Christoph Eckert wrote:
If it is not
open enough to take advantage of then is there
any truly open soundfont-like standard anywhere on the
planet?
for me it somehow looks as there's need to define a new,
absolutely free file format for sampling data.
It would be excellent if there was a totally free instrument
spec developed along the lines of ogg vs mp3. Some have
dismissed the idea as irrelevant because it would not have
the inertia of sf2, or even guspats, and certainly not the
respectability of professional Gigasampler formats.
Unfortunately, this would need lots of time and
resources, and
as 'we' are not enough users, it wouldn't be spread widely
enough.
A new soundfont patch standard would obviously include a
translator between it's own format and any other format.
However, a new standard would be possibly be greatly enhanced
over current specs as it would incorporate the best of what
is now available (knowledge wise) a decade after sf2 2.01
was designed and released. And here I mean specifically the
point of dealing with 24/96 and even 64/192 samples in at
8 least channel (7.1) formats. We could also take advantage
of Akai and Gigasampler format "tricks" that are unpatented.
So there are at least two very valid reasons to develop a
new soundfont spec regardless of the time frame and hurdles;
a) an unconditional no compromise free and open spec
b) to accommodate 21st century sound spec developments
by the end of this decade, cheap motherboard sound chips
will *probably* come in 7.1 channel 24/96 versions sounding
as good as todays mid-range priced semi-pro cards.
Greg Lee has extended the guspat format to accommodate 5.1
samples here...
ftp://ling.lll.hawaii.edu/pub/greg/gt-0.4.tar.gz
--markc