Are MPE Controllers a fad ?

[WatchTheGuitar]

But, isn't it clear that burning aviation fuel doesn't burn hot enough to remove the virus from the 5G towers that the grays and reptiles control from Area 51?

[Noumena]

Do you honestly think that we don’t know that you have no idea what you’re talking about? That you can just keep typing and that some magic will change things so that you are celebrated as a genius? Everyone knows, man. I feel sorry for you. If your heroes were members of this community they would be shaking their heads at the way you conduct yourself. Have some dignity and just stop. Times are tough all over, maybe it would be good for you to make a change.

Would you like some help configuring your rig so that you can do polyphonic expression? Mono is nice but poly can be nicer. There are people here that could help.

[pdxindy]

Now: MPE is new and it is hard to get working in a lot of cases, and even when it works it needs a lot of extra configuration and practice to make it work well. And it's only when it is working well that you can start to play and gain competency and feel, and it is then that what is really interesting starts to happen. I just spent four hours figuring out how to make a template in Bitwig that would allow me to quickly wire up instrument racks of mono Kontakt instances with pressure and timbre controlling dynamics and mic x-fades in Spitfire Audio sample libraries. I spent two hours working out how to smooth the pressure control releases using DC offset devices, envelope followers and math. And now it works remarkably well. It plays like an instrument would, like my viola does. It makes me want to learn to play better, to practice, it has shown me that over days and months my work slowly results in my being able to play better and more interesting music. I can't buy these sounds or results -- I have to really work for them. I can't demo them, I have to develop them myself. Its really exciting and deep and it feels like both the future and what Vangelis was working on figuring out his whole life. You'd better believe that HE would have all this figured out to the n-th degree if we had these tools in 1982. I can IMPROVISE string quartets, with the flowing dynamics.... fade up a chord one note at a time from silence in real time. This is all he ever wanted and so much more.

I would like to find myself in community with people that are working on the same thing: people smarter than me, people with different goals than mine... others that I can share understanding with so that I'm not just working on my own. There seem to be precious few that are doing this... and my feeling is that they are discouraged by the kind of argumentative thread trolling that is constant when discussing this tech here. Let's keep talking and working despite all that.

There are many uses for MPE... my ways of using it slowly keep expanding and I steadily improve in my ability to make good use of them. It takes practice.

One cool new MPE possibility is the updated arpeggiator in Bitwig 3.2. It responds to per note pressure and uses that pressure level to control note velocities as the notes play. So I can hold a chord and as the arp plays, hold one of the chord notes harder and those arp notes play louder (or whatever you want velocity to control). One can create various accents that way, slowly change the pressure from note to held note and the arp evolves. It gives beautiful results!

I was exploring controlling the velocity curve in the Grid so that notes only sounded above a certain pressure so I could drop notes out entirely by keeping a light pressure. All this makes for an arpeggiator more realtime expressive than was previously possible.

With the new Instrument Selector modes in 3.2, I can use a note range to switch layers while playing. If one holds a note, then switches to a different layer (sound), the held note still stays on the previous sound and per note pressure, pitch and timbre still work as long as you hold the note. If you have an Arp on one of the layers and hold some notes, then switch layers, the arp keeps playing and the arpeggiated notes still respond the MPE input... even while you can be playing a different layer with the other hand and switch that layer. This opens up so many expressive possibilities using Bitwig and an MPE controller (Linnstrument in my case and waiting on an Osmose).

[Echoes in the Attic]

There are many uses for MPE... my ways of using it slowly keep expanding and I steadily improve in my ability to make good use of them. It takes practice.

One cool new MPE possibility is the updated arpeggiator in Bitwig 3.2. It responds to per note pressure and uses that pressure level to control note velocities as the notes play. So I can hold a chord and as the arp plays, hold one of the chord notes harder and those arp notes play louder (or whatever you want velocity to control). One can create various accents that way, slowly change the pressure from note to held note and the arp evolves. It gives beautiful results!

I was exploring controlling the velocity curve in the Grid so that notes only sounded above a certain pressure so I could drop notes out entirely by keeping a light pressure. All this makes for an arpeggiator more realtime expressive than was previously possible.

