[comp.sys.apple2] GS Sound underestimation

unknown@ucscb.UCSC.EDU (The Unknown User) (05/06/90)

In article <11669@netcom.UUCP> lyle@netcom.UUCP (Lyle Fong) writes:
>The sound on the IIgs is among the BEST in the PC industry, capable of playing
>15 sounds simultaneously with Orchestraic quality.  It is better than most
>stock Macs, and better than IBM's even with AdLib sound cards on them.
	Well, you have underestimated it! As far as I've read/been told,
the GS's sound chip is capable of 31 channels simultaneously, yet through
"legal" means, 15 is how many you get... As we've all seen though, trickery
gets you much more than going through the legal means.

-- 
  _    __       __           __       _     /-------------------------------\
 /_\  |__| |    __|    |_|  |_  \  / |_|   |    unknown@ucscb.ucsc.edu       |
/   \ |    |__ |__       |  |__  \/  | \   | unknown%darkside.com@ames.arpa  |
Computer engineering student seeking a job. \-------------------------------/ 

haris@wam.umd.edu (Chris Mazzola) (05/07/90)

Actually, the Ensoniq chip used in the Apple IIgs has 31 oscillators, but they are each paired to give a richer sounding machine.  So, on the GS, only 15 at once, but on an Ensoniq keyboard (which are WONDERFUL), 32.

-Chris Mazzola
-haris@wam.umd.edu

rlw@ttardis.UUCP (Ron Wilson) (05/08/90)

In article <3133@darkstar.ucsc.edu>, unknown@ucscb.UCSC.EDU (The Unknown User) writes:
>In article <11669@netcom.UUCP> lyle@netcom.UUCP (Lyle Fong) writes:
>>The sound on the IIgs is among the BEST in the PC industry, capable of playing
>>15 sounds simultaneously with Orchestraic quality.  It is better than most
>>stock Macs, and better than IBM's even with AdLib sound cards on them.
>	Well, you have underestimated it! As far as I've read/been told,
>the GS's sound chip is capable of 31 channels simultaneously, yet through
>"legal" means, 15 is how many you get... As we've all seen though, trickery
>gets you much more than going through the legal means.

It's NOT really trickery.  The Ensonic chip used in the //gs does indeed
have 32 logical oscilators.  The //gs sound toolkit uses one those as a
timing source.  30 are then used in a paired mode of operation, leaving
one unused by the sound toolkit.

In the paired mode of operation (which IS designed into the Ensonic chip),
one "oscilator" plays the initial or attact portion of a sound, then the
other "oscilator" is automatically triggered (in the chip, no software
intervention required) to play the sustain portion of the sound (looping
until the software tells the chip to stop the "oscilator").

This mode of operation, while "wasteful" of potential "polyphony," was
devised (by Ensonic for their Mirage sampling "synth") to provide a more
realistic sound within the constraints of 64K of RAM (even on the Mirage,
no single sound can occupy more than 64K - the Mirage, however, can access
multiple banks of SOUND RAM, 2 at a time - the //gs only provides 1 bank of
*SOUND* RAM - the sound chip can't access any the "system" RAM).

I known that "Music Construction Set" uses this mode of operation,  HOWEVER,
some other programs (such as both version of Music Studio I have (1.2 and 2.0))
do not use this feature - In fact, it's very obvious: the higher the pitch of
note being played, the shorter it's duration (ie: notes can not be sustained).

One program, Diversa-Tune, implements this feature in software so as to
make use of all 32 oscilators.

It would have been nice if Apple's sound toolkit does what Divers-Tune does,
but I guess they (Apple) thought the CPU overhead would be to large for
anything but a dedicated music program to use.  With the paired oscilator
mode of operation, the program only has to worry about starting the note and
then stopping it - it doesn't need to "reprogram" the oscilator partway
through the playing of the note.  (of course, if it WANTS to reprogram the
oscilators on the fly, it IS free to do so - Apple's sound toolkit just
doesn't provide a defined way to do this)


----------------------------------------------------------------------------
"We were wondering when somebody would discover that one."
	- The response of a customer support rep to a customer's phoned in
	  bug report (name of company with held to protect the guilty).

rlw@ttardis	rel.mi.org!ttardis!rlw	  sharkey!cfctech!ttardis!rlw