Since the start of this issue - several workarounds have been listed including, in the mix, hardware-only solutions such as the MIDI Solutions gizmo (among others).

Some of these solutions (like the MIDI Solutions one) do not easily translate to dynamic configurations. In other words - if you want anything other than one re-route choice, it's not trivial to setup a new/different set of re-channel assignments.

Something like Karma For Motif (... and making this available for Montage/MODX) would require both a way to dynamically assign MIDI channels (through MIDI messages) and also not have to support many 3rd party solutions (meaning the keyboard having this built-in is a quasi-requirement for Karma). Karma doesn't describe all situations - but some may want Performance-to-Performance variation in the setup which would require something other than the MIDI Solutions box. An iPad would likely work with proper software. The check is in the mail for some of this - but there should be other solutions out there one could leverage as well.

I'm not assigning any "severity" to this lack of flexibility. All I can do is provide what I know are workarounds for various use cases when the subject bubbles up. Hopefully it helps some.

Some of these solutions (like the MIDI Solutions one) do not easily translate to dynamic configurations. In other words - if you want anything other than one re-route choice, it's not trivial to setup a new/different set of re-channel assignments.

Okay, but for 2-board gigging, I think just one setup should address pretty much anything you're likely to need (that one setup being MIDI Channel 1 rerouted to all of channels 9 through 16). I mean, the fundamental issue is that MODX wants your second controller to be able to transmit on multiple MIDI channels, and many "second boards" can only transmit on one at a time, right? So what we're doing with the MS box (or alternatively, an iOS app) is turning the single-channel second board into a multi-channel board.

So then you'd use the first 8 Parts of a Performance for the MODX's own keys (you can't assign more than 8 Parts to that anyway), and you'd use Parts 9 through 16 for whatever sounds you want played from the external keyboard on that particular Performance. (You could even "page over" to parts 9 through 16 on the MODX display, and dynamically change which parts of 9-16 are and aren't active, adjust their levels, etc., on the fly.)

Assuming I've got that right, I have a hard time imagining any gig scenario where your there's something you could do by (the unavailable option of) assigning Parts to alternate channels, that you cannot do by just putting the sounds you want on Parts 9 through 16. I mean, sure, you might do them differently, but is there a result you cannot achieve this way?

Answering my own question about whether there is "something you could do by (the unavailable option of) assigning Parts to alternate channels, that you cannot do by just putting the sounds you want on Parts 9 through 16" (assuming you have a MIDI Solutions box or IOS app routing everything from your channel-1 second controller to all of channels 9-16 on the MODX)...

As I understand it, as soon as you put ANY sound on parts/channels 9-16, you lose the Seamless Sound Switching, even if you haven't exceeded four Parts total.

Realistically, if you intended to use your MODX to play sounds from two keyboards at the same time with at least two parts being triggered externally (because there's already no issue if you only want to trigger one externally), there's a good chance you were going to exceed four parts total anyway. But still, this does point to a limitation of the universal 9-16 approach.

Anything else?

At the risk of beating a dead horse (it's been established already, I believe, that there is merit to MIDI channel flexibility):

Just because you are using an external keyboard does not mean you have to use different channels than the 1st 4.

One way to do this would be to use splits on a single PART instead of consuming more PARTs. Splits as defined by element note limits.

Using element note limits - you could map a piano on C-2 to D#0. That's 2+1/4 octaves below the lowest note on a 76-key synth. With the default octave setting, your local piano keys cannot reach these notes. However, you can have an external keyboard controller MIDI target these notes.

That's a specific example of a portion of the general idea - no need to use up PARTs which would invalidate SSS (if that's a goal).

Since elements are limited, you may have to handle layering with an external keyboard by using a 2nd PART with the same element split idea.

There are different ways to work around this - but that does not invalidate that there are plenty of reasonable/valid configurations which could leverage sharing MIDI channels to split - and also keeping the PART count low so that SSS works fine.

... still, one must work within reality and avail themselves to the workarounds until, if ever, options change.

At the risk of beating a dead horse (it's been established already, I believe, that there is merit to MIDI channel flexibility)

THAT is the information I am missing, those "established merits to MIDI channel/flexibility" if you will, specifically in a 2-board gig context. My question again, was, "I have a hard time imagining any gig scenario where your there's something you could do by (the unavailable option of) assigning Parts to alternate channels, that you cannot do by just putting {the sounds you want triggered from the external keyboard} on Parts 9 through 16 {and using a device to route a single channel controller to 9-16}. I mean, sure, you might do them differently, but is there a result you cannot achieve this way?" And the one thing I came up with is, "as soon as you put ANY sound on parts/channels 9-16, you lose the Seamless Sound Switching, even if you haven't exceeded four Parts total." (Though if you're playing 2+ parts from your external controller, odds are good that you'd exceed 4 parts total anyway.) Can you point me toward any other limitations of this approach (for 2-board gigging)?