With the new Instrument Selector modes in 3.2, I can use a note range to switch layers while playing. If one holds a note, then switches to a different layer (sound), the held note still stays on the previous sound and per note pressure, pitch and timbre still work as long as you hold the note. If you have an Arp on one of the layers and hold some notes, then switch layers, the arp keeps playing and the arpeggiated notes still respond the MPE input... even while you can be playing a different layer with the other hand and switch that layer. This opens up so many expressive possibilities using Bitwig and an MPE controller (Linnstrument in my case and waiting on an Osmose).

This is not useful. And if you insist that it is, then I will call you close-minded for not being open to how it is not useful to me.

But seriously, yeah there are tons of ways to use per note expression in ways that aren't even just polyphonic playing. The example that I often give is a monophonic MPE sound where you are gliding one note while another (or several) plays on and off quickly, so that you just hear the one note pitch gliding in between the notes that are cutting on and off. This kind of thing would be very difficult to do by regular means, the automation would be a nightmare. And certainly impossible live. The funny thing is that many of these playing techniques aren't even imagined by some of the leading MPE hardware or software developers. I've argued with Roger Linn and U-He both about the usefulness of MPE for mono sounds (which is something that Roli clearly get based on their MPE preset patches). There are clearly cases where separate tracking of expression per note is useful where you only want to hear one note at a time (an arpeggiator being another), but it's still so new that people aren't seeing all these possibilities yet.

[pdxindy]

But seriously, yeah there are tons of ways to use per note expression in ways that aren't even just polyphonic playing. The example that I often give is a monophonic MPE sound where you are gliding one note while another (or several) plays on and off quickly, so that you just hear the one note pitch gliding in between the notes that are cutting on and off. This kind of thing would be very difficult to do by regular means, the automation would be a nightmare. And certainly impossible live. The funny thing is that many of these playing techniques aren't even imagined by some of the leading MPE hardware or software developers. I've argued with Roger Linn and U-He both about the usefulness of MPE for mono sounds (which is something that Roli clearly get based on their MPE preset patches). There are clearly cases where separate tracking of expression per note is useful where you only want to hear one note at a time (an arpeggiator being another), but it's still so new that people aren't seeing all these possibilities yet.

Keep arguing with them

Then there is 'sorta' monophonic playing... but where notes have a bit more release and there is a subtle amount of overlap. Bend the pitch with a regular synth and the bit of tail of the still sounding note also bends... no problem with MPE! That tiny bit of overlap makes the character of the sound quite different.

I also like doing things like playing a chord, but starting one of the notes a bit flat (turn off quantization). I've had to work on that technique as it requires some skill but sounds great when done well.

Sliding a chord up a semitone is different with MPE. Sliding a chord up a semitone with a regular pitchwheel, all the notes are perfectly aligned. Using an MPE slide, there are subtle differences from note to note that just sounds better to my ear.

We are just at the beginning of MPE development.

[excuse me please]

It's Vincent Clarke, not Clark, dummies.

[Echoes in the Attic]

Then there is 'sorta' monophonic playing... but where notes have a bit more release and there is a subtle amount of overlap. Bend the pitch with a regular synth and the bit of tail of the still sounding note also bends... no problem with MPE! That tiny bit of overlap makes the character of the sound quite different.

Yes this exact scenario is one of the major reasons I was annoyed for so long about how Bitwig used to rearrange the midi channels on playback if the notes don't overlap and reuse the same channel too quickly. The same thing still happens with the touch keyboard in Bitwig and it's very annoying since it bends the tails of previous notes that are still heard. But at least hardware plays back correctly now. The weird thing is that some software doesn't care if the same channel is used again for a note before the previous one is finished. It still separates the expressions. I don't know how but that's how the Roli synths work for example, and Bitwigs own synths. So there obviously are lots of differences between how devs implement MPE in software. Especially in how they play mono sounds, if even possible (*cough* u-he).

[Noumena]

I’m very much looking forward to VCV rack v2 to assist with a lot of these things, too, and hopeful that it will have proper MPE out as well. MPE control into a proper modular env where you can do most anything... I’m also hopeful that Bitwig will implement a notefx-grid and a modulator-grid soon as well.

[BONES]

The reasons why polyAT were not in general use for a long time have little to do with precent or desire. Synths were really expensive and didn't have the power to process the density of MIDI data that polyAT generated for quite a while.

