Re: Tie bug in Lilypond 2.7.18 for Windows still not resolved

classic Classic list List threaded Threaded
15 messages Options
Reply | Threaded
Open this post in threaded view
|

Re: Tie bug in Lilypond 2.7.18 for Windows still not resolved

seng_liang
Hi, it seems that the tie problem is still not solved in version 2.7.18
(Windows 98 version).
Here is an extract of the file:
\version "2.7.18"
\relative c' {
<c' e,>1 < d g,> <c c,>~<c c,>
<g g,>4 <g g,>8. <g g,>16 <g g,>4 <g g,>
<g g,>1~<g g,>~<g g,>  r2 <g e>2 r2 <g e>
\bar ".|." }

If you generate this file using 2.7.18 for Windows ( 98 ), you will find
that the lower ties in bar 6-8 are not vertically aligned correctly. Why is
this well documented bug keep recurring?
I recall that the stable version of 2.6.x generated perfect ties. Why is
there the need to reformat the tie feature?




_______________________________________________
bug-lilypond mailing list
[hidden email]
http://lists.gnu.org/mailman/listinfo/bug-lilypond
Reply | Threaded
Open this post in threaded view
|

Re: Tie bug in Lilypond 2.7.18 for Windows still not resolved

Erik Sandberg
On Friday 16 December 2005 15.45, liang seng wrote:
> Hi, it seems that the tie problem is still not solved in version 2.7.18
> (Windows 98 version).

As Han-Wen said, the problem should be fixed in v2.7.19 and later.

BTW, if the bugs in the unstable 2.7.x series disturb you, it might be a good
idea to stick with 2.6, and wait until 2.8 comes out before switching.
(unless, of course, you explicitly need the new features of 2.7)

--
Erik


_______________________________________________
bug-lilypond mailing list
[hidden email]
http://lists.gnu.org/mailman/listinfo/bug-lilypond
Reply | Threaded
Open this post in threaded view
|

Re: Tie bug in Lilypond 2.7.18 for Windows still not resolved

seng_liang
Hi, I see. But I'm curious to know that how can you be certain that the tie
problem will be fixed once and for all in the upcoming stable series 2.8.x?
Reitering my previous question, I would like to know why the tie formatting
is being rewritten in 2.7.x which causes all these problems in the first
place when there is nothing wrong with the tie format 2.6.x which generates
good ties?




_______________________________________________
bug-lilypond mailing list
[hidden email]
http://lists.gnu.org/mailman/listinfo/bug-lilypond
Reply | Threaded
Open this post in threaded view
|

Re: Tie bug in Lilypond 2.7.18 for Windows still not resolved

Han-Wen Nienhuys
liang seng wrote:
> in the first place when there is nothing wrong with the tie format 2.6.x
> which generates good ties?

Many people, including me, disagree.  See eg.

   http://lilypond.org/bugs/v2.6/


--
  Han-Wen Nienhuys - [hidden email] - http://www.xs4all.nl/~hanwen


_______________________________________________
bug-lilypond mailing list
[hidden email]
http://lists.gnu.org/mailman/listinfo/bug-lilypond
Reply | Threaded
Open this post in threaded view
|

Re: Tie bug in Lilypond 2.7.18 for Windows still not resolved

seng_liang
Hmm, I see. It's just that the tie problem I highlighted in my example does
not appear in 2.6.x, so I assume that this is not such a big problem in the
stable series. By the way, Mr. Erik Sandberg mentioned that this should be
fixed once and for all in 2.7.19. Is this true?




_______________________________________________
bug-lilypond mailing list
[hidden email]
http://lists.gnu.org/mailman/listinfo/bug-lilypond
Reply | Threaded
Open this post in threaded view
|

Re: Tie bug in Lilypond 2.7.18 for Windows still not resolved

seng_liang
In reply to this post by seng_liang
Hi, after installing version 2.7.23 for Windows, I have to say that the tie
reformatting feature is still not satisfactory. I've attached 2 files for
your reference.
Testa.pdf is generated by version 2.6.3 for Windows (the source file content
is at the start of this thread) while Testb.pdf is generated by version
2.7.23
As you can see, there is a big difference in the ties in both files. I don't
know about the horizontal distance between the end of ties and the
noteheads, but don't you agree that the ties generated by 2.6.3 look better?
If you take a downward tie as an example, the 2.6.3 version looks better
because the tie always starts and terminates on or below the lowest point of
the notehead. While in the 2.7.23 version, they start and end in the
midpoint of a notehead. While this is correct for ties in notes within a
chord, this is clearly not correct for ties attached to the outermost notes
of a chord or when there is only a single note. (Which in the case of the
attached files, upward and downward ties show the same problem in Testb.pdf
)
Clearly, a mechanism has to be created in order to tell Lilypond how to
differentiate between whether to apply a tie to a note either as the note is
inside a chord or as the outermost notes of a chord. I'm not a computer
expert, but is there a way to tell Lilypond this? I hope I'm making myself
clear as this is beginning to annoy me.


