Re: [SPAM] Re: LilyPond benchmarks - request for proposals

[Urs Liska]

Am 01.03.2014 14:04, schrieb David Kastrup:
> Urs Liska <address@hidden> writes:
>
> > Hello LilyPond users,
> >
> > ...

> > And I think that - even with exposing our "imperfections" - we're not
> > giving away because this default output scores will always be soo much
> > better than most of the competition could hope for.
>
> I am not sure that "input without manual adjustments" is a natural
> concept for all WYSIWYG programs, so I'm not sure that the comparison
> with other programs should be a prime motivation.

It _isn't_, maybe I didn't make myself clear enough. All I wanted to say 
is that there is no need to hide such a comparison just because we're 
exposing weaknesses by it. I think LilyPond can be very proud of what it 
has achieved already in this respect.

> It's more like seeing
> for ourselves where we could do better.

That was exactly my motivation. So far we don't have a corpus of 
representative examples to measure against.

> Of course, it would be nice to
> see the progression (hopefully) for 2.0, 2.2, 2.4, 2.6, 2.8 ... 2.18.
>
> > Today I'd like to ask for suggestions: a) to assemble a set of score
> > _types_ to be used in that benchmark series, b) possibly for concrete
> > suggestions.
> >
> > Of course the example scores should expose (more or less) exemplary
> > input style.
>
> I suppose that rules out my favorite bad tie demonstration subject (see
> below).  But it probably can be converted to "canonical" non-computed
> form by using \displayLilyMusic on the generated music and paste the
> results.

No it doesn't rule that out. This actually is a very good example, and 
it will be useful to show any progress (or the lack thereof).
Of course we can have such a "generated" example, I just wanted to say 
we should take care not to expose sloppy coding here.
My example probably has to be cleaned up to match that claim, too ;-)

> > My ideas so far are:
> >
> > - classical piano piece
> >
> > - string quartet
> >
> > - solo instrument
> >
> > - moderately complex orchestral score
> >
> > - SATB choir with piano
> >
> > - complex song
> >    -> We'll probably be able to use one of our Fried songs for that
> >
> > - complex score from 20th century (not graphic yet)
> >    -> I have a proposal for that already:
> >
> > https://dl.dropboxusercontent.com/u/49478835/lily-examples/variations-and-theme-default.pdf
>
> Well, the following tie-rich example is a Jazz chord progression.  Jazz
> chords are likely a good contestant for multiple tie configurations.

I'll definitely take it to the list.
Of course it doesn't matter either that it isn't a full-fledged composition.

Urs