MIDI Association Announces MIDI 2.0 Prototyping (hackaday.com) 133
MIDI was introduced at the 1983 NAMM show as a means to connect various electronic instruments together. Since then, our favorite five-pin DIN has been stuffed into Radio Shack keyboards, MPCs, synths, eurorack modules, and DAWs. The standard basically hasn't changed. Now, ahead of the 2019 NAMM show, the MIDI Manufacturers Association (MMA) in conjunction with AMEI, Japan's MIDI Association, are announcing MIDI 2.0.
From a report: The new features include, "auto-configuration, new DAW/web integrations, extended resolution, increased expressiveness, and tighter timing." It will retain backwards-compatibility with MIDI 1.0 devices. The new initiative, like the release of the first MIDI spec, is a joint venture between manufacturers of musical instruments. The company lineup on this press release is as follows: Ableton/Cycling '74, Art+Logic, Bome Software, Google, imitone, Native Instruments, Roland, ROLI, Steinberg, TouchKeys, and Yamaha.
Google (Score:1)
Does Google just sign up for every multi-company initiative as a matter of course? Are we going to hear next about their participation on the ratings panels for small gas engines or feminine hygiene products?
Re: (Score:2)
Re: (Score:2)
Does Google just sign up for every multi-company initiative as a matter of course?
Yup.
Seriously, yes they do. They want to either be driving everything or, failing that, at least in a position to influence it to go the way they want it.
About time! (Score:5, Interesting)
Re: (Score:2)
Also, the DB15 VGA connector - introduced in 1987. Dropped out of standard usage but still not uncommon yet.
Re: (Score:1)
Incidentally, the DB15 connector isn't just a "joystick connector".
On pin 12 and 15 of the game port you have the signals used to connect the MIDI-interface.
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Also, the DB15 VGA connector - introduced in 1987. Dropped out of standard usage but still not uncommon yet.
People use it in spite of itself! In 2010 a whole host of manufacturers said they will drop it by 2015. [engadget.com] While it might be a "lowest-common denominator" connector, I'm surprised how many people still use it. Existing concealed wiring to a 10 year old 1024x768 projector? Sure, VGA is fine. Trying to use a crappy old monitor, or crappy old computer, sure VGA is fine. Same with an old KVM switch.
But I'm boggled by how many people use a Displayport-VGA adapter, or HDMI-VGA adapter, so they can connect to a 1080p
Re: (Score:2)
"Hey, they put a dongle in the box, I must be supposed to use it!"
And, then they ask IT for a VGA cable so they can use their HDMI to VGA dongle.
And don't mock the warm, fuzzy picture or else hipsters will cotton on to it and start preferring VGA over DisplayPort.
Re: (Score:2)
No need to ask IT for a cable. The monitor either comes with one, or they just find one. People usually keep the power and VGA cable off their old monitor "Just in case". Even if they already have 6 cables in their drawer.
Re: (Score:2, Interesting)
Midi is used by content creators. The music industry tries not to fuck with them. It's consumers that the music industry is waging a war with.
Secondly midi is used by professionals. Equipment manufacturers try hard not to fuck with them either and keep to the same standard for a very long time to keep customer loyalty. Like Canon who kept with the same lens mount and memory card standard for a decades, they dont change plugs and ports very often at all. The only outlier I see in this is Apple.
Re:About time! (Score:4, Insightful)
Midi is used by content creators. The music industry tries not to fuck with them. It's consumers that the music industry is waging a war with.
Secondly midi is used by professionals. Equipment manufacturers try hard not to fuck with them either and keep to the same standard for a very long time to keep customer loyalty. Like Canon who kept with the same lens mount and memory card standard for a decades, they dont change plugs and ports very often at all. The only outlier I see in this is Apple.
As someone who works in the professional music industry, manufacturers have no problem fucking with creators. The connections for MIDI have changed (more often than not, you'll see USB MIDI instead of traditional MIDI DINs), but the underlying protocol hasn't changed at all.
Re: About time! (Score:2)
Yes, and my Yamaha keyboard uses USB midi with a non standard protocol. Thanks Yamaha, now I cant use it as a controller.
Re: (Score:3)
> I'm trying to think of something else that's deeply computer integrated and has remained largely unchanged in the past 20 years.