_______________________________________________
bug-lilypond mailing list
[hidden email]
http://lists.gnu.org/mailman/listinfo/bug-lilypond

Testa.pdf (25K) Download Attachment
Testb.pdf (22K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Tie bug in Lilypond 2.7.18 for Windows still not resolved

Erik Sandberg
On Wednesday 21 December 2005 13.21, liang seng wrote:
> Hi, after installing version 2.7.23 for Windows, I have to say that the tie
> reformatting feature is still not satisfactory. I've attached 2 files for
> your reference.
> Testa.pdf is generated by version 2.6.3 for Windows (the source file
> content is at the start of this thread) while Testb.pdf is generated by
> version 2.7.23

In the future, please send compilable .ly snippets as well; that makes it
easier for us to reproduce your problem.

> As you can see, there is a big difference in the ties in both files. I
> don't know about the horizontal distance between the end of ties and the
> noteheads, but don't you agree that the ties generated by 2.6.3 look
> better? If you take a downward tie as an example, the 2.6.3 version looks
> better because the tie always starts and terminates on or below the lowest
> point of the notehead. While in the 2.7.23 version, they start and end in
> the midpoint of a notehead. While this is correct for ties in notes within
> a chord, this is clearly not correct for ties attached to the outermost
> notes of a chord or when there is only a single note.

Are you sure about this rule? Try e.g.:
\relative { << {c'2~c} \\ {c,~c}>> <c c,>~<c c,> }

The ties between the chords look good IMHO (do you disagree?)

Your problem might be that the chord has no stem; this makes it look like
polyphony. You can always use polyphonic notation (<<c\\c,>> iso. <c c,>) if
you want polyphony-style ties.

(disclaimer: I am no typesetting expert, so I don't know what ties are
supposed to look like)

--
Erik Sandberg
Maintainer of the Lilypond bug CVS archive,
http://savannah.gnu.org/cgi-bin/viewcvs/lilypond/lily-bugs/bugs/
http://lilypond.org/bugs/v2.7/


_______________________________________________
bug-lilypond mailing list
[hidden email]
http://lists.gnu.org/mailman/listinfo/bug-lilypond
Reply | Threaded
Open this post in threaded view
|

Re: Tie bug in Lilypond 2.7.18 for Windows still not resolved

Han-Wen Nienhuys-2
In reply to this post by seng_liang
liang seng wrote:

> note is inside a chord or as the outermost notes of a chord. I'm not a
> computer expert, but is there a way to tell Lilypond this? I hope I'm
> making myself clear as this is beginning to annoy me.

At this point Steve Doonan is sponsoring me in monthly installments, so
the next time I will have a look at his gripes in January.  If it really
annoys you, you can sponsor further work on this.

thanks,


--

Han-Wen Nienhuys - [hidden email] - http://www.xs4all.nl/~hanwen

LilyPond Software Design
  -- Code for Music Notation
http://www.lilypond-design.com



_______________________________________________
bug-lilypond mailing list
[hidden email]
http://lists.gnu.org/mailman/listinfo/bug-lilypond
Reply | Threaded
Open this post in threaded view
|

Re: Tie bug in Lilypond 2.7.18 for Windows still not resolved

seng_liang
In reply to this post by Erik Sandberg
Hi, here is the source file again:
\version "2.7.23"
\relative c' {
<c' e,>1 < d g,> <c c,>~<c c,>
<g g,>4 <g g,>8. <g g,>16 <g g,>4 <g g,>
<g g,>1~<g g,>~<g g,>  r2 <g e>2 r2 <g e>
\bar ".|." }

Yes, I do agree that it looks good. However, the problem is they look good
only in polyphony. In my case, I cannot change the polyphony or break notes
into two voices since I have to faithfully follow the Urtext. Your solution
also has limitations for string writing. As notes belonging to the same or
different stems mean that they are performed differently.
Thanks.




_______________________________________________
bug-lilypond mailing list
[hidden email]
http://lists.gnu.org/mailman/listinfo/bug-lilypond
Reply | Threaded
Open this post in threaded view
|

Re: Tie bug in Lilypond 2.7.18 for Windows still not resolved

seng_liang
In reply to this post by Han-Wen Nienhuys-2
Hi, ok. But is there a way to sponsor this without using a credit card?




