I have thought about this for a while, and have written the first page of a design document at http://chalisque.com/Music/mi32design.pdf.

MIDI does not fulfil the design requirements (being able to handle high bit depths and multiple word languages within a bitstream, with precision timing to the bit) and neither does OSC, relying as it does on string handling and complex digital data.

If I want to trigger a sound, and a hardware synth knows exactly the parameters to the sound, I should, in principle, only need to transmit a single bit of information. MI32 is designed around this idea of minimalism together with the ability to have multiple wordstream languages of different bit-depths. At present it is highly unstable as I have not thought about bi-directional error detection and renegotiation protocols: that is for later.

As an example, if I want to modulate a synth parameter and my bitstream is clocked at 1MHz,
I should be able to accurately signal when down to a single bit, not even a word. Neither MIDI nor OSC allows this kind of precision and, for what I am developing, this precision is necessary.
I need to be able to signal high frequency modulation commands to discrete digital circuitry to convert to analogue and the route that into an analogue synthesiser. With instabilities in the circuitry, and it is precisely these instabilities I wish to explore, precision timing of events is essential.

For now there is what you find at the address mentioned earlier.

Any comments?

John

--------------------
-- S.E.Ch... chalisque.com...

Post Extras