Hello, Well, that's basically what Max does. It defaults to the original 0-127 and then you can specify a "hires" mode to use all 14-bits rather than just 8-bits like 0-127 did. We just implemented this for midiparse and midiformat in cyclone, the current version only had the 0-127. So yes, it'd be pretty easy to do. Just default to the 0-127 and have a flag or have a message change the mode to the "hires" mode. I think in terms of a software's life cycle, if you want it to keep getting better and better, something along the line has to break at some point when people figure out better and more efficient ways of implementing things. Look at Python 2 vs 3 or Lua. But I suppose that argument is for a different time and a different thread and you wouldn't necessarily have to break things here anyways =).
Derek On Sep 04, William Huston wrote: > I agree with Alexandre about what seems sensible, but introducing a change > like this will break existing patches. > > Is it possible (for all changes like this) to introduce a compatibility > mode to get the old behavior? > > Yes, any seasoned Pd hacker will know how to fix the scaling. I'm just > imagining someone's patch breaks after upgrading and they just want a fast > way to get old behavior. > > Thanks, > > On Sunday, September 4, 2016, Alexandre Torres Porres <por...@gmail.com> > wrote: > > > > > > 2016-09-04 13:35 GMT-03:00 Giulio Moro <giuliom...@yahoo.it>: > >> > >> Is this a [bendin] or a [bendout] bug? > >> [bendin]'s current implementation is closer to the actual MIDI messages > being transmitted. > > > > but in actuality, the not raw standard is from -8192 to 8191 right? > > I just care they both are the same, but it seems to me that -8192 to 8191 > is the sensible choice > > cheers > > -- > -- > May you, and all beings > be happy and free from suffering :) > -- ancient Buddhist Prayer (Metta) _______________________________________________ Pd-list@lists.iem.at mailing list UNSUBSCRIBE and account-management -> https://lists.puredata.info/listinfo/pd-list