The closest (*) are probably:
* TCP/IP
* USB
(*) they have each had a few revisions, hence "closest".
Re: (Score:2)
> I'm trying to think of something else that's deeply computer integrated and has remained largely unchanged in the past 20 years.
RS-232 hasn't changed since 1960. Most modern computers don't offer an external port, but there are usually serial port pins on the motherboard.
You can also use an RS-232 to USB adapter to talk to a 60s-era teletype machine.
Re: (Score:2)
TCP/IP has changed MASSIVELY. IPv4 hasn't, but now we have IPv6.
USB has changed MASSIVELY. Enumeration is still the same, but everything else is different. 2 was highly similar to 1, but 3 is very different.
There is one thing that hasn't changed much:. VGA. Sure, it's rarer, but it's still around and while it's gotten faster, it hasn't changed in other ways in absolutely ages. Oh yeah, and RS-232. The only real change there is the introduction of ports that are 5v tolerant. A 10 volt swing between high and
Re: (Score:2)
Oh yeah, and RS-232. The only real change there is the introduction of ports that are 5v tolerant. A 10 volt swing between high and low doesn't meet the RS-232C spec formerly used by everyone.
I'm surprised at how much old serial stuff, that expects and produces +/- 12V, still happily talks to +/- 5V devices.
Mice, keyboards, ASCII/utf8. Anyone want midi cabl (Score:1)
> , what modern tech hasn't evolved in that timeframe? We've even done away with BIOS at this point, I'm trying to think of something else that's deeply computer integrated and has remained largely unchanged in the past 20 years
Mice, keyboards. I have Model M (1984) on my wishlist, as an upgrade from my current keyboard.
ASCII is still king - over 95% of web pages are ASCII. In the Unicode wars, UTF8 (which is ASCII, plus more) won because it's technically superior in a strict sense - it does everything b
Re: (Score:2)
Mice have moved from trackballs to optical sensors primarily, keyboards have adopted USB as the standard connectivity, ASCII has evolved into UTF8 as you implied, but even beyond that, ASCII was last updated to the current ANSI standards in '86 as I recall (so damned close).
I don't have much MIDI cable oriented these days, believe it or not, CV is coming back stronger than MIDI right now.
Re: (Score:1)
ASCII is still king - over 95% of web pages are ASCII.
Wrong. Even back in 2012, 60% of the web crawled by Google was Unicode. That figure would have only gone up since.
https://googleblog.blogspot.co... [blogspot.com]
Slashdot is in a tiny minority of ASCII-only websites.
Re:Mice, keyboards, ASCII/utf8. Anyone want midi c (Score:4, Insightful)
ASCII is still king - over 95% of web pages are ASCII.
Wrong. Even back in 2012, 60% of the web crawled by Google was Unicode. That figure would have only gone up since.
https://googleblog.blogspot.co... [blogspot.com]
Slashdot is in a tiny minority of ASCII-only websites.
You've confused "ascii" with "ansi" and confused "unicode" with "utf8".
That's pretty damn impressive considering you only wrote three sentences.
Re: (Score:2)
ATX chassis
GPIB
I2C and SPI
It goes with out saying, RS-232 and RS-422/285, which will still be in use 100 years from now
Re: (Score:2)
Why fix what isn't broken?
Re:About time! (Score:4, Funny)
Because you are Apple?
Re:About time! (Score:5, Insightful)
Where'd the idea come from that "if it's old, it must suck"?
If it's lasted this long, obviously it's working fine. Too many working techs get "updated" for no reason other than to satisfy the egos of the people who get to put a new bullet point on their resumes.
Re: (Score:2)
If anything, the more recent history taught us that the formula is rather "It's new, it must suck"
Re: (Score:2)
MIDI does kinda suck though, it's just that it's had various vendor specific extensions and hacks to keep it usable.
The low data rate and chaining is a problem when you have lots of MIDI devices. It becomes difficult if not impossible to synchronize them all. The stuff for describing instrument metadata and controlling instrument banks is all proprietary and not very good too.
I haven't read the new spec but it will be interesting to see if they can fix any of that stuff. If it's backwards compatible then th
Re: (Score:2)
I'm trying to think of something else that's deeply computer integrated and has remained largely unchanged in the past 20 years
Ummm ... the 3.5mm headphone jack? Oh, wait, that's being "upgraded", too.
