Comment by dwringer | Hacker News Reader

Comment by dwringer | Hacker News Reader

Parent Full thread

Portamento is a fairly standard feature on many synthesizers - there's no need to use MIDI to spell out the full range of sliding one note into another, or to use a pitch bend. On supporting instruments (such as a great number of vintage and modern analog synthesizers, and perhaps a majority of software synthesizers/samplers) MIDI can send a control to enable portamento (and to set a desired speed) right before moving from one set of notes to another. After the slide is done, portamento can be disabled with another MIDI control signal. This can be used precisely for the effect of sliding a minor chord into a major chord, for example. Admittedly, this does require that the instrument be designed with portamento support in mind, which does seem like it could be an arbitrary limitation on instruments that may otherwise be capable of only simple pitch bends (though since these typically affect all notes at once I'm still not sure how trivial the portamento idea would be to implement in general).

But having the concept of per-note pitchbend built into MIDI itself would be a much easier way of achieving this.

Their point wasn't "sliding notes in a chord is 100% impossible, prove me wrong!", their point was that the only reason it's not trivial in the first place is because MIDI is biased towards piano keyboards, and that has had a huge influence on a lot of the music released in the past few decades.

I appreciate your response, though my point wasn't to "prove [the above poster] wrong", but that the ability to slide notes in chords is not exclusively dependent on a specific FL Studio plugin and has always traditionally been the domain of synthesizer builders rather than the MIDI standard. Perhaps the effect is more uncommon in music through the past few decades as a result of this, but it's not absent and dedicated musicians with enough resources were still able to make it work.

I do agree that it is a potentially limiting influence on the music most readily produced, but for that matter (IMO) so is the traditional adherence to 12-tone equal temperament that is often implicit in how the standard is implemented. It would be nice to have finer control here without having to resort to various hacks, but finer resolution means more bits required to represent a given frequency. Finding a different better way to encode MIDI data that could be used as a standard going into the future is something that I believe requires great consideration and might do well with a significantly different approach (but I don't presume to have given it enough thought to say one way or the other).

TL;DR: it doesn't quite work[1].

------------------

>Portamento is a fairly standard feature

On monophonic synthesizers.

There's no standard on how that would work with polyphony (that would be pretty much equivalent to polyphonic pitch bend). With your trick, you'd still need to specify which note goes to which new note with portamento/glide, and there is no standard for that.

The problem is usually worse with analog gear, where oscillators are commonly assigned to keys via a round-robin scheme. So enabling polyphonic portamento means that even if you play the same chord over and over, you'd hear the notes slide! Not quite what you intend.

So while some polyphonic synths have portamento/glide in poly mode, lack of support in the protocol means that it acts unpredictably: you only have control of where the notes are sliding to, as a set.

You can see the problem very clearly, using Behringer Deepmind 6 (6-voice analog with poly portamento) as an example here[1].

At this point, you break out your polyphony into voices, put each voice on its own track / MIDI channel, and pitch bend / portamento that (and record separately, if needed). And then you have the ROLI solution. Which is not compatible with any other MIDI controller.

As I said, MIDI is hackable enough to allow one to do pretty much everything, but it's going to be non-standard. And standards and defaults matter.

For that matter, that's why we have MPE[2], which is an extension of MIDI, but it's very, very new (2018-new).

[1]https://www.youtube.com/watch?v=o9cvMCHgB2s

[2]http://www.rogerlinndesign.com/mpe.html

Well for polyphony the "which note goes where" issue could, with creative sequencing, be solved by using different channels for each individual note to override the synthesizer's configured behavior [also enabling the use of simple pitch bends for the effect] - but I guess any good solution here will end up requiring even more hardware or software to allow this [it looks like this is essentially the goal of MPE] and, as you imply, arguably be a "nonstandard hack". But music IMHO is often about finding value and personal expression in the nonstandard along with the standard (perhaps I'm a little biased from too much jazz, because admittedly this isn't the case for all players in all styles). And musicians have certainly employed MIDI to use these techniques for decades.

Regardless of my possibly misguided attempts at defending a 36-year-old spec, I absolutely still agree with you, however, that an improved standard could certainly make having more creative control simpler and easier, requiring fewer "hacks".

Reply on news.ycombinator.com