MIDI is pretty great, I am about as musically inclined as a wet rag, but I recently bought a bank of midi sliders as it was the cheapest way to get a bunch of dials, buttons and sliders. At first I was looking for libraries to interface with it. But after looking at the spec, it turns out midi is dead simple and designed to be easy for hardware to parse. Super easy to read and generate. So I am sitting here happily hacking away building the dispatch interface for this thing. The icing on the cake is that it turns out openbsd's sndiod includes a midi transport and control system. Which was just one less thing to worry about.
Update: that was not a great commands introduction, here Are my thoughts. if bit 1 is set it is a command otherwise it is data. commands are split into two half bytes(nibbles?) the first 4 are what command it is(first bit is high so 3 bits or 8 commands) the second four are the channel. some commands have two following data bytes(low first bit remember) and some have one. The main exception is the system command. which is complicated
Novation has a web browser editor for customizing the 8 midi banks, using what I assume is the web USB protocol so it can be customized on tablets too. It’s been a joy to use and set up custom workflows, you one for dawless, one for ableton etc.
MIDI should be studied by anyone interested in protocol design. Very few protocols had its staying power - despite not being perfect for every type of music, it has done its job admirably for close to 50 years due to its simplicity and flexibility.
I think we are lucky it was created when it was. The temptation to make a much more complex and fragile protocol would have been harder to avoid if they weren't constrained by what could be cheaply and efficiently implemented on microcontrollers in the early 1980's.
MIDI as a non-proprietary, non-patented, non ad supported or privacy invasive protocol opened so many innovation from synth and others electronic musical instruments constructors or makers. We really need this type of candid innovation back.
Midi being an “artist” tool places it more as a medium like paint. The artist was able to use it to build complex worlds.
MIDI’s cousin RS-422 existed in a technical environment but probably enjoys as much use as MIDI but technicians have many protocols to choose from. the artist just needs to get it speak MIDI to feel in control and the artist generally has fewer stakeholders, which I believe helps maintain continuity of an implementation.
MIDI is just good enough for the vast majority of musical needs. Any more complicated and the protocol would have been too hard to implement on early computers.
MIDI was revealed to the world, on the Sequential Circuits stand at the NAMM show in January 1983 (precisely one year since Smith’s proposal got the brush off)
Very cool article! I had no idea that the idea to use a 5-pin DIN connector for MIDI came from the Japanese HiFi world. One very minor nitpick: 'Running Status' actually refers to a feature in the MIDI specification where multiple sequential messages of the same type can omit the 'status byte'[0]. Stupidly, I didn't know about this feature in the specification when I was writing the MIDI handling routine for a synth[1]. This introduced a bug that took me ages to squash. I was struggling to figure out why everything worked fine when playing notes from my DAW, but notes were getting stuck when directly connected to a synth.
That makes me wonder how often one runs across a synth that doesn't support running status. I'm working on a MIDI controller myself, and that's a thing I probably ought to enable if I want the lowest latency.
It seems the standard MIDI libraries that Arduino uses don't enable it by default, but it's a configuration option you can turn on, along with a note not to try to use it with USB[1].
> Korg noted that the use of jack plugs to transmit the required signals would cause noise in the system.
No more detail on this, but it was huge for MIDI not to end up "becoming a nightmare of endless fine-tuning better suited to electronics engineers than musicians" like CV/gate was.
Mieda-san from Korg responds on behalf of the Japanese companies to the meetings at the 1981 Gakki Fair with Sequential and Oberheim and confirms the discussion that were they had at the Gakki Fair.
• 19.2kbps is too slow
• ¼” Jacks will have ground loop problems
• There is no concept of synchronization, clock or the ability to start and stop sequences
---
Ground loops are the bane of musician's lives and sometimes the mitigation is the end of their lives
That’d match the analogy from TFA if it only ran on, say, a 2014 mbp, so while it’s true Logic runs only on macs or ipads (I guess just pros for the latter though I don’t know), I think it’s quite a stretch to say that’s a certain make and model of computer.
"freeing for the first time the act of composition from the ability to play an instrument". I think the player piano, invented in 1896, might have something to say about that. There are modern recordings of piano rolls (eg, Scott Joplin) but I'm wondering if they were ever used like that historically (or in concert). You could make your own piano rolls or record a performance and edit out mistakes (eg, Scott Joplin)
The master rolls for player pianos were made by actual pro players playing in real time. They needed to have the performance chops, even if they maybe were not the composer in every case.
This quote about "separation of composition from performance" applies when you are working with a sequencer because the composer can "step-edit" the musical data note-by-note into the sequencer and then have the sequencer play it back in proper musical (real) time.
One of the things I like about modular is that the v/octave CV standard imposes no assumptions on what kind of scale you're using. 12-tone equal temperament is pretty easy to use, but there's nothing forcing you to that scale if you don't want to use it. (Some modules and MIDI to CV devices have built-in scale assumptions, but it's not baked in to Eurorack modular at the protocol level.)
I use MIDI in projects when I have to, but I usually lean on OSC because I can send it over wifi from a microcontroller and send floats. It just makes a lot of what I do so easy. That said, my stuff isn't very plug-n-play. Would love to see more development of new protocols like O2 and O2lite to build networked ensembles.