That may have been true when the CS-80 was released but for the last 20 years or more, processing large amounts of 8 bit data has been a doddle. Despite that, still no PolyAT.

I'm grateful that some manufacturers are pushing things forward with more expressive hardware and really excited about what is happening in software, with Bitwig, especially. Now the market is more oriented towards people that can't play keyboards well at all -- that are not performers... but that doesn't mean that there are not a lot of people with a lot of art to make that are not looking for more expressivity in performance. Just because one things exists doesn't mean that others do not.

You being personally interested doesn't mean that it won't be just another in a long line of fads.

Yeah I think people who point to poyAT not having been that huge in many consumer products as a predictor of MPE popularity are missing several things. As you point out they were more expensive so many keyboard makers left them out.

That hasn't changed, have you seen the price of a Haken Continuum or a Linnstrument or a Roli Seaboard? My tiny, little Seaboard Block cost more than all but one of my hardware synths (Analog Keys).

Plus aftertouch on regular keys is quite different to the instant pressure type MPE controllers.

This has nothing to do with MPE, it is about the extra "dimensions" of touch that you can take full advantage of without MPE, using a standard MIDI set-up. I don't have to have MPE enabled on my Seaboard to take full advantage of all 5 dimensions of touch it offers. And I don't have it on because it creates too many hassles with my other hardware.

Aftertouch on keyboards was a very different effect in that you press into the bottom of the key after you've played a note.

As I do with my Seaboard, too. Yes, you can use it differently but it also works just like it does on my KeyStep, Ultranova and Analog Keys.

It was never an obvious effect, you had to know it was there and care about adding some kind of expression after you've played the initial sound.

The same could be said of velocity or even a mod wheel but you make the effort to learn, don't you? Ultranova is the first synth I've ever owned with aftertouch, and I only bough tit 8 years or so ago. It was hardly a big adjustment.

My first exposure to that style of expression was actual the Keith Mcmillan QuNexus where I discovered how great it was to play pads expressively by pressing into the sound as opposed to an envelope giving it a soft attack. I think I was playing around with imposcar where polyAT was part of a preset when that sort of sunk in.

You never used a mod wheel or joystick?

But once MPE came along, it took it to a much different level, primarily in adding the per note pitch control, which is vastly more useful than a pitch wheel.

No, it didn't. Again, what took it to another level were the extra dimensions of touch. The ability to run your finger along a key instead of reaching for the mod wheel or pitch bend.

Natural polyphonic instruments do not bend each simultaneous note exactly the same ever.

Which instruments would that be? Piano? The pedals affect all notes equally. Guitar? The Whammy Bar works on all the strings equally. Yes, you can bend a string with your finger but, as I've explained too many times now, you can set up a keyboard split easily enough to do the same with a synth.

Subtle differences in vibrato has a huge effect, Same for ensembles of mono instruments like orchestral strings and brass, which synthesis in some ways emulates.

Those subtle differences are in the phase/frequency of the vibrato, not in the amplitude, so MPE is going to recreate that.

Not to mention of course larger pitch slides of strings or instruments independently. So along with the pitch per note, and adding the up/down expression, plus control surfaces that encourage you to press into the surface expressively or strike hard for all kinds of combinations of articulations, now we have something much different than poly aftertouch, and the benefits of having different expressions per note become much more clear to (most) people than it was with a simple expression that happens after a note has initially played.

I think you are confusing what is technically possible with what might be musically useful.

So no, you can't predict the success of MPE by the past success or relevance of poly aftertouch and it is certainly not a fad.

Of course it's a fad. There hasn't been a new MPE controller since the Roli Seaboard in 2016. If it wasn't for the prototype of that one that was getting people excited last year, I'd say it was dead and buried already. You have to remember that the Continuum has been available for 21 years now - I don't know about you but I have never seen one. The Linnstrument has been around for going on six years now and I've never seen one of those, either. And Roli seem to be in all sorts of financial trouble at the moment, too, so who knows how long they can survive. None of those things point to any kind of long term future.