Re: (Score:1)
Re: (Score:2)
" It's kind of impressive that the music industry is still entirely reliant on 35+ year old tech that hasn't changed"
They are innovative. They abandoned cat-gut strings years ago, ditto for ivory keyboards.
I guess you'd think that trumpets need batteries to annoy the neighbors?
You're wrong.
Re: (Score:2)
Hey, drummers are still hitting things with sticks.
Some things just seem to work : )
Re: (Score:2)
MIDI hasn't been updated in over 20 years, and even then it was incremental type updates. The spec itself hasn't really changed since...1983?
Because musical instruments, which MIDI was designed to control, don't change that quickly.
Re: (Score:2, Interesting)
Are you completely retarded? Next up, Mr. "On the Brain Transplant Waitlist" tells us that RS232 is decades old and is going to be replaced by this newfangled "Universal Serial Bus."
RS232 is still very common in industrial applications where you need a reliable, robust interface but not high bandwidth for control purposes. Certain low-latency and fault-tolerant Ethernet protocols are making headway but stuff like Modbus are not going away anytime soon.
Re: (Score:3)
OK I'm old (Score:3)
Re: (Score:2)
Anybody remember MIDI adapters for sound card joystick ports? I have one of those somewhere. Worked good.
Pretty sure I still have two or three in my basement.
Re: (Score:2)
Re: (Score:2)
Re: (Score:2)
Anybody remember MIDI adapters for sound card joystick ports?
I sure do. Back in high school my friend and I both played sax and toyed around with MIDI keyboards. He had a Mac LC, and had a hard time finding a MIDI adapter for it. I can remember that when I got my first MIDI compatible keyboard, the first thing I did was start shopping for a MIDI adapter.....only to find out that my IBM compatible already had it built into the Sound Blaster joystick port :-)
Re: OK I'm old (Score:2)
Oh yeah. Thanks for bringing back the memory. Once popped over a friends house and he was playing Privateer and used an actual keyboard with Midi as opposed to my Sound Blaster and Gravis Ultrasound. I was suitably impressed.
Limited physical space (Score:2)
I still don't get why they didn't just add a real MIDI port to the sound card.
Limited physical space on the card. You can't even fit MIDI connectors on an ISA slot.
(Even professional midi adapter back then used a flat connector between the ISA board and an external box that had the actual MIDI DIN connector.
Using an adapter on the gameport wouldn't feel that out of place)
Also, back when the first Sound Blaster was launched there were many more gamers likely to plug a cheap $20 joystick into their cards, than gamers owning an expensive MT-32 MIDI synth.
So it made much sense for Creati
Re: (Score:2)
Story of this port (Score:4, Interesting)
Pretty sure those were MIDI ports that we plugged joysticks into.
Back at the time of the IBM PC, these were dedicated game port, featuring 4 analog axis and for button (and thus usable for 2 players, with an analog stick and 2 button each, provided the correct Y splitter cable. Or much later more complicated 1 player peripherals, with the Gravis Gamepad being the first with such popularity).
It came on a dedicated separate expansion card.
IT way much later, once MIDI got added to the port that a few jostrick decided to use it for extra feature (I think Microsoft Sidewingers ?)
That’s why they were on your sound card... or was that a joke
The idea of packing MIDI + Game port together came from Creative.
Like I said before, the game port began as a dedicated expansion card under IBM.
This ate a whole expansion slot just for analog sticks.
So instead some manufacturer tried to make multi function expansions that packed together game ports with a few other functionnality (e.g.: my parents' 386 back hten had serial, parrellel, floppy and game all on a single expansion board, with a few extra headers to the ports that couldn't be held on the card's edge.)
Creative is the first one who had the idea of making cards for gamers, containing many functions on the same board.
The first one was Sound Blaster: an OPL2 FM synth popular with the then popular AdLib, a game port, a DAC to playback samples, and a MIDI interface (for external synths such as the then popular MT-32 from Roland).
Instead of needing 4 different expansion card eating basically every single available port, it's just 1 single card. A real space saver.
(or even more, when they also started putting CD-ROM controllers).
