And in that respect, a greedy algorithm is actually best, because it involves less work to override the decision of a "greedy" algorithm than a "fair" one (just add appropriate page breaks). So there is really no substitute for an expert human making this determination. It's actually one of the more subtle arts of good music engraving. It usually is more important to think about things like, how easy physically is it for a musician to turn the page at that point (after a long rest is perfect), how difficult would it be to remember the last few measures if physically it makes sense to turn earlier, etc. The problem is, this isn't actually a good way of going about deciding where page breaks should go in multi-page scores. A less "greedy" algorithm certainly has merit, and actually wouldn't be hard to program in the face of changing system sizes - you just need to think about evening out the system distance rather than evening out the number of systems.
#Two line arranger lilypond full#
However, it is a "greedy" algorithm - one that basically makes each page as full as possible, not caring if that leaves the last page less full than it need be. Page by page, the current system actually does quite well in providing two simple controls that offer a great deal of flexibility across a wide variety of score types.
(I mean that the "minimum system spacing means the minimum value for system spacing" type of answer doesn't work.) (This would not work well for scores which keep jumping around with different numbers of parts per system, but that's good because the latter is generally antisocial.)Ĭan you tell me where I can find a definition of the effect of seting "minimum" and "maximum" values for system spacing. Again, an independent function (which could probably just be a plugin I don't know) called "Paginate" could do this by calculating the default total length of the score, rounding up to a whole number of pages, then dividing the score evenly into the resulting number of systems. Yes, of course you are right - the last time I played in an orchestra was the 1960s, and I was thinking of piano music particularly, but also things like vocal scores.Īrguably, a "justify / fill" option would be a useful idea by "fill", I mean that by default the score would fill a whole number of pages. This means that the first page only has two systems, and there doesn't seem to be a way to down-align the second system.
#Two line arranger lilypond pdf#
Publication as a PDF file means it is a good idea to have the title and arranger etc info on the first page, so it can't get chopped off. The immediate prompt for this is an arrangement I made (Nutcracker final waltz and apotheosis for 2P4H). Personally I would write using exclusively Continuous view, which is the logical form of the score, then paginate later, and ideally I would like to choose the number of pages, then the system default would paginate to fill these. I suppose that the beginnings of computer set music had lots of limitations, and we started getting scores consisting of two pages, one system, and a two-bar fragment, which I (personally) find ugly. I prefer this because it looks like "real music". The only part of this that seems to be supported is setting the "Last system fill threshold" to zero, so there are no part lines. I guess that the traditional method was for the engraver to look at the manuscript, and if all bars are more or less that same size, decide on a number of pages, divide into the total number of bars, and then divide the bars on each page into the number of systems, with adjustments for repeats (as far as possible arranged to match page breaks). I can't find an option for this in Style, but would much prefer to have it. The traditional typesetting of music means that any score you buy has n pages, and every page has (roughly) the same number of systems, and every page including the last is basically filled with score. In other words there are no part-filled pages. What I mean by this is that the systems are evenly spaced so that the bottom ones line up on all pages.
0 kommentar(er)