(Also, assuming there are other limitations, I'm really interested in things that could reasonably be expected to be somewhat commonly considered useful, rather than weird oddball things that may be technically impossible but are also unlikely to be needed. IOW, I'm looking for examples of "real world" limitations, not theoretical impediments to things few if any people would ever need to do. Though I understand, that distinction may not always be obvious.)

The reasons I'm focussing on the 2-board gig scenario are (1) that's what I've seen discussed most often when this issue comes up, and (2) in a non-gig scenario, it is common to have access to a computer (which is the rationale for Yamaha moving other old Motif functionalities to software like Cubasis and SampleRobot), and once you have a computer in the mix, you may easily be able to solve the problem other ways.

Just because you are using an external keyboard does not mean you have to use different channels than the 1st 4.

One way to do this would be to use splits on a single PART instead of consuming more PARTs. Splits as defined by element note limits.

Using element note limits - you could map a piano on C-2 to D#0. That's 2+1/4 octaves below the lowest note on a 76-key synth. With the default octave setting, your local piano keys cannot reach these notes. However, you can have an external keyboard controller MIDI target these notes.

That's a specific example of a portion of the general idea - no need to use up PARTs which would invalidate SSS (if that's a goal).

Yes, split-by-element could be a useful thing to do, but "not using up parts that would invalidate SSS" is not actually the goal. The goal is to use a second keyboard to play whatever combination of parts you want (okay, up to 8), while playing up-to-8 parts from the internal keys as usual, without losing your other standard MODX functionalities... even if your second keyboard only transmits on one MIDI channel.

So, okay, you could say element-based splits gives me back my otherwise lost SSS functionality (from the 9-16 plan), but then you're losing a bunch of other stuff, so the goal still isn't met, it's just not-met in a different way. I mean, say you work around that SSS limitation by working at the element level to keep everything for both boards in parts 1-4, and then maybe you use the first 2 parts for your internally triggered sounds and parts 3-4 for your externally triggered sounds... you've got your SSS back, but can you really imagine anyone wanting to do that on a global basis? (i.e. you'd have you MIDI Solutions box programmed to re-route channel 1 to channels 3-4 for the entire gig.) I don't think this is a realistic fix to the problem of the SSS limitation in the "re-route to 9-16" approach. You'd no longer have access to all the same sounds and ability to mix-and-match all the factory supplied presets at will. You'd have to do a bunch of time-consuming configuration for every combination you might want. You no longer have access to 4-part sounds on the internal keys while having the external keys playing something else. Basically, while it's good to know you can do this kind of thing to solve a special case problem, I don't think it's realistic to look at that as a "standard" way of interacting with the board. It does not really eliminate the limitation I described, it just provides a somewhat complicated and limited workaround, trading off one limitation (loss of SSS) for numerous others.

So anyway, with the actual goal now fully spelled out, is there in fact any "unfortunate limitation" of the "re-route the external board to 9-16" approach for 2-gig live performance, other than the loss of SSS (which you'd be somewhat likely to run into anyway, since it's easy to exceed 4 parts when you're splitting and layering sounds across two boards regardless)? Because if, in fact, that's the only thing you lose, it seems like this could be a viable, compact, simple-to-implement, not-too-expensive solution, as long as you can live with that one limitation. And it's not even like you lose SSS completely, you only lose it on those Performances where you have sounds assigned to the external board (where again, you'd be somewhat likely to lose it anyway, since it is easy to go beyond 4 parts in that scenario, regardless of which combination of the 16 part locations you are using).

Scott, you are totally right that everything which the assignable channels solve could be solved by addition of hardware of some sort which can re-route input data to different channels (let's call it the processor). However at what price? And I am not talking about the money...

I am playing a hardware synthesizer because of the simplicity and reliability on stage. So I don't want a computer. But also some MIDI processor adds complexity. The box, 2 MIDI cables and probably a power supply. That's more stuff to take with you and to wire up. Stuff you could forget, stuff which breaks and so on. But that's the not the really bad part. Worse is the programming.