When MPE started getting introduced to VSTs a couple years ago

Not a "couple of years ago", 21 years ago. Even Roli's Seaboard has been around since late 2015. If it was going to be successful, there would be mainstream variations on the theme from Roland, Korg and Yamaha by now, for sure. But it remains a rarity, an expensive, out-there curiosity only made by a couple of niche manufacturers. I will be extremely surprised if anyone is making them a decade from now, which is a real shame because I love my Seaboard Block.

[Echoes in the Attic]

^ Most of these individual responses are just nitpicking arguments and really don't require a rebuttal, and I think it's obvious to anyone reading that you have your view of 5D and MPE and it's pretty mixed up so I'll leave that alone at this point. But with respect to whether it's a fad, which could be a reasonable theory, I think the mistake in this thinking is to see the current controllers as the end of the line and not imagine that any more will be made. The success of MPE in general will not depend solely on the success of Roli or the linnstrument, they will keep making them, new ones with slight variations like Osmose will continue to be developed and different designs will appeal to different people. Some who don't like squishiness of seaboards may love the regular key feeling of the Osmose while still getting the vibrato and instant pressure on the keys, which yes, is obviously extremely different from aftertouch. In fact Osmose has both instant pressure and aftertouch at the end of the key depth, which is something a lot of people will appreciate. Who knows, maybe MPE+ will take off! If you think aftertouch (which begins at the end of the key depth) and instant pressure (which begins immediately on contact) are the same thing, then you must not have much experience playing with them. Aftertouch, beginning at the end of the key depth is nowhere near as expressive. An absolute novice can feel the difference immediately. So it would be foolish to try to judge MPE proliferation by polyphonic aftertouch adoption in the past. What makes these new and interesting is not just because there are extra dimensions of touch. Regular midi could always accommodate 3 dimensions (velocity, aftertouch, release velocity). It's not two extra dimensions that was a game changer. It's the fact that pitch (and later Y-axis) became possible per key, along with those 3 because they decided on a new way to split up notes by channel. So now that all of this expression was possible per key, it is a much bigger additional benefit than just polyAT, or release velocity, as these on their own were incremental and somewhat obscure. But once pitch became per note, along with all the other expressions per note, now the benefits become much more obvious. And it becomes worth actually having new note input devices in interesting and different form factors, like funny squishy keywaves or grids like the linnstrument, to take advantage of all that power. Could they have designed something to allow pitch bend from a struck note instead of from a pitch wheel without MPE? Sure, but without individual expression per note it would turn into a complete mess when you bend from multiple notes at a time or play fast, as there would be conflicts. And actually, the Continuum was one example that already had multiple dimensions of touch and allowed pitch bend from where you struck the note. However it was it's own complete system and hardware synth, not made to work with other products. There was no standard protocol when it came out to help separate software and hardware companies make things that could work together. This is where MPE helped make it a less obscure concept and encouraged a market for these products.

MPE type controllers will continue to be designed. And the prices will continue to fall. Does anyone remember the cost of the initial batch of seaboards. before the Rises? They were up to $10,000! And this is before Y axis modulation even. These days many of the newer MPE capable controllers are often taking the form of smaller portable controllers like the Sensel Morph or seaboard blocks. They will never be as cheap as a simple basic keyboard with only velocity of course. More features mean more complex technology. The last couple years has seen a drastic increase in instrument compatibility though (and when I said a couple years I meant when many more instruments started adding MPE compatibility, not when MPE started, obviously), both VSTs and hardware. I see no reason why the current trend wouldn't continue. It is similar to other areas of technology. You can not predict success based on the usability/success of the current offerings only. That means not taking into account the products that haven't been made yet.

[pdxindy]

Of course it's a fad. There hasn't been a new MPE controller since the Roli Seaboard in 2016. If it wasn't for the prototype of that one that was getting people excited last year, I'd say it was dead and buried already.

The Osmose is nearing production. The initial pre-order run sold out in days and they made a second run due to high demand. The pandemic delayed it, however I am scheduled to receive mine in the next 3-4 months.

MPE is not a fad... it is here to stay because there are people who want what it offers and are willing to pay for it. I love my Linnstrument. I've had it for some years now and its value and use to me has not diminished. To the contrary, as my skill with it has improved it has become even more useful.