Given the limited amount of space on the side of the ISA slot, MIDI was routed through game port pins, and required one extra adapter cable.
That's the point at which a couple of manufacturer jumped in and decided to use MIDI as one possible way to expand the possibilities of joysticks beyond the 4 analog + 4 digital channel offered by the port. (While at the same time still retaining compatibility with games relying on the classic interface).
(The other strategies where using a completely different port : some complex joysticks used a serial connection, but this was at the cost of lost compatibility with older games.
Yet a different strategy was joystick communicating solely over the classic game port, but being able to switch into a proprietary protocole that send ditigal packets of data over the digital pins instead of straight 4 axis, 4 buttons - Logitech's ADI protocole is an example. This had the benefit of working with pre-MIDI game port, and still also be compatible with old games when the joystick isn't switched into ADI protocol mode).
Re: (Score:2)
Sidewinders never used MIDI. Instead, they did what everyone else did, and they used the second joystick inputs to add additional features. That gave two more button inputs plus two more axes, so even without doing anything tricky you could implement four buttons and four axes. But by using the four button signals to make a binary number, you could either send four-bit numbers synchronously, or three-bit values asynchronously. I believe both approaches were used, but I'm not 100% on that. Later sidewinders
Source (Score:2)
I managed to find the sources of where does my vague remembering of MIDI joysticks comes back from....
Sidewinders never used MIDI.
They did [github.com], the force feed-back is sent as MIDI messages.
Instead, they did what everyone else did, and they used the second joystick inputs to add additional features. That gave two more button inputs plus two more axes, so even without doing anything tricky you could implement four buttons and four axes.
...which would still limit you to 4 axises and 4 buttons. Going exclusively from the joystick to the PC.
Sidewingers rely on MIDI-out for force feed-back (sending information from the PC to the Joystick).
(Meanwhile, Logitech ADI protocol relied on rhythmically querying the port in some pseudo-morse-like patterns to trigger behaviors)
But by using the four button signals to make a binary number, you could either send four-bit numbers synchronously, or three-bit values asynchronously. I believe both approaches were used, but I'm not 100% on that.
Several joysticks us
Re: (Score:2)
...which would still limit you to 4 axises and 4 buttons. Going exclusively from the joystick to the PC.
Well, no. It wouldn't. You'd be limited to 4 real axes, but you could have more axes which were reported digitally. Not so good for directions, but fine for a throttle. It would be one-way, though, unless perhaps you had a way to toggle power for the second joystick connection. Thanks for the correction.
And yeah, I did leave out those who encoded digital signals on analog axes, like buttons with resistors on.
Just brand OSC as MIDI2 already... (Score:1)
https://en.wikipedia.org/wiki/Open_Sound_Control
Summary is in error. (Score:5, Informative)
Summary is incorrect when it says "MIDI was introduced at the 1983 NAMM show... The standard basically hasn't changed." The first half is true. The follow-on isn't.
MIDI received a major upgrade in 1991 in the form of General MIDI, which dealt with many of the things previously left open to interpretation by manufacturers, such as what order the instruments should go in the patch bank. That's the original MIDI 2.0, we just weren't quite as keen to use that particular notation (outside of software) yet. [wikipedia.org]
I don't object at all to extending MIDI, but I think they should have called it Global MIDI or Universal MIDI or Modern MIDI or something in keeping with the General MIDI naming tradition.
Re:Summary is in error. (Score:4, Funny)
Field marshal MIDI?
Re: (Score:3)
Field marshal MIDI?
Probably... whatever, they were both in NAMM together.
Re: (Score:2)
ROTFL that was a good one
Re: Summary is in error. (Score:1)
What about MIDIoE so much uses ethernet as layer 1 today why not MiDI, bonus no more specualiced cables needed just gram any old network cable, 1Gbps shuld be enugh for midi right? Aded bonus you could ad dante over the same port cutiing down the number of caples snaking across the stage, and possibly ( if the kepoed can stay within tight power constraints) you might ne eble to powerr it via PoE as well, hmm. Now that would cut down on the mess quit a bit insteed of 2 xlr cables 2 di boxes an unbalanced c
Re: (Score:1)
MIDI was already de-facto extended with Roland GS, Yamaha XG, and although I didn't notice a lot of it available on newer instruments... there was also GM2 (General MIDI 2). Newer generations probably don't remember (but it should be in Wikipedia by now) there were daughtercards/daughterboards for existing soundcards up to Turtle Beach Santa Cruz PCI, not just legacy ISA and PCI SoundBlaster series. WaveBlaster, Roland SoundCanvas, Yamaha DBXG, Ensoniq Soundscape, etc. all could be used by MIDI sequencer so
Re: (Score:2)
Super MIDI (power level over 9000)
MIDI That Goes to 11
Acronym abuse. (Score:2)
Ultima 6 Yamaha memories (Score:2)
..