If the Montage / MODX had the flexible channel assignment then you could program all the routing, all the splits, all the layers in the MODX. Just add some e-piano which happens to be around, a MIDI cable and things will just work. With the MIDI processor, you have to do some programming in the MODX and some of it in the processor. And they have to be in sync. That adds complexity, effort, and more things which can go wrong. I have used such a setup for a long time and I had my share. Having only one thing to program is a major relief...
I currently use a 2 keyboard setup, the MODX 8 just as a dumb controller driving the Montage 6 which is in single mode. The MODX 8 keyboard is just a mirror of the Montage 6 keyboard which I use when I want the piano keyboard. With flexible channel assignment I would immediately start to have the MODX play only some parts, e.g. the piano and do other stuff on the Montage. I could get rid of quite a few splits... That would be very nice but I want to avoid the overhead of an additional processor (I actually do have one which I built myself and used in countless gigs).

And I believe (based on 35 years of software development) that the flexible channel mode should be easy to add to the MODX. The hardware can certainly do it (otherwise the external box would not help) and this should really be an easy filter at the place where the MIDI signals come it. They are doing it already because you can switch between single and multi MIDI mode. The step to add configurable channels should be really easy. And if I am right, then why the hell should people be burdened with extra cost, extra complexity, extra things which can go wrong? And why would the MODX and Montage be discounted by some many reviews and forum comments for the lack of flexibility which my synthesizers had 30 years ago???

I see your point about simplicity, but I think it's not as bad as you make it out to be.

some MIDI processor adds complexity. The box, 2 MIDI cables and probably a power supply. That's more stuff to take with you and to wire up.

Make that the box, and 1 MIDI cable. You'd need the 2nd MIDI cable anyway even if you were just making a direct board-to-board connection. And the box doesn't need a power supply as long as the MIDI device you're connecting provides 5v, which most do. The box is small enough to leave velcro'd to the keyboard, then it's not something else to remember to take.

But that's the not the really bad part. Worse is the programming.

If the Montage / MODX had the flexible channel assignment then you could program all the routing, all the splits, all the layers in the MODX. Just add some e-piano which happens to be around, a MIDI cable and things will just work. With the MIDI processor, you have to do some programming in the MODX and some of it in the processor.

The effort for the MODX programming (setting up which sounds you want playing from internally and externally) is identical either way. All you're adding is programming the box, which you'd only have to do once (to take channel 1 in, and send it to each of channels 9 through 16 out).

And I believe (based on 35 years of software development) that the flexible channel mode should be easy to add to the MODX. The hardware can certainly do it (otherwise the external box would not help) and this should really be an easy filter at the place where the MIDI signals come it. They are doing it already because you can switch between single and multi MIDI mode. The step to add configurable channels should be really easy.

Things always look easier to the people who don't have to do it. 😉 With all the clamor, and considering that Yamaha has come out with numerous Montage updates (with what appear to be some more elaborate enhancements than this), I suspect that if it were really easy, they'd have done it. If that's the case, we might feel better if some Yamaha engineer came here to explain why the architecture does not actually easily support doing this, and we'd all feel better, but nothing would change. I hope they're working on some way to do this too, if it isn't in fact architecturally impossible. But absent that, I think what I've described might be a very workable alternative, to generally do what people need, at pretty minimal effort or expense.

@Scott - I'm with you.

I've outlined some examples. Karma outlines some. Other users have outlined specific examples (there was a specific outline of using two keyboards in tandem - sometimes one or the other assuming master depending on the sound or performance). Like you, I've also encouraged specific examples to be "aired" to give concrete examples of how setting MIDI receive channels solve real-world problems that otherwise could not be done another way.

Eventually goals either get let-go or other avenues are found. I think, for the most part, players with these instruments are making due regardless of the gap between wishes and implementation.

It's easy enough to say that your car has no handle to open the door from the inside. You need to go out and buy a $0.75 hanger from the store so you can make a handle for your car to get out. Just because it's easy to do - this doesn't invalidate the request to build a handle into the door.

I am personally not in the pool of users who needs this function as I only carry one keyboard and am also not "into" arpeggios. At least not enough to miss Karma or a similar implementation.

Scott you are right, some MIDI boxes do not need the separate power. But that’s not necessarily the case for all of them, especially the more complex ones which have a display might need a separate power supply, because they exceed the power provided by MIDI.

You are also right that if you use the static routing of channel 1 to 9-16, then yes, there is no further programming on the box needed. I think I did not understand that proposal earlier. And the idea is kind of neat, I might use it with my MODX / Montage combination. I would not even need the box, I can just program some performance on the MODX to do that - the MIDI output part of the MODX is really powerful. So yes, this makes things easier. Still it seems like a wasted opportunity for me.