_______________________________________________
bug-lilypond mailing list
[hidden email]
http://lists.gnu.org/mailman/listinfo/bug-lilypond
Reply | Threaded
Open this post in threaded view
|

Re: Tie bug in Lilypond 2.7.18 for Windows still not resolved

Erik Sandberg
In reply to this post by seng_liang
On Wednesday 21 December 2005 14.07, liang seng wrote:

> Hi, here is the source file again:
> \version "2.7.23"
> \relative c' {
> <c' e,>1 < d g,> <c c,>~<c c,>
> <g g,>4 <g g,>8. <g g,>16 <g g,>4 <g g,>
> <g g,>1~<g g,>~<g g,>  r2 <g e>2 r2 <g e>
> \bar ".|." }
>
> Yes, I do agree that it looks good. However, the problem is they look good
> only in polyphony. In my case, I cannot change the polyphony or break notes
> into two voices since I have to faithfully follow the Urtext. Your solution
> also has limitations for string writing. As notes belonging to the same or
> different stems mean that they are performed differently.
> Thanks.

I think formulated myself poorly. What I meant, was that in your case, you can
substitute
<c c,>1~<c c,>
for
<<{c1~c}\\ {c, c}>>

What I meant with the half notes, is that
<c c,>2~<c c,>
in fact _does_ look good (therefore only whole notes are problematic, and the
above fix is lossless since it doesn't change any visible stem directions)

However, there are still problems with stuff like
<c c,>1 ~ <c c,> ~ <c c,>2
but again, I don't know what's considered beautiful in those cases.

BTW, you can probably use \hideNotes tricks to get polyphonic ties for chords
(create invisible, tied, polyphonic voices, and non-tied chords)

--
Erik


_______________________________________________
bug-lilypond mailing list
[hidden email]
http://lists.gnu.org/mailman/listinfo/bug-lilypond
Reply | Threaded
Open this post in threaded view
|

Re: Tie bug in Lilypond 2.7.18 for Windows still not resolved

seng_liang
In reply to this post by seng_liang
Thanks. I will look into this. By the way, you mentioned that there is an
ongoing improvement on the tie reformatting. Does this mean that the tie
feature will get better and the bugs fixed in later versions and will be on
par with the stable 2.6 version, if not better?




_______________________________________________
bug-lilypond mailing list
[hidden email]
http://lists.gnu.org/mailman/listinfo/bug-lilypond
Reply | Threaded
Open this post in threaded view
|

Re: Tie bug in Lilypond 2.7.18 for Windows still not resolved

seng_liang
In reply to this post by Erik Sandberg
Yes, I agree with you. Regarding to your question about what consitutes a
good tie, I think if we take a look at some editions of scores, the pdf
generated by 2.6.3 certainly looks a lot like the editions than the pdf
generated by 2.7.23. Again, I understand that the tie reformatting has to be
done because of some underlying problems with the stable version. I'm just
saying that certain tie formatting rules applied in 2.6 (vertical distance
between tie and notehead) should be mantained.




_______________________________________________
bug-lilypond mailing list
[hidden email]
http://lists.gnu.org/mailman/listinfo/bug-lilypond
Reply | Threaded
Open this post in threaded view
|

Re: Tie bug in Lilypond 2.7.18 for Windows still not resolved

Han-Wen Nienhuys-2
In reply to this post by seng_liang
liang seng wrote:
> Thanks. I will look into this. By the way, you mentioned that there is
> an ongoing improvement on the tie reformatting. Does this mean that the
> tie feature will get better and the bugs fixed in later versions and
> will be on par with the stable 2.6 version, if not better?

the idea is that everything will be better.


--

Han-Wen Nienhuys - [hidden email] - http://www.xs4all.nl/~hanwen

LilyPond Software Design
  -- Code for Music Notation
http://www.lilypond-design.com



_______________________________________________
bug-lilypond mailing list
[hidden email]
http://lists.gnu.org/mailman/listinfo/bug-lilypond
Reply | Threaded
Open this post in threaded view
|

Re: Tie bug in Lilypond 2.7.18 for Windows still not resolved

seng_liang
Hi, thanks again. I guess that is the ultimate aim in this development
series. Looks like we'll have to wait for the tie problem to improve. By the
way, how much does it cost to sponsor this?




_______________________________________________
bug-lilypond mailing list
[hidden email]
http://lists.gnu.org/mailman/listinfo/bug-lilypond