Hacker News Reader

At this layer of note information, fact is the data will be either sent to a MIDI device or a plugin as a MIDI event, so you might as well keep it in the native format and not convert to a frequency.

I feel like we're only months/years away from software and file formats being obsolete for this for this sort of thing, and instead songs are composed by humming a tune to ChatGPT and have it add all the other instruments, some nice vocals, and output the MP3.

If you want to make changes, you give ChatGPT the MP3 file and say "switch out the trumpet for a piano and make the singer a guy".

As someone who makes music as a hobby that's about as enticing as outsourcing sex. I don't care if the machine can do it better.

That's an interesting point. I don't get much value from art personally (not that i don't see why people value it, i do!), so the idea of AI generated art isn't disgusting to me.

Music however.. i adore music. A big (but not essential) thing i love about music is the story that got the artist to that point. Pain, joy, emotion is transferred. Yea, often it's not essential so maybe AI Music has a place in my life, but easily 60% of my music is loved partly, if not heavily, because of the emotions that got the artist to that point.

I too am not interested in outsourcing this.

... music is art, no?

In English speaking world "art" can be used in two ways, as a synonym for "visual art" or to mean really any art (including music, cinema etc).

This is a hilarious take and juxtaposition with MP3 (lame encoders anyone?) and some far fetched gtp application. It would be a musak equivalent of the bored ape nft’s.

Like midi note transcription or waveform modeling will be cool machine learning tools. An end to end composition and mastering bot may make something passable like other current derivative music out there, being generous by saying passable.

Show me the leaps and bounds in self-driving cars that were only 3-5 years away 8 years ago.

There is still need for better daw, music transcription, and audio file formats in the next century. Lol humans want to make art in their free time not play around with pretend chatbot personalities.

P.s. people still record to TAPE and it sounds awesome even though there are albleton plugins

We might soon have good auto-generated music, but that won't mean that everyone else who was making music the regular way is going to stop.

You don't need to convert to frequency, you can use fractional MIDI pitch values. E.g. 60.5 would be a quarter tone above middle C.

In fact, the VST2 and VST3 SDKs support microtonal offsets per note event (measured in cent). Unfortunately, only few plugins and hosts support it.

I mean, you could, but that would get really gross when implementing certain scales that are nowhere similar to the MIDI tuning (like working out the fractions for a 13-note scale)

Eh, it's fine. With two decimal points of precision you're basically working in cents, which is a pretty standard way of representing tuning.

This doesn't need to be a trivially human-readable format, but I think maybe it would be nice to allow for multiple pitch representations: floating point frequency, floating point fractions of a semitone in any arbitrary EDO, fractional frequency as used in just intonation, etc..

You could also allow for annotations, like maybe a note pitch is represented as "63.86" in 12-EDO, but it's not just some weird in-between note, it's actually the E that's 5/4 above C, so the file could have an annotation that says that's what the note actually means in this context.

Isn't the issue that pitch bends in midi apply to every sound on that channel, so if you play a chord with a different temperament and send the bend data of each note the result is that the whole chord will be tempered as normal but with every note the same amount higher or lower?

I've heard the recent midi 2 can get around this but it's still a pain with the 40 years of midi equipment that surrounds us.

Yes, that's a problem with MIDI 1.0. There are a few ways to work around it. Some synths support custom tuning tables, and there's even a MIDI spec for custom tuning tables called MTS that hardly anyone actually implements as far as I know.

One way that works for multitimbral synth (which a lot of 90's romplers are) is to only play one note at a time on each channel, and use multiple channels for polyphony. This allows you to individually control the pitch of every note.

The one-note-per-channel trick works okay, but it's kind of awkward. If you want to use all 16 channels, it means setting every channel to the exact same patch, which is tedious. Also, you have to know what the pitch bend range of the synth is if you want to bend by the exact right amount. So, eventually MPE was adopted as an official MIDI standard to provide an easier, more user-friendly and standardized way to do one-note-per-channel MIDI.

MIDI 2.0 has added per-note pitch bend. For anyone making microtonal instruments right now, though, I'd say MPE is probably the best/easiest option.

Thanks, that sas a helpful response and I'll surely have a fun time looking into these different potential solutions.

