15 Sep
2008
15 Sep
'08
9:03 p.m.
Hi all,
For fluidsynth-devel list people, here is the complete thread of this message:
http://thread.gmane.org/gmane.linux.audio.users/50260
First of all: thank you very much, Roberto, for sharing your soundfont! Great
stuff!
On Sunday, September 14, 2008, Roberto Gordo Saez wrote:
Some problems have been detected in FluidSynth regarding dynamic range when
triggered by MIDI velocity parameters, and have been fixed recently in the
FluidSynth development repository by S. Christian Collins, who is also a
soundfont designer and author of the excellent GeneralUser GS SF2.
http://www.sccmusic.250x.com
I've been listening with your SF some pieces recorded with real Disklavier
instruments from this page: http://www.piano-e-competition.com/midi_2008.asp
some of them having a big range of dynamics, for instance, this one:
http://www.piano-e-competition.com/MIDIFiles/2008/Khmara02.MID
I think that the current development version of FluidSynth sounds much better
than 1.0.8 with these kind of music, but I would like to hear your opinions.
Regards,
Pedro
On Sat, Sep 13, 2008 at 03:02:19PM -0600, Stephen
Doonan wrote:
QSynth is only a GUI front-end for the FluidSynth library. Any soundfont that
can be used with QSynth is also valid for the FluidSynth commandline client,
and any other FluidSynth front-end.
I downloaded the soundfont and imported it into
qsynth, to play with my
external MIDI keyboard. Is this the way you yourself would use the
soundfont (with qsynth)? Or would you load the soundfont into some other
software or hardware?
Yes, it can be used in qsynth. I also use qsynth myself. In qsynth, the
base layer (piano) is triggered on MIDI channel 1, and
the other velocity layers (piano layer 1, 2, 3, 4 & 5) are assigned (by
default, anyway) to MIDI channels 2, 3, 4, 5 and 6.
Actually, only the channel 1 is needed, it does contain all the layers
included into it. They will be triggered automatically by programming
different velocity levels in a sequencer or playing your external MIDI
keyboard at different pressure levels. It is very easy, just attach
your MIDI keyboard to the channel 1 and start playing.
The other channels in the soundfont are there for debugging or special
purposes. For example, if you want a particular track of your sequencer
to always sound bright, you could attach it to the channel 2. Under
normal conditions you probably want to use the channel 1 only.
Note that the pressure levels may vary between keyboard models. In my
cheap keyboard I can only reach around velocity 110 no matter how strong
I press the keys. Other people could find the opposite: that the strong
layer is reached extremely soon, at soft levels. So that is what I was
asking for feedback on the different keyboards (my keyboard is not a
good reference).