Re: Q: Applied Engineering Sound Card

[spec@vax2.concordia.ca (Mitchell Spector)]

In article pubpc1@library.ucla.edu writes...

>spec@vax2.concordia.ca (Mitchell Spector) wrote:
>>In article ToasterKing <toastjuice@aol.com> writes...
>>>Third, I'm pretty sure the card only plays 16-bit audio
>>>and not the crappy audio from 8-bit programs, which is played through
>>>the internal speaker.
>>
>>    There is _no_ such thing as 16-bit sound resolution on an Apple
>>IIgs.  The Ensoniq is limited to 8-bit resolution, and thus, that is
>>only what software supports (it should be possible to convert 16-bit
>>waveforms into 8-bit, but I'm not aware of any software on the IIgs
>>to do that).
> 
>I think he isn't saying that the Ensoniq is doing 16-bit resolution.
>He is saying the Ensoniq connector for the sound card outputs "audio"
>from "16-bit" programs while the "audio" from "8-bit programs" won't
>be played through the sound card, but only through the internal
>speaker.

    Reading over his post a second time, I thought that as well. 
Regardless however, the statement "plays 16-bit audio" does in
fact imply 16-bit sound resolution, and I wanted to correct that.
(the poster may know that is wrong, but someone else reading it 
may very well take what he said literally).

    You have to be careful not to mix terms and keep distinctions
separate. For example, on the PC you can have an 8-bit ISA sound
card which has a 16-bit DAC for digital sound (the two are entirely
separate and have little or nothing to do with each other). The
same is true of the Apple IIgs, it being a 16-bit machine has no
relation or influence on the sound chip which is 8-bit. Again it's
just a measurement ("bits") of two entirely _non-related_ things.
Incidentally, you can write an 8-bit application that accesses the
Ensoniq chip, NoiseTracker GS is one example (I've even seen some
DOS 3.3 programs do things with the Ensoniq, but of course they
won't work on anything but a GS). I think the term "native audio"
is a better suited and a less confusing term than "16-bit audio"
when discussing the Apple IIgs.

>>    The internal speaker outputs both audio from the Ensoniq chip (it
>>sounds pretty crummy, but what wouldn't from that tiny little 8 Ohm
>>^^^^^^^^^^^^^^^^^^^^
>>non-amplified speaker) and speaker toggling techniques used by 8-bit
>>Apple II software in emulation mode. A stereo card such as the AE
>>SonicBlaster doesn't enhance the sound capabilities of the IIgs, it
>>merely taps the left/right channels from the Ensoniq and produces a
>>true stereo signal (Apple never connected stereo to the back port,
>>even though the chip supports it; same with it's built-in ADC).
> 
>Um, I take it you mean that the *tiny built in speaker* sounds
>crummy.  The Ensoniq 5503's output itself is quite nice (about the
>quality of a good high bias cassette tape, IMO), but you just need
>to connect a pair of decent high fidelity speakers to the headphone
>jack or sound card to hear it at full quality.

    Certainly that's what I meant, the little speaker isn't much good 
for anything beyond the simple 'beeps' and 'clicks' produced by classic 
8-bit Apple II+/e/c software (and in certain cases, even then it can be 
barely suitable at times :). Since the first day I brought my Apple IIgs 
home, one of the first things I did was plug in a pair of headphones or 
amplified speakers into the 1/8" audio jack to listen to native audio.
Of course within a few months I purchased a stereo-output card so I
could get the full effects of the chip (nothing too exciting to show
off stereo during the days of Music Construction Set and Music Studio,
but that soon changed once Soundsmith and synthLAB come out, or more
so with Beatbox and Shellplay! :)

    As for the Ensoniq 5503 DOC, it is a very decent sound chip, but
Apple seriously crippled it within the Apple IIgs. This included *not*
providing stereo-out or sound-in for the user (mono only, and no place
to insert a microphone to record sound, despite the chip supporting 
both stereo and digitizing; you had to buy a 3rd party add-ons). Also
the fact that they supplied only 64K of DOC-RAM, with no way to expand,
when the chip can easily address up to 128K. They further limited the
chip by not isolating it properly from the motherboard circuity, so lots
of "noise" leaked into the Ensoniq (this was improved with the ROM 3
motherboard, but not entirely fixed, which I suspect _was_ with the 
never-released "Mark Twain" prototype, judging by what I saw and read).
The final blow to the chip was the low-pass cut-off filter that the
GS added. Apparently, from what I'm just finding out now, regardless
of whether the chip or software played a sample at 44.1 kHz (or other
high frequencies) the filter would essentially drop the quality down
to around 26 kHz.

    Another annoyance about the chip, was the lack of software to take
full advantage of it, especially coming from Apple Computer. It wasn't
until 1992 that Apple packaged MIDIsynth (the first decent and useful 
sound tools IMO) with System 6.0. MIDIsynth was also the first, and so
far best way, to do MIDI playback using the GS as a synthesizer. Still
it's far from perfect, MIDIsynth limits you to 7-voices and only 64K
of bank patches (8-bit resolution with that low-pass filter), of which
an instrument can be no more than 32K in size. It's fairly passable
for MIDI sequences that use _few_ voices and instruments, but just 
barely these days. Don't expect GM/GS MIDI support here. ;) It's still
leaps and bounds above those FM-synthesis boards for PC, but that's
almost moot since the PC world has gone wavetable (as for the Macintosh,
sans avec synthesis, that's a different story ;-)

Mitchell Spector
spec@vax2.concordia.ca