Re: [LAD] "enhanced event port" LV2 extension proposal

From: Jeff McClintock <jef@email-addr-hidden>
Date: Thu Dec 06 2007 - 23:25:14 EET

>> All of MIDI can be reduced to two messages: set-control (addressed by
>> Channel, Voice and Controller ID) .. and SYSEX - for data-dumps.

>What you say is of course a true statement, but it is ignoring what the
>founding fathers of MIDI attempted to accomplish regarding efficiency on
>a bandwith limited connection and the likelyhood of certain messages
>appearing more often than others ,,,

Hi Jens,
Very true. MIDI was an excellent solution at the time and made good use of
limited bandwidth.

  But I think as MIDI got extended it got too complicated, and now with
software music systems we don't have such limited bandwith anymore. The
8-bit messages are looking dated as far as simplicity and especially
resolution goes.

The MMA are addressing the resolution aspect with "High definition MIDI",
but it's implemented as SYSEX messages and therefore just adds more
complexity to writing a comprehensive MIDI implementation.

 I hope in future we'll have a simplified system...instead of CCs and RPNs
and NRPNs and SYSEX Controllers, have a single unified controller with a
32-bit Controller-Number range. Map all the existing MIDI messages into
that range.

NEW-MIDI would have only one message type, it would look like..

[Timestamp][Channel][Voice Number][Controller-Number][Value]

Anyway, I got way off-topic. My intention was to say that LV2 may not need a
per-note control extension because MIDI already has it via "Key Based
Instrument Controllers", which is a SYSEX message for sending a CC to one
specific note-number.

Best Regards,
Jeff

_______________________________________________
Linux-audio-dev mailing list
Linux-audio-dev@email-addr-hidden
http://lists.linuxaudio.org/mailman/listinfo/linux-audio-dev
Received on Fri Dec 7 00:15:09 2007

This archive was generated by hypermail 2.1.8 : Fri Dec 07 2007 - 00:15:09 EET