On Saturday 11 July 2009, Ilan wrote:

> It seemed strange to me that almost every note is tenuto. It seems like the
> sensitivity for setting it is too high.

Rosegarden tends to write tenutos where it should have written slurs.  It 
almost never writes slurs.  That algorithm could definitely do with some 
tweaking, but there are just too many higher priorities right now.

This would be a good project for these new people who have expressed an 
interest in doing something to take on.  There's a lot of room to improve it, 
and its performance is so abysmal already that there isn't much chance of 
making it worse.

> I saw there is Add Tenuto but no 
> remove. Then I found there was Remove all marks which did the trick.

Which reminds me that I really should finally get around to making a toolbar 
action for that.

> However there is still a small problem that Ctrl-A will select only the
> first track. I had expected that if I selected notes in the first bar of
> the second track and then did Ctrl-A it would select the whole of the
> second track. I couldn't find a method to select the whole of the N-th
> track where N>1.

It's really the segment boundary rather than the track boundary that matters 
here. The notation editor can lay out multiple segments in the same edit view 
(each track gets a separate staff), but you can only actively edit one of 
them at a time.  You can't make selections that span multiple segments, 
whether they also span multiple tracks/staffs or not.

You can do this from the main window segment canvas, though if you wanted 
three notes from this segment and five notes from the other one, this 
wouldn't be of much help either.

> Oops, it included the first note of bar 2... I tried this on other bars and
> generally it makes the error, but not always (depending upon which bar is
> chosen).

What's happened here is this is a notation-quantized performance where the 
performance and display durations are different.  If you look at the matrix, 
you can see that the last note in bar 1 of the first segment extends across 
the barline slightly.  A more literal rendering of this in notation would 
have a short extra note tied across the barline, and it would mangle the 
second bar.  Rosegarden has done right to simplify the performance into neat 
display notation, but that extra length is throwing off the slur.

I'm not sure why that is off hand.  I started looking at the raw events, and I 
realized I don't actually know how slurs work.  I've never had to fool with 
indications before.  It looks like it has a start time and a duration, and 
the figure that's off here is the duration.  It could be as simple as this 
code used performance duration instead of display duration to make its 
calculation.  We've killed dozens of bugs like that, and it's entirely 
possible this is another one we've missed.

I hope it's that simple.

Nice catch, Ilan!
-- 
D. Michael McIntyre 

------------------------------------------------------------------------------
Enter the BlackBerry Developer Challenge  
This is your chance to win up to $100,000 in prizes! For a limited time, 
vendors submitting new applications to BlackBerry App World(TM) will have
the opportunity to enter the BlackBerry Developer Challenge. See full prize  
details at: http://p.sf.net/sfu/Challenge
_______________________________________________
Rosegarden-devel mailing list
[email protected] - use the link below to unsubscribe
https://lists.sourceforge.net/lists/listinfo/rosegarden-devel

Reply via email to