The fact that the Atari STs had MIDI built in was the main reason that they were so successful as computers in recording studios (and that they were much cheaper than Macs certainly helped a bit as well)
I remember going to Sequential Circuits on 1st street in San Jose (on my lunch hour) to buy a copy of the MIDI Specification (I think it was $10). Probably late 1986.
I think the Lumatone uses MIDI 2.0, but I haven't actually seen one of those in person.
I'm working on my own microtonal MIDI controller, and for now it makes more sense to just use MPE. Maybe I'll support MIDI 2.0 eventually, but I'm kind of waiting to see if MIDI 2.0 is a flop first.
[+] [-] somat|1 year ago|reply
https://www.akaipro.com/midimix
https://computermusicresource.com/MIDI.Commands.html
http://man.openbsd.org/sndiod#MIDI_CONTROL
Update: that was not a great commands introduction, here Are my thoughts. if bit 1 is set it is a command otherwise it is data. commands are split into two half bytes(nibbles?) the first 4 are what command it is(first bit is high so 3 bits or 8 commands) the second four are the channel. some commands have two following data bytes(low first bit remember) and some have one. The main exception is the system command. which is complicated
[+] [-] squarefoot|1 year ago|reply
http://www.ucapps.de/
https://www.elgato.com/us/en/p/stream-deck-mini
https://github.com/streamdeck-linux-gui/streamdeck-linux-gui
[+] [-] dmix|1 year ago|reply
Novation has a web browser editor for customizing the 8 midi banks, using what I assume is the web USB protocol so it can be customized on tablets too. It’s been a joy to use and set up custom workflows, you one for dawless, one for ableton etc.
[+] [-] glimshe|1 year ago|reply
[+] [-] TazeTSchnitzel|1 year ago|reply
[+] [-] makapuf|1 year ago|reply
[+] [-] detourdog|1 year ago|reply
MIDI’s cousin RS-422 existed in a technical environment but probably enjoys as much use as MIDI but technicians have many protocols to choose from. the artist just needs to get it speak MIDI to feel in control and the artist generally has fewer stakeholders, which I believe helps maintain continuity of an implementation.
[+] [-] em3rgent0rdr|1 year ago|reply
[+] [-] chrisjj|1 year ago|reply
Absolutely. A shambles hopefully never repeated.
[+] [-] T-A|1 year ago|reply
MIDI was revealed to the world, on the Sequential Circuits stand at the NAMM show in January 1983 (precisely one year since Smith’s proposal got the brush off)
[+] [-] JKCalhoun|1 year ago|reply
[+] [-] ajxs|1 year ago|reply
0: https://midi.teragonaudio.com/tech/midispec/run.htm
1: https://github.com/ajxs/yamaha_dx97
[+] [-] elihu|1 year ago|reply
It seems the standard MIDI libraries that Arduino uses don't enable it by default, but it's a configuration option you can turn on, along with a note not to try to use it with USB[1].
[1] https://github.com/FortySevenEffects/arduino_midi_library/bl...
[+] [-] morganw|1 year ago|reply
No more detail on this, but it was huge for MIDI not to end up "becoming a nightmare of endless fine-tuning better suited to electronics engineers than musicians" like CV/gate was.
https://midi.org/midi-history-chapter-6-midi-begins-1981-198... spells it out in more detail:
Mieda-san from Korg responds on behalf of the Japanese companies to the meetings at the 1981 Gakki Fair with Sequential and Oberheim and confirms the discussion that were they had at the Gakki Fair.
• 19.2kbps is too slow
• ¼” Jacks will have ground loop problems
• There is no concept of synchronization, clock or the ability to start and stop sequences
---
Ground loops are the bane of musician's lives and sometimes the mitigation is the end of their lives
https://en.wikipedia.org/wiki/Ground_lift#Safety
[+] [-] everyone|1 year ago|reply
[+] [-] fipar|1 year ago|reply
[+] [-] ksherlock|1 year ago|reply
[+] [-] tanseydavid|1 year ago|reply
This quote about "separation of composition from performance" applies when you are working with a sequencer because the composer can "step-edit" the musical data note-by-note into the sequencer and then have the sequencer play it back in proper musical (real) time.
[+] [-] golergka|1 year ago|reply
[+] [-] steve1977|1 year ago|reply
A popular one was made by Doepfer - who were also a strong force behind the modular renaissance with their A-100.
[+] [-] elihu|1 year ago|reply
[+] [-] worik|1 year ago|reply
Yes you can. I do
[+] [-] _spduchamp|1 year ago|reply
[+] [-] fyt2024|1 year ago|reply
[+] [-] steve1977|1 year ago|reply
[+] [-] happyweasel|1 year ago|reply
[+] [-] tanseydavid|1 year ago|reply
[+] [-] memalign|1 year ago|reply
https://chiptune.app/
[+] [-] worik|1 year ago|reply
[+] [-] elihu|1 year ago|reply
I'm working on my own microtonal MIDI controller, and for now it makes more sense to just use MPE. Maybe I'll support MIDI 2.0 eventually, but I'm kind of waiting to see if MIDI 2.0 is a flop first.
[+] [-] readyman|1 year ago|reply