Re: Change stem direction based on position of note in staff?

[Lukas-Fabian Moser]

Hi Stefano,

> Yes, but it gets more complicated.  For example, this is something I
> transcribed in drumburp:
>
> Cr|X-------|--------|X-------|--------|
> Hh|--x-x-x-|x-x-x-x-|--x-x-x-|x-x-x-x-|
> HT|--------|--------|--------|----oooo|
> MT|--------|--------|--------|-----o--|
> Sn|--o---o-|--o---o-|--o---o-|--o---o-|
> FT|--------|--------|--------|--------|
> Bd|o---oo--|o---oo-o|o---oo--|o---oo-o|
>    1+2+3+4+ 1+2+3+4+ 1+2+3+4+ 1+2+3+4+
>
> HT, MT, FT are the toms (high, mid, and floor).  Cr is a crash cymbal,
> Hh is hihat, Sn is Snare and Bd is kick drum.  Not shown, but also
> common are pedal hihat and other cymbals, cowbells, woodblocks,
> triangles, etc.  Each instrument may get it's own line, but sometimes
> different note heads can indicate different instruments on the same
> line.
>     
    Yes, thanks for this more feature-rich example.

> I don't want to enter notes that way though.  Typing tabs is not fun.
>  Aligning all the ascii sucks.  That's why drumburp is a nice program
> because graphically, it sucks much less.
>
> Better would be some lilypond like syntax with different lines for the
> different instruments like the tab I wrote above.
>     
    So you want to enter (for instance) the Cr part as a separate
    LilyPond voice? In this case something like 1 R1 1 R1 or even \breve
    \breve? (The point being is that you want to enter durations and
    rests, even though they are in some sense arbitrary?)
> > What's *not* easy is:
> >
> > - convert a series of zero-time events (kicks, combined kicks) into
> > musical notation.
> >       
> It's not just that.  It's interweaving the kick and combined kicks with
> the other instruments grouped within that same voice.
>     
    I disagree insofar that I think that this isn't actually a
    complication. Interweaving instruments simply means forming a
    set-theoretic union, so to speak: We get a sum voice consisting of
    "a kick wherever one of the involved instruments has a kick". Then
    it's just a matter of representing the kick as one or multiple note
    heads, i.e. a note or a chord at the right point of time.


> Agreed.  But as I said it's not just about getting the kick notes
> correct.  Take for example something easy:
>
> Cr|X-------|
> Hh|--x-x-x-|
> Sn|--o---o-|
> Bd|o---o---|
>    1+2+3+4+
>
> Here if you have the bass drum grouped with snare (drums and cymbals),
> then each beat is on quarters:
>
> bd4 sn4 bd4 sn4
>
> But if the kick drum is all by itself (hands and feet), then you could
> have:
>
> bd2 bd2
>
> Although that would be weird.  I think:
>
> bd4 r4 bd4 r4
>
> would be more appropriate.
>

To demonstrate what I mean by set-theoretic union: If you want to group as drums+cymbals, we get the following diagram:

Cy|C-h-h-h-|
Dr|B-S-B-S-|
   1+2+3+4+

where C means Crash, h means Hihat etc.

If we want to group as hands+feets, we get:

Hd|C-Z-h-Z-|
Ft|B---B---|
   1+2+3+4+

where Z means "Snare plus Hihat" (and will be realised as a LilyPond chord <snare hihat>).

So you see, the actual problem is really just that of converting a series of "kicks" [i.e. virtual kicks comprising one or multiple instruments being played at the same point of time] into a sequence of durations, rests and ties.

Your last remark then forms one first rule for this: For |x---x---|, the LilyPond realisation as 4 r4 4 r4 seems so be preferred over 2 2.

This is a very general problem - algorithmically converting a sequence of points in musical time into a sequence of notes (/rests/ties). There are certainly lots of possible styles, depending on musical genre, time signature, instruments etc., but it's a problem that might be interesting to tackle. One just should think carefuly about the desired degree of configurability etc.

> Agreed.  But there are common ways to do it.  Perhaps those common ways
> could be used to steer decisions.
>
> If you're interested in how one project has tackled it, have a look at
> drumburp:
>
> <https://whatang.org/>
>
> Specifically this file (also attached):
>
> https://github.com/Whatang/DrumBurp/blob/1.1.2/src/Notation/lilypond.py
>     
    Thanks for the links, I hope I'll have time to dive into the source
    soon.
> Just while looking at drumburp, there are a number of example tabs of
> whole songs in the git repo.  I've attached the output of one of the
> songs in lilypond format, tab format, and pdf.  In the lilypond file, I
> see some interesting functions defined:
>
> #(define (rest-score r) ...
> #(define (merge-rests-on-positioning grob) ...
>
> But I don't yet know what they do.
>

In the example you posted, they seem not to do anything. But from reading the source, I think they are supposed to act in situations where LilyPond would create two simultaneous rests, in order to decide which one to keep.

Lukas