So much skepticism (Score:5, Interesting)
So, the reason why MIDI has served the test of time is because it's a relatively simple protocol that does one(ish) thing and does it well - it provides data from musical instruments that is easily readable by other things which understand this very-well-defined spec, and allows things to be daisy-chained so that data can be manipulated and the data altered as a function of its sequence on the bus. It's very well understood, and that data can be manipulated however the devices wish to do so. Its simplicity is why it has stood the test of time, it is well-documented and as such is easily implemented by anyone, and I'm unaware of any royalties required to pay anyone for its use, making it possible to use in everything from open source projects to multi-thousand-dollar DJ controllers and keyboard workstations. Let's see what they're going to replace it with...
auto-configuration
There is no real configuration needed in most modern MIDI implementations, to my knowledge. The configuration is primarily on how the software interacts with it, and let's be real - that's the sort of thing the user should be doing anyway.
new DAW/web integrations
Ehm...this sounds like code for "a protocol that can allow for things like Mainstage to run in a web browser rather than on the device", to which I'd generally say, "stop making a web browser into an operating system". I really don't see the point of web integrations otherwise, and I don't see how "DAW integrations" is a problem to be solved. A DAW that doesn't support MIDI is like a word processor that doesn't support printing - fails at its core purpose and thus generally doesn't exist. If the plan is to be able to connect DAWs to each other, that's already a solved problem with Rewire...or existing MIDI files...or bouncing tracks...or running a DAW as a VST plugin. I fail to see the unsolved problem here.
extended resolution
I mean, I guess...but that's like saying we need a new standard for smaller MicroSD cards. Sure it's possible, but usability starts being counterproductive. Is there really that pressing a need for a higher-res MIDI protocol such that its human interface justifies it? Current MIDI doesn't seem to be a problem for DJs on controllers or have note limits that extend beyond what human hands can accomplish; anything much more than that I'm pretty sure is already solved in data-to-data sorts of ways that don't require MIDI.
increased expressiveness,
What even does this mean?
and tighter timing.
Again...*maybe*...but I'm open to scenarios where current MIDI timing is an actual-issue. It's like saying that RS-232 serial at 115,200 is too slow. It is if the intent is bulk data transfer, but it's plenty quick for its most common contemporary use case - typing commands into Cisco routers and similar appliances; it's far faster than I can type commands or read output and its simplicity means I am not limited to a particular solution.
Re: So much skepticism (Score:3)
Extended expressiveness is big nowadays. Go look at the Roli Seaboard and MPE. As for DAW integration, you donâ(TM)t know what you are talking about. Go look at NKS for a good implementation It will be good to see NKS become a standard as it is currently Native Instruments proprietary.
There are currently lots of ideas for extensions to MIDI around and being implemented by different companies. It is time to come together and sort things out a bit.
Re:So much skepticism (Score:5, Informative)
Timing is definitely an issue, even in the early 90's. the bit rate is so low that you can hardly play multiple instruments without noticing serious lag in rythm. which is why daisy chaining doesn't actually work and why your need all your equipment connected to a sequencer with lots of separate outputs.
Re:So much skepticism (Score:5, Informative)
More bandwidth is definitely a big thing - MIDI runs at around 31 Kbit/sec and it's fairly easy to swamp it, especially if you're chaining instruments on a single bus.
Increasing the resolution is Really Big Thing. MIDI is 7-bit, which means that if you do something like sweeping a cutoff filter, you only have 127 possible values which gives you very noticeable stepping artifacts (often called 'zipper noise'). Some manufacturers try to interpolate in software. Others bond two controller streams together so that you get 14-bit precision, or send custom NRPM values but since there are a number of incompatible ways of doing this, you have to have a controller keyboard which works the same way as your synthesizer. Setting out an actually standardised way of doing this would be really handy.