[pdxindy]

The last couple years has seen a drastic increase in instrument compatibility (and when I said a couple years I meant when many instruments started adding MPE, not when it started, obviously), but VSTs and hardware. I see no reason why the current trend wouldn't continue. It is similar to other areas of technology. You can not predict success based on the usability/success of the current offerings only. That means not taking into account the products that haven't been made yet.

Hydrasynth has PolyAT, so does Arturia's Microfreak. Behringer announced a PolyAT keyboard (CS-80 clone). A number of iOS apps support MPE with touchscreen keyboards. Various smaller and/or less expensive controllers support PolyAT or MPE. Haken released the Continuum Mini and has also redesigned the Continuum. And of course Osmose. There is plenty of new stuff happening.

There are a growing number of hardware synths that support MPE. The Black Corporation synths do (Xerxes, Kijimi, Deckards Dream). UDO Super 6 has solid MPE support. It was delayed by the pandemic, but it should be shipping soon. There are others...

There are dozens of MPE capable softsynths now.

If someone wants to argue that MPE will stay somewhat niche, I wouldn't disagree... but fad? no, it aint going anywhere cause there are a solid core of musicians who want what it offers. And there will be a new musicians coming up who take the tools and develop their skills and use them. I'd wanted a MPE controller for many years... before I ever heard of MPE, I'd imagined it. Now that I've had my Linnstrument for some years, my interest and use of it has not faded in the slightest.

[Echoes in the Attic]

Indeed. Good lists for Desktop, iOS and hardware here:
https://www.rogerlinndesign.com/support ... ded-sounds

[Noumena]

I would like to develop an MPE focused MIDIfx VST plugin to provide some utility functions that I feel are needed. I’ve been doing a lot of this work already in Bitwig with modulators, but there are some functions that simply cannot be implemented in this way. I feel like we might get some of what we need when we get the MIDI Grid… and hopefully the Modulator-Grid and Notefx-Grid devices in BW, but a more general solution, implemented outside of that environment feels useful.

Blue Cat’s Plug n’ Script seems like it might be a useful platform for this work. You can write MIDI VST plugins that can be exported as standalone and shared with people that don’t have the development software, which is itself only $99. It uses AngleScript, which is close enough to Javascript that I think I can code with it. I would like feedback from this board as to better alternatives.

Here’s what I wish to implement:

This plugin would be placed between the MPE controller and the MPE synth, It is a mediator — takes an MPE stream in and sends an MPE stream out. This allows it to modify the MPE control stream in various ways. The plugin would be grouped with the synth plugin when saved… allowing these configurations to be part of the patch or snapshot. This would allow the user to improve the MPE implementation of synths (which have inconsistent and often shallow MPE implementations.)

Attack and Decay lag processors for pressure. As a ROLI user this is my most desired function. I would like to have the ability to dial in the right lag for the pressure control, which is jittery. I need smoother operation and the various plugins that support MPE generally don’t have this. Audiodamage ones do and it works really well.

The ability to implement a MIDI-based envelope for release. I would like to be able to do the following:

press note -> note on event
modulate pressure -> pressure cc messages flow
release key -> process a release where pressure cc falls according to an envelope pattern AND THEN, when it hits zero issues the note-off
random modulator of parameters. Taking inspiration from the Random modulator in Massive, I would like to be able to generate a random value 0-127 on every note-on event that can be used to modulate pitch, velocity, pressure, etc. For instance: I have configured random to modulate pitch by 2%. Each time I press a key the pitch of that note (modulated via MPE pitchbend) is altered by the random value generated on key-down * .2 * actual pitch.

host automation -> value mod. The plugin will have several host automation targets (knobs) that can be modulated by the host (in Bitwig via modulators) or by external controls. These will be in a mod-matrix internal to the plugin and would be able to modulate the MPE controller valunes. Use case: external LFO -> MPE MIDI plugin -> internally this host automation control is mapped to modify pressure, making a tremolo effect.

What do you folks think? Feasible? I’ve not worked with midi fx plugins yet and wonder.

[WatchTheGuitar]

What do you folks think? Feasible? I’ve not worked with midi fx plugins yet and wonder.

If you're just starting with this and have some C/C++/C# programming experience you may also want to consider JUCE as you can use the personal edition for free with (the also free) Visual Studio Community Edition while you get to grips with it.
https://juce.com/