initiated synthesis which will complement it will this year. But we aren't designing an interchange format. Just another fun and cool open source instrument, which in this case starts with a sample.
But I also agree with Andreya! Our goal is definitely to have a fantastic instrument which supports loads of unique features. But the 'format' we are using to stream the internal state is, by design, closely tied to that internal state. You will certainly have files you can share which SCXT will load, but they would be pretty inscrutable to other synths and, most importantly, we don't want to have the burden of 'streaming format' maintenance beyond doing it at load time.
We instead have import from sfz, sf2, and the open bitwig/presonus multisample (which works today) and will almost definitely have a (fidelity reducing) export to those formats also (and maybe to decent also).
And I say fidelity reducing because, as Andrea points out, there are going to be some things SCXT does which isn't really a standard, it's a feature. I'm (right now - or would be right now if I wasn't typing on KVR instead
) working on the inline zone and group processors and, for instance, the weird custom pitch shifting / phasing / modulating thingy we have is certainly not something you'd expect in sfizz or decent or so on.But anyway I do agree it is a cool and exciting project, and we have some other adjacent ideas for sample-initiated synthesis which will complement it will this year. But we aren't designing an interchange format. Just another fun and cool open source instrument, which in this case starts with a sample.
Statistics: Posted by baconpaul — Sat Apr 06, 2024 10:38 pm