[Coco] Sound/Speech Cart

[Alex Evans]

That is inconsistent with the code you presented (which works). WDELAY
waits for BUSY*, then it calls BDELAY2.

The way I read the docs, BUSY* should be low when the PIC is busy (not
ready to accept input). However, as your code takes into consideration,
this is not the case. Maybe I misunderstand the docs.

Was the ~250us delay determined experimentally or some other way?

Oh, the switching the MPI at the beginning of the code is unnecessary as
the SSC is mapped at $ff7d, $ff7e which is not in the range $ff40-$ff5f so
it doesn't even have the option of using SCS*

On Thu, Feb 6, 2025 at 8:01 AM R. Allen Murphey via Coco <
coco at maltedmedia.com> wrote:

> On Wed, 2025-02-05 at 18:25 -0500, Alex Evans via Coco wrote:
> > Thanks! looking at this when doing direct register writes, busy* goes
> > high before it should?
> > I'll play around with this. The funny thing is the attached works for
> > speech.
>
> The BDELAY (big delay) and WDELAY (wait delay) are not there because BUSY*
> does
> something it's not supposed to - BUSY always seems to do the right thing
> for
> me, even if the "right thing" is "make me wait forever and 2 extra days
> before
> it's my turn again."
>
> There are 3 chips in the Speech Sound Pack you're dancing with:
> The PIC (and its two left feet)
> The Narrator (for speech)
> The PSG (for sound)
>
> I use the 22/44 loop delays to give the PSG time to latch the register
> address
> or data off of the PICs internal bus once the PIC has finally decided to
> get
> around to doing its part.
>
> The SSC is plenty fast enough to keep up with BASIC, especially when using
> the
> built in buffers for simple preset triggered events.
>
> But the Speech/Sound Pak is slow as molasses in winter on an icy moon in
> deep
> space when you're hitting it from machine language.
>
> Every delay in the PIC's setup piles up on you ... both ways ...
>
> --
> R. Allen Murphey <exile at weylan-yutani.com>
> Weylan-Yutani Corporation
>
> --
> Coco mailing list
> Coco at maltedmedia.com
> https://pairlist5.pair.net/mailman/listinfo/coco
>