denemo-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: resizing the virtual MIDI keyboard


From: Bric
Subject: Re: resizing the virtual MIDI keyboard
Date: Sun, 9 Feb 2020 13:38:58 -0500 (EST)

> On February 9, 2020 at 12:57 PM Richard Shann <address@hidden> wrote:
> 
> 
> On Sun, 2020-02-09 at 12:38 -0500, Bric wrote:
> > > On February 6, 2020 at 8:01 AM Richard Shann <address@hidden.c
> > > om> wrote:
> > > 
> > > 
> > > On Wed, 2020-02-05 at 18:09 -0500, Bric wrote:
> > > > I am running denemo with a touch-screen monitor right now.
> > > > 
> > > > It occurs to me that it would be totally awesome to make the
> > > > virtual
> > > > MIDI keyboard resizeable such that the keys actually expand to
> > > > greater pixel dimensions, to one's liking
> > > > 
> > > > With touch-screen you can actually use your fingers to tap the
> > > > screen
> > > > and play (!). Except right now it does not re-scale (only lets me
> > > > resize the window horizontally, and with that it simply fills the
> > > > greater width of the window with *more* keys, rather than expand
> > > > the
> > > > keys' size), and my finger tips are too big for the keys drawn on
> > > > my
> > > > screen.  If i resized the keys by just 30%, say, it would be an
> > > > awesome way to input data.
> > > 
> > > That sounds like a good idea. It is in fact already possible: if
> > > you
> > > type
> > > (d-VirtualKeyboard 2)
> > > into the CLI box of the Scheme window and hit Return you get a two-
> > > octave keyboard which you can then resize to fill the width of your
> > 
> > This works!  I invoked it with three octaves (changed the 2 to 3),
> > and that works fine with my 1600x900 resolution. When I resize
> > horizontally, the width of the keys also extends to where you can
> > avoid fat finger touch misses)
> > 
> > So, nice touch-screen play/input. Might come in handy for some
> > scoring, at some point. (Right now, the purely typing input speed is
> > still very competitive with the keyboard input - given the need to
> > change duration and a slew of other properties)
> 
> The duration change is well handled by putting in multiple durations -
> a whole bar, line or movement before starting to play on the MIDI
> keyboard. The numeric keypad is set up so that slurs can be added
> without extra keypresses when entering the durations and ties require
> only one keypress on the tied note. For my work I also insert dynamics
> and ornaments as I enter the durations - I make sure these require only
> key strokes to do. Then I can start playing the pitches and the cursor
> automatically goes back to where I started entering durations.
> I noticed some pitch-spelling errors in the Denemo file you sent - Eb
> for D# etc, which is a hazard if you are playing the notes on a MIDI
> keyboard, but for tonal music this is mitigated by sounding augmented
> and diminished intervals on a different channel.
> 

The denemo + lilypond engraving handles enharmonic accidentals much more 
intelligently, than, say, MuseScore, or even Rosegarden. (Although Rosegarden 
has the convenient command "re-spell as [sharp|flat]"). MuseScore is sloppy, 
spelling what should be flats as sharps, etc., with the little bit of tinkering 
i've done with it.

I don't fully understand your denemo input process you describe above. You 
input a sequence of just durations somehow, ahead of playing a midi keyboard, 
and denemo then attaches each duration in the sequence to the sequence of 
midi-entered notes?
 


> > 
> > I should also note that "Virtual MIDI Piano Keyboard" (VMPK) does not
> > respond to touch-screen input for some weird reason - only mouse
> > clicks.
> Are you referring to
> http://vmpk.sourceforge.net/

Yes. vmpk doesn't seem to respond to touch.
So, the denemo one is better coded, it appears.
Or perhaps the framework it's built into.

But, speaking of framework - something needs to be done with the info balloons 
still. They are a great nuisance. The info balloon gets locked, such that it 
can only be dismissed by hover over /it/, and nothing else.  The mouse thus has 
to travel back to the (accidentally triggered!) balloon to dismiss it. (And 
nothing functions until it is dismissed) A great pain.



reply via email to

[Prev in Thread] Current Thread [Next in Thread]