Hi,
On Wed, 15 Dec 1999, Rickard Lind wrote:
> I have dissected most of the patch.001 format (42 bytes still unknown),
> and it should be possible to map most MT-32 instruments to GM ones. It
> looks a quite bit better to me than relying entirely on patch.002.
Agreed.
> The
> MT-32 has 64 timbre memory locations holding userdefined instruments which
> will have to be mapped by "name" (like patch.002) or possibly emulated.
> The end result will hopefully sound better than the current solution,
> especially for Conquest of Camelot or Trial by Fire.
If I recall correctly, patch.001 is provided with all sound-supporting SCI
games, since MT-32 support was continued through all of SCI1 (while
patch.002, as you pointed out, is not present in Camelot and QG2, and
probably missing in SCI1, too).
> I'll write some documentation as soon as i'm finished.
Thanks! :-)
However, we're not actively working on audible sound support right now, as
this is just a "feature". Sound will probably not be enabled before the
0.3 release series (well, I've heard of no one wanting to write it before
that, and I'll start working on it only after the kernel function support
is complete).
> Is anyone else working at this right now?
No one I know of, but Francois-R Boyer ([EMAIL PROTECTED]) used to
work on MT32 for SCI0 and has written a "virtual device driver" for
ia32/NT. However, it only covers the Sysex data, not the header
(obviously). He sent me the sources; if you're interested, you can have a
look at them.
> /Rickard Lind
Chalmers university... you don't happen to know Magnus Reftel?
llap,
Christoph