MPE exists, and can handle microtuning of pitch. MIDI 2.0 supports per-note pitch bend. Old multitimbral MIDI synths can be used in an MPE-like fashion by using the one-note-per-channel and pitch bend trick. Modular synthesizers (and virtual modular synthesizers like VCVRack) use control voltage directly, which has no built-in assumptions about how or if the voltages are quantized.

Microtonal music, or even just plain regular traditional music that's tempered to be more in tune with itself, should be trivially easy to do with electronic instruments, and yet due to some historical decisions to base the one near-universal music protocol we've been using for the last 40 years on a piano-centric representation, it's a lot harder than it ought to be.

I think we should not disregard microtonal music just because it isn't cleanly backwards compatible with a protocol that's over 40 years old.

Idunno, half the instrument VSTs I have seem to support some form of microtuning or alternate tunings...

How do they achieve that?

I don't know what all the options are, but MTS-ESP is one. If I understand it right, it's a sort of out-of-band channel where you can apply tuning tables to all the MIDI synths that use the MTS-ESP API from one place.

I think it's not sent over MIDI so it only works if everything is running on the same computer.

https://oddsound.com/mtsespsuite.php

I don't know the mechanics, my guess is they translate the MIDI notes to the scale given to them by a tuning file

Here's a pretty comprehensive-looking list of software instruments supporting microtuning:

https://en.xen.wiki/w/List_of_microtonal_software_plugins

A DAW that utilises ODD-SOUND's MTS-ESP should be something all DAW's support. International music styles are suffering in DAW-land imo :(

https://oddsound.com/

This is built into MIDI 2, fwiw

Yeah, MIDI 2.0 has per-note pitch bend. They kept with 7 bits for note numbers, though, which is kind of inconvenient. If you have an instrument with more than 128 keys, you have to do a kind of dynamic allocation thing where you find the nearest unused key and bend it to pitch.

I think for most use cases, MPE is actually simpler. (Also it's probably supported by more instruments and synths at this point.)

All the more reason to include microtuning as a supported feature in this new format.

Not per note pitch bend. Note On/Off messages support an extra 16 bits of "attribute" data that can optionally be used as an unsigned 7.9 fixed point pitch offset in semitones. Note numbers are also expanded to 256 since they can use the full 8 bits of the note number.

MPE is a super limited hack, I doubt anyone is going to use it once MIDI 2 becomes available in synths (it's a future technology, fwiw - it will be a year or two before you can buy any controller that uses it)

> Not per note pitch bend. Note On/Off messages support an extra 16 bits of "attribute" data that can optionally be used as an unsigned 7.9 fixed point pitch offset in semitones.

> Note numbers are also expanded to 256 since they can use the full 8 bits of the note number.

You might be right on the attribute data, but I thought they had per-note pitch bend as well. I'm skeptical that 8th note number bit is available, but I'm going off of memory since the MIDI association has decided for whatever reason to require a login just to see the specs as if the MIDI specs are some kind of secret, and that section of their website is throwing SQL errors right now.

> MPE is a super limited hack, I doubt anyone is going to use it once MIDI 2 becomes available in synths

Maybe MIDI 2.0 will be adopted, but so far it seems to be getting very little traction, at least in the hardware synth/controller space. I'm less familiar with software synths; maybe it's getting picked up there.

MPE is kind of a gross hack, but it works pretty well and is supported by most of the expressive controllers out there and at least some synths. The only expressive controller I'm aware of that uses MIDI 2.0 is Lumatone. I think Roland also makes a regular keyboard controller with MIDI 2.0 support. Other than that the major music incumbents seem to be staying away, and the smaller expressive instrument manufacturers seem to mostly be sticking to MPE.

I'd be in favor of just ditching MIDI entirely, and use a different protocol that's more like what MPE would be if it didn't have to be mostly backwards-compatible with MIDI 1.0. I'd also be in favor of using CAN-bus instead of 31.25 kbps serial for anything that's not using USB.

Ah you're right, I missed in the message layout that the MSBit is reserved for note number. That's kind of pointless.

And it is a bit absurd that they're so intent on doing things behind closed doors. But the spec is very comprehensive and decently polished.

It's not really "out" yet in either hardware or software. The last update made some notable changes. Allegedly Korg is releasing a line of MIDI 2 controllers soon. I think NAMM in January is going to have a lot of MIDI 2.0 demos.

In the software world, some people (Steinberg) don't even want to support MIDI in synths at all. VST3 barely supports MIDI 1, and it will not support MIDI 2.0.