On Friday, September 30, 2011, Niek van den Berg wrote: > Automatic compression, sounds interesting. Can I find some more information > about this or can you give hint where I have to look?
I have about as much of an idea as you do at the moment. > For plain repeats this would work but how about endings? Endings have only the most rudimentary and horribly kludgey support. > I wander it is a good idea to have such difference between the playback and > notation behavior... It's a terrible idea, and a pile of crap to work with, but it's what we have to use for a tool to get this done. Before, it was just totally impossible. Now it's a complete abomination, but at least it does something. I was the one who came up with this monstrosity, and it was the best I could think to do under the circumstances. > However maybe the new linked segments may help. Imaging a composition with > segments > > A - B - A - C > > and the A segments are linked segments. This can be noted as > > |: A | B (ending 1) :| C (ending 2) | > > A forced use of the Alt-x Lilypond directives is not required any more and > the notation matches the replay. With the "compression" we're talking about, that could indeed go some good way toward solving the problem more elegantly. I'm really not quite sure what Yves has done with that, and my frame of reference for all this is still the old, traditional way. From that frame of reference, you had to have two completely separate versions of everything if you wanted to hear something and print it properly as notation. I think we're some way closer to being able to do everything with the same data now, but I just haven't had time to go digging around to see exactly what it does, or how it works under the hood. > Unfortunately this can not be extended to implement Dal Segno/Capo because > the symbols have a negative subordering, placing the symbol before a note. If you can see a way to use all of this to work some new magic by changing the subordering, the subordering of the Symbol class can be changed. This is one case where I wouldn't worry so much about breaking compatibility with old files, and likewise anything that makes use of the assorted LilyPond hacks that all exist to detour around limitations in Rosegarden. I don't think those assorted hacks have been widely used anyway, because they're so awful. I may well be the only one who ever bothered to try to do anything with them, and I'm happy to rework any affected files if somebody is finally smart enough to figure out a way for Rosegarden to have its cake and eat it too. > matches with what is really sounding. It seems what is noted is not > necessarely not what you hear when the play the composition. The question > what should take precedence? Traditional thinking would have it that sound matching notation is an impossible luxury, so a notation-oriented export function should produce the printed result the composer intended, even if it doesn't match playback. Maybe we can have some new thinking. I'm not closing the door on any of that by any means. I am, however, rather sick still today, and I'm not up for any serious thinking about anything. My mind is way too foggy right now to take any of this too seriously, and I could be spewing complete drivel right now without even realizing it. -- D. Michael McIntyre ------------------------------------------------------------------------------ All of the data generated in your IT infrastructure is seriously valuable. Why? It contains a definitive record of application performance, security threats, fraudulent activity, and more. Splunk takes this data and makes sense of it. IT sense. And common sense. http://p.sf.net/sfu/splunk-d2dcopy2 _______________________________________________ Rosegarden-devel mailing list [email protected] - use the link below to unsubscribe https://lists.sourceforge.net/lists/listinfo/rosegarden-devel
