Just a note from the sound design aspect of this.

In order to implement what Peter is trying to do for everyone the default
soundfont would have to be modified extensively in order to have the
multi-velocity splits and other bells and whistles available to be acted on.

With Igevorse's work on MidiAction dialogue merged with the rest of the code
this would be perfectly possible without a huge superset of XML based
configuration.

As FluidSynth is based on the SF2 specification it is at it's heart a MIDI
synth. Although MuseScore can load soundfonts into it and specify the Ports
and Channels it receives on it still has to send MIDI messages to get
FluidSynth to play the notes.

All synths are like this whether they are hardware, VST3, or SFZ- they are
black boxes into which you send MIDI messages (or maybe in some cases OSC)
which then do their thing inside them to produce audio.

Speaking from my experience as a backing track programmer I identify
completely Luis Garrido's comment about "Score Editors Are Not Sequencers."
He described my workflow pretty accurately in that initial scoring of
orchestral backing tracks would be done in Finale (I hadn't yet discovered
MuseScore at the time) and then transferred via MIDI file to my DAW where
finishing of the audio would take place.

Even with the most sophisticated soundset file it is impossible to produce
audio capable of going for mastering without further tweaking in a DAW. For
example - the order the notes are played in a guitar chord are different
depending on whether the strum is up or down, and this has a subtle effect
on a guitar part, which will sound wrong if it is not rendered like this
from the MIDI file. There are many minutiae like this which have to be taken
when producing something more than just demo standard.

If the community wish to have a default soundfont like this, then I am quite
willing to begin to implement it. Indeed it could become an important part
of music and audio FOSS, but it is by no means a trivial task, and could
require funding to get some of the samples recorded - I have an idea about
that btw which is more suited to discussion in the open forum rather than
the developer list, but I need to do a little more work before I release the
idea into the public domain.

So, I suppose the question is: "Would there be enough interest from
MuseScore's general user base to warrant the enormous commitment in terms of
time and resources to produce such a soundfont?"

I do actually support Peter's idea btw despite my so far critical view of
it. I'm just not convinced that his implementation proposal is on a firm
enough footing regarding the basics of communicating with modern synths.

Regards
Michael



-----
Regards
Michael
--
View this message in context: 
http://dev-list.musescore.org/Playback-abstraction-layer-tp7579762p7579800.html
Sent from the MuseScore Developer mailing list archive at Nabble.com.

------------------------------------------------------------------------------
Find and fix application performance issues faster with Applications Manager
Applications Manager provides deep performance insights into multiple tiers of
your business applications. It resolves application problems quickly and
reduces your MTTR. Get your free trial! http://pubads.g.doubleclick.net/
gampad/clk?id=1444514301&iu=/ca-pub-7940484522588532
_______________________________________________
Mscore-developer mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/mscore-developer

Reply via email to