Toughts about a Propeller based synthesizer

[6581]

The day I've discovered Ahle2's SIDCog, I was sure I'm going to build
a SIDCog based synthesizer. I've started quickly to design a PCB to try
out the SIDCog and Propeller capabilities. The board itself contains
MIDI I/O (Input and Output - no MIDI thru), a SD-Card slot, 24LC256
EEPROM, 5 (yep, 5 not 6 - due to board size limitations) line-level
Audio Outputs, VGA (for debugging purposes) and PS/2 keyboard and
mouse connectors (for no obivious reason :lol:). Quickly I've started to
dive into the SPIN language. With the help of some users here on the
forums I've successfully written a little SIDcog based MIDI synthesizer
firmware supporting basic features as note on, note off, pitch bending,
modulation wheel support, mapping the SIDcog parameters to MIDI
midi controllers (using the controller change events). Since MIDI is
was designed around the 80ies, it's a quite simple and easy to implement
protocol - especially on a 8-cog microcontroller leaving a lot of space
for other functions. The next step I did was adding some HD44780 2x40
LCD to the propeller, an 12-bit 8 channel analog-to-digital converter and
hook up a potentiometer to it.

This was the first thing I was playing with. Now, I've decided to build a much
more extended version of my experimental synthesizer with about
30 potentiometers, a lot of LEDs, a HD44780 LCD, some 7-segment
displays and a higher quality audio output - all modular. This is what
I'm doing right now.

This is how the prototype does look like:
http://loom.no-route.org/sidthesizer/1.jpg

http://loom.no-route.org/sidthesizer/2.jpg

http://loom.no-route.org/sidthesizer/3.jpg


What is your idea for a Propeller based synthesizer? Do you like the
idea?

[Humanoido]

It's a great idea. The propeller has plenty of capability for sound synthesis. You could also design it to reproduce phonemes with some extra switches and circuits.

[Invent-O-Doc]

Very neat project! The SID approach is quite innovative. Check how memory and COG intensive the SID project is- often emulators are hogs. The reason I say so is that a good voice synthesizer will ideally use only one COG and will not consume much memory.

[Ahle2]

@6581
It looks really nice... :)

@Invent-O-Doc
As the name suggests, SIDcog uses ONE cog.
It doesn't need much memory.
It has got a full fledged ADSR Envelope for each channel.
It has got a sweepable multi-mode filter (any combination of lp + bp + hp).
Other interesting features are ring modulation and oscillator synchronization.
For what he is trying to do (subtractive analog synthesis), other sound objects for the prop isn't sufficient at all.

[erco]

Man, I hope you guys keep on this. It's beyond my capabilities, but it should certainly be doable at the corporate level (right, Parallax?). Text-to-speech synthesizers were a dime a dozen back in the 1980's, for the C64, Vic 20, TI99-4A, etc. It's really surprising now that there are no cheap plug & play solutions for this today. Speakjet are expensive & hard to find. A complete Prop-based text-to-speech converter module under $50 would really put Parallax on the map, even with the Ardweeny crowd.

How great would THAT be, when those guys ponied up to buy one. Parallax could charge extra for one in a "shield" configuration.

[Cluso99]

IIRC Chip was working on a phonome?? project for the prop to speak. Anyone recall this???