Re: (Score:2)
Auto config seemed to be aimed at just detecting and understanding devices connected via MIDI without having to manually configure them. Also helps with chains and multiple devices not conflicting.
The web stuff just seems like a way to produce DRM infested software that you have to subscribe to. It's too hard to make software running on the client expire or prevent piracy, so they want to move it all server side.
The timing stuff is because if you have lots of MIDI devices it becomes difficult to keep them i
Re: (Score:3)
It's obvious what kind of autoconfiguration is desirable. For example, having devices announce what type of device they are, how many of everything they have, names of samples or patches, etc. I'm not a musician and it's still obvious.
Web integration does seem stupid. Midi will be around long after the web.
Tighter timing and more bandwidth both have obvious utility (and are essentially the same thing.) Midi bandwidth is already a limiting factor, and long has been. And a finer time scale enables additional
Re: (Score:1)
I've struggled some with trying to generate MIDI files from code. The encodings are a PITA but I've had more problems with understanding why my files become malformed when I increase duration beyond a few measures. Feels like I obey the spec but I've missed something basic.
uhm.. (Score:2)
Re:uhm.. (Score:5, Informative)
Don't know why they still kling to the old DIN-port, USB(-c) was the way to go for MIDI 2.0, backwardcompatibility would be done through USB to DIN (which already exist and work perfectly).
Firstly, MIDI is opto-isolated. Without that, you get weird ground-loop effects like the data leaking into the audio, which happens quite a bit when using USB MIDI.
Secondly, MIDI is peer-to-peer whereas USB has a host and a guest. You cannot plug a USB MIDI keyboard into a USB sound module and expect it to work, you have to have a computer somewhere to act as a broker. With MIDI you can take two cables and link three machines together. USB1.1 doesn't work that way, and that's what the USB-DIN adapters are all using. USB-C might be better in that regard, I'm not sure.
You've also got a very large installed base (probably millions of machines) which are using DIN and USB1.1, switching to USB-C isn't going to. They bent over backwards to ensure that MIDI 2.0 is going to work with your $10000 OB-X with Kenton board.
Re: (Score:1)
Durability.
Have you ever watched a band set up, perform and take down their equipment?
USB cables would be lucky to survive for more than a show or two (and with some bands, even surviving one song would be remarkable).
The USB connector on the instruments would probably outlive a few cables, but would eventually fail as well. Do you honestly think all those starting bands who tour out of a van and sleep in rest stops because they can't afford a motel every night can afford multiple backup instruments to kee
Re:uhm..USB sucks (Score:1)
The spec isn't the problem (Score:3)
I code MIDI quite a bit. The spec could use some more modern updates, but at the same time it simply suffices because it is quite flexible. A much bigger problem is the lack of device specific documentation; most MIDI controller vendors fail to publish details on how to configure their products over MIDI and simply refer to OS-specific configuration software or don't even bother to do that and just rely on DAW vendors to fix it. That includes some big players like Korg, Novation and M-Audio (the latter being the company that's been selling a broken MIDI interface for years without even bothering to fix some major but easy to fix bugs).
Great news ... but why JSON ? (Score:1)
Great news that 2.0 is on the way !
But why on earth do they've chosen to go to JSON for some data interchange ?
It is not compact: takes lot of bandwith for little. There are better alternatives such as ASN.1 or even EXI.
It is no straight forward outside the web. You will need heavy marshaling (a good parser and generator). It will consume a lot of memory compared to alternatives such as FlatBuffers or ole Protocol Buffers.
It is neither easy to secure nor secured by default. In a close loop Din 5 Midi config
can't wait (Score:4, Funny)
for a new Atari ST with midi 2.0 ports!
DAW Song FIle Format (Score:2, Interesting)
Now if the DAW manufacturers could just come up with a similar standard for storing song data. Trying to share songs between DAWs is still a total pain in the ass involving stemming the audio, lots of time normalising MID then writing it off to new files etc.
Best of the bunch is Reaper as at least it's song file format uses plain text.
it really does sucks balls trying to collaborate on projects where you both have different DAW software.