In a nutshell it is a means to allow devices to know which kind of other devices they are connected to, like a smart midi learn in their words.
1 Like
A device like an Iconnect ICM4+ or MIO4 gives you routing, filtering, and remapping options to make much more possible with a midi setupā¦
Gives you lots of control to route things to specific places, block them from others, etcā¦
With the right tweaks can make it seem like you have many more channels as you find ways to make them not interfereā¦
Comes with a new headscratcher being able to control so much, but provides worlds of flexibility once you wrap your head around itā¦
1 Like
Really interesting stuff!
Letās see if I understand it correctly. The main things are:
-
MIDI Polyphonic Expression
ROLI Seaboard is most famous for implementing this atm? Anyone know of other interesting gear currently under development? -
MIDI-CIā (Capability Inquiry)
Is this basically an automated way for gear to communicate with each other? And has future improvements in mind? -
Backwards compatibility
The linnstrument is MPE, and the haken continuum. Both pretty cool if you havenāt seen them. Roger linn has written a list of MPE capable synths on his website somewhere. Iāll post if I find it
edit: hereās some on the roll website. https://support.roli.com/article/mpe-compatible-synthesisers/
Roger linn claims there are 55+ hardware synths with MPE. Cant find the list though
This was already well-defined in the market, but itās now official.
It actually adds very little new, itās basically an agreement on using separate channels fo each note played. Here is a good introduction.
Itās again an way for devices to find out from eachother what they are, what they can do, and how they should be controlled.
It opens up the way for an updated version of MIDI at a higher speed with more channels and higher resolution data because devices can first figure out if they both support that new standard before switching to it over the same connection.
2 Likes
these type of little thing give your computer 4x16 midi tracks http://www.m-audio.com/products/view/midisport-4x4-anniversary-edition
@Holografik I already have the 2x2 version of this. But it wasnāt enough in the end. So I have just invested in a patch bay solution; a Roland A-880 and a Motu Midi Express XT. But it still needs some routing to avoid conflicts. So more unique MIDI channels to choose from would have made it much easier 
My 2004 Yamaha keyboard gives me 32 MIDI channels over USB.
This sort of thing has always been possible but generally not implemented.
Modern hardware often seems to be fairly stripped down, e.g. no digital I/O.
By defining two MIDI ports, I assume?
Lots of good info here.
Maybe the most simple answer as to āwhy 16?ā is because MIDI is based on Hexadecimal numbers.
Hex numbers are base-16.
So the numbers just make it work out really simply for 16.
Channels 0-1-2-3-4-5-6-7-8-9-A-B-C-D-E-F.
A note-on message for the first channel is 90, for the next channel 91ā¦the last channel is 9F.
ps. Hex works out really good for music (16 ticks per beat anyone?)ā¦something I love about using trackers to make music.
Yeah of course two ports, individual MIDI messages can only address 16 channels.
A bit of a stretch I think, it is probably more accurate to say that MIDI is a binary system and Hex is just the most efficient way to display binary numbers.
I suspect the 16 channel limit was simply based on efficiency of MIDI messages. The next option being to allocate a complete byte for MIDI channel identification, which would have provided 128 channels (i.e. way too many for the time).
because 16 bit in one word at the begining a Long Time ago.
Since most synths made in the last 10 years and onwards have USBā¦ it is time for a protocol that can handle the density of MIDI data that DAWs can send out. . . MIDI needs a re-write yesterday. When it was written there was barely any (?) CC control surfaces in the market so they were designing it for people playing on a keyboard and sending the odd mod wheel, pitch bend, and aftertouch message.
Thereās no protocol-defined speed limit for MIDI messages sent over alternative hardware transports such as USB. In practice, the limit is the speed at which the receiving device can process incoming messages.
Also, MIDI is the protocol, so it doesnāt really make sense to refer to a āprotocol that can handle [ā¦] MIDI dataā.
Thatās is incorrect. They very much did consider other input devices and included Control Change Messages to make sure the system was extensible.
Youāre aware the āCCā stands for Control Change, right?
DAW and USB are really not reliable for midi
Yes, but like I wrote people didnāt use CCs very much back then and thereās only so much you can send from a keyboard or the primitive MIDI sequencers of back thenā¦
I donāt know much about networks and such tech, so I donāt know the correct jargonā¦ but itās obvious that in 2018 and with USB 2 having been around for ages, the current MIDI speed/ bandwidth/ whatever is a relic that is dragging everything down, and certainly the reason why a lot of people gave up on MIDI and moved to working exclusively with DAWs and plugins. Making music on a setup consisting of interconnected MIDI hardware has become somewhat of a quaint relic from the past.
MIDI clock / sync issues are also the other related (and huge) issue that needs resolving. We can have multi-channel digital audio streaming at super high bit rates yet computers are having trouble with MIDI clock ā¦ in 2018.
There is obviously something wrong with the MIDI USB as implemented by various manufacturers because in many cases you get more reliable results just using the traditional MIDI DIN connection.
Not convinced that MIDI sequencers were primitive back then.
This.
I wouldnāt call it a stretch - seems like a perfectly logical and informed decision, but yea I see what yer saying. Thereās a lot of programmatic reasons to base off 16 is what I think it comes down to.
Either way, all speculation on our part so we might never know. Iāll ask Dave next time I see him if he has any insight, but I think everybody makes some good points! ;]