Re: Sega sound boards...

>Space Fury 8 bit 22khz 1.2mb
>Star Trek 8 bit 22khz 1.0mb
>Zektor 8 bit 22khz 1.2mb
>
>the samples were created at 48khz 16bits and bandpass
>filtered prior to sample rate converison

They're all just voice-band stuff, no? Anyone have a convincing reason why
I couldn't resample these to 8KHz? (I don't recall the frequency response
of the SPO-250, but I doubt it's over 4KHz-- and even if it is, human voice
energy rolls off real fast past 1.6KHz...)

If I can fit 'em into 1Mbyte or less a "sample player" card is kind-of a
no-brainer. (1Mbyte is my max "pain threshold" for EPROMs now-a-days.)
Simple RLE compression would probably do pretty well on the samples too...)

I'm thinking of a 2051 controller with 2 27C040's, 4 74ls193's, a DAC and
some glue logic... The controller gets the sample commands from the G-80
I/O port, then parallel loads the start address of the appropriate sample
in the cascaded counters (193's). Then with each strobe of an I/O line on
the 2051 to the counters we get a new byte of sample data which can be
decompressed and output to a DAC (just an R2R ladder with an output filter
is probably fine).

(If you don't care about the compression you can hook the DAC more of less
directly to the EPROM.)

The good bit about this is that it's small and gets away from the SPO-250
nicely.

Any thoughts? Zonn-- you did this before, right? I did one back in
college without the ability to address the start address of a sample, but
it worked OK...

-Clay

Clayton N. Cowgill Engineering Manager
_______________________________________________________________________
/\ Diamond Multimedia Systems, Inc. clay@supra.com
\/ Communications Division http://www.supra.com/
Received on Wed Dec 3 13:51:27 1997