[PD] Re: [PD-dev] We've got to undo the MIDI revolution! - Where isOSC?!

[Martin Peach]

geiger wrote:

>On Wed, 15 Mar 2006, Mathieu Bouchard wrote:
>  
>
>>FUDI wasn't designed to "undo" the MIDI revolution.
>>
>>Is there OSC over RS232 ?
>>    
>>
>
>OSC doesn't specify the transport layer AFAIK, so yes, there is.
>If your question is if there is an implementation, then I would
>say I don't think so. You will see more implementations of the
>MIDI protocol over RS232 (MIDI hardware != RS232) because MIDI
>is easier to implement.
>
>In that sense FUDI is very nice in that it is human readable,
>but its implementation is quite complicated as you need a
>parser for the number values. And it needs a bit more bandwidth
>under most circumstances.
>
>MIDI is a hardware- and protocol specification, OSC
>is just a protocol. With a decent 8-bit clean morse alphabet
>you could even transmit it by drumming :)
>But encoding and decoding it is pretty complicated.
>  
>
Complicated compared to what?
It could be made simpler for low-speed serial systems such as MIDI and 
RS232 by simply sending the whole thing as ASCII text instead of 4-byte 
padded binary. Bit 7 (the MSB) can then be used to delimit packets. In 
MIDI it could be sent as a sysex message without having to pack and 
unpack 8 bits to 7 bits. It would also be human-readable for debugging.
As far as PD goes, it would be nice to have the OSC objects use the same 
code as netsend and netreceive, that way TCP could be used as well as 
UDP, and all the networking would use the same code, i.e. the code in 
x_net.c.
At the moment the PD OSC objects use Adrian Freed's htmsocket.c which 
handles its own (UDP or UNIX) sockets.

Martin