JMC wrote:
There are ugens floating around written by people who have moved on are probably never going to recompile them. These will be broken forever. OK, Well they didn't release the source code, so maybe it is deserved.
Actually existing binaries shouldn't be broken if all we are doing is renaming things. The binaries will still work unless the struct layouts change. Correct me if I'm wrong, but I think the only breakage will be with the source code.
A GraphDef is not a SynthDef. They are different types. One is the result of a transformation of the other. In pseudo Haskell notation you'd have this: SCLang :: SynthDef -> array of bytes SCSynth :: array of bytes -> GraphDef I think it would be slightly more confusing to have the SCSynth object and the SCLang object have the same name since it would then require an additional word to disambiguate which you are talking about. But I don't really care that much.
I see.It's difficult for me to make an assessment about which is less confusing in general. Does anyone else have an opinion about making things more/less understandable?
In my view:- I understand the distinction James makes above but for me it would have been a lot easier to understand the server source if Graph and GraphDef had been named such that I could related them to the domain concepts of Synth and SynthDef.
- "Graph" is a pretty generic term. Given that we already have another graph (of Nodes) it get's pretty confusing.
- The compiled scsynth GraphDef is more of a StaticExecutionScheduleAndBufferAllocation than a Graph.
Comments anyone? (please!) Best wishes Ross.