I realize some differences are inevitable, and different styles settings can cause differences even given the same layout algorithm. But I am seeing differences that go well beyond what I'd expect. The easiest example I can point to is to create a new score for a single instrument, 32 measures, using all default settings. Do this in 1.3 and then in 2.0. Remove the instrument names from the 1.3 version to match 2.0. I get 22 empty measures on the first system with page size set to Letter in 1.3. I get only 18 in 2.0 (also with page size set to Letter).
I've gone through to compare style settings to see if there is something different that I can adjust to make these match more closely, but I'm not finding anything that makes a difference. See http://musescore.org/en/node/24271 for a forum post I made recently on the subject. Of course, it's not just empty scores, although that's where I see the biggest difference (and maybe that's OK, because who cares if empty scores look the same, and I think 1.3 made empty measures too small anyhow). But many "real" scores take more space in 2.0 than 1.3. See, for example, my "Reunion" demo. I know there were tweaks made to address this some time ago, and they helped. But differences were being masked by the fact that 1.3 scores, upon being loaded into 2.0, had a much too small distance from barline to first note in measure. The default in 1.3 and for new scores in 2.0 is 1.2sp, but 1.3 scores loaded into 2.0 were getting only half that, thus masking extra space in the layout elsewhere and allow many scores to have the same number of measures per line as in 1.3. But at the expense of a crowded barlne. When I changed this so that 1.3 scores would get the same left barline distance in 2.0 that they had in 1.3 or that new scores would get in 2.0 (1.2sp), it became apparent there are still significant differences. I'm not saying the 2.0 layout is worse - it's probably better. But if people put work into their 1.3 layout, they expect to see it preserved as well as possible. Do we know the source of the differences? If it's something specific we can fix with some custom handling in read114.cpp, great. If it's a lot of small things adding up, I might propose we try something like deliberately reducing the overall style / general / measure / spacing parameter by, say, 10% upon import of 1.3 scores. That's the one quick-and-dirty thing I found we could do that would allow 1.3 scores to load more like how they looked in 1.3, without introducing obvious bad side effects like reducing the left barline distance did. -- View this message in context: http://dev-list.musescore.org/Layout-differences-versus-1-3-tp7578579.html Sent from the MuseScore Developer mailing list archive at Nabble.com. ------------------------------------------------------------------------------ Managing the Performance of Cloud-Based Applications Take advantage of what the Cloud has to offer - Avoid Common Pitfalls. Read the Whitepaper. http://pubads.g.doubleclick.net/gampad/clk?id=121051231&iu=/4140/ostg.clktrk _______________________________________________ Mscore-developer mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/mscore-developer