You write that things always sound easy for those who do not have to do them. That‘s usually my line 😀 . But of course I can use that line in reverse, here...

Using the MIDI processor with your solution sounds maybe much easier than it is in the end. Due to the static routing (which really does improve things!) some of the things which I like to use won‘t work. Like the volume faders for the different parts. I am talking about the faders on the Montage, because the MODX knobs and sliders are hidden from my view in my two keyboard setup. With the assignable channels I could use parts 1-8 freely how I need them and would be able to use the sliders to control their volume. Also when programming the performance constantly switching between the parts 1-8 and 9-16 is a hassle.

All in all workable but maybe more difficult for the ones which have to use them :D. Thanks a lot anyway for the idea!

It's easy enough to say that your car has no handle to open the door from the inside. You need to go out and buy a $0.75 hanger from the store so you can make a handle for your car to get out. Just because it's easy to do - this doesn't invalidate the request to build a handle into the door.

Love the analogy! And I certainly never meant to imply the request for Yamaha to address this was not legitimate. It's just that sometimes I've read where people refused to buy a Montage/MODX because of this limitation (or sold their Montage), and you know, if it's really great for everything else you need, maybe you should suck it up and just buy the $0.75 hanger, you know? 😉 But people do have to know that the hanger is available and can work. That's where I feel that some other responses I've come across have been less than fully helpful, basically saying, well, it's just not designed to do that, or you could buy a different second controller, etc. Let people know about the hanger (i.e. a relatively cheap and easy workaround), maybe they still buy/keep the car.

Actually, since Yamaha is into enhancing their keyboards with iOS apps anyway, it would be nice if they came out with their own app to facilitate this, free for Montage/MODX owners, to simply allow a single channel external board to trigger whatever combination of channels you want on your Montage/MODX, with Montage/MODX specific interface/instructions to make it as simple as possible.

As an update to my post above: Using the MODX without the router box does not cut it, unfortunately. The problem is the sustain pedal. I want the sustain pedal to affect parts played from either keyboard. That works, but would require too many zones on the MODX. One per part played from the Montage and one per part played from the MODX. Unless I am missing something...

Sustain cc64 is a Channel message

Stefan, I'm glad you found value/interest in my basic idea, which was aimed at the problem of not being able to play multiple Montage/MODX parts from an external keyboard (while playing other parts on the Montage/MODX's own keys) if your second board could transmit on only a single MIDI channel. Since your "second controller" is, itself, capable of transmitting on multiple channels, yes, the external MIDI Solutions box would not be needed. But I don't really understand the sustain pedal issue you're talking about. It sounds like you're trying to use a single pedal to sustain all the sounds being played from either controller? Which keyboard is the sustain pedal plugged into? Assuming each keyboard's triggered sounds respond to (or can be set to respond to) only the sustain pedal plugged into that keyboard (which is what I would want it to do, just as if they were two fully independent keyboards), you could do what people sometimes do when they want to sustain two separate keyboards together... get a pair of the square-style sustain pedals (one plugged into each board) and mount them side-by-side so you'd have the option of using one foot to press one, the other, or both pedals together. (Also, how many parts need to be sustained in total? If it's no more than 8, then using up zones shouldn't be an issue, but I'm not sure that's the best approach anyway.)

I have an FC7 and one sustain pedal. That's as many pedals as I can handle. In my setup I have the Montage 6 as the sole sound generator and use both the Montage 6 and the MODX as keyboards. What I want is that the sustain pedal is affecting the sounds played on both keyboards. I can then switch sustain on for those parts in a performance where I want it. I do the same when I play only with one keyboard: I have complex splits and layers and some of the parts are receiving the sustain, others are not.

In this setup I need to attach the sustain pedal to the MODX (which is just a controller) and send the sustain to the Montage. I need to put the Montage in Multi mode to be able to play multiple different parts from both the Montage and the MODX. Let's assume I control 8 parts from the Montage (e.g. TX816 e-piano) which should have sustain. So in order for the 8 TX parts to receive sustain I would need already 8 zones on the MODX which only send the sustain but no keys. That leaves nothing for playing sounds from the MODX keyboard, unfortunately. And since I want to change nothing on the MODX during a gig that does not work out.

To be very clear: I am very fine with the Montage and the MODX. I can absolutely make do with my current setup. I have now actually added a MIDI router which does what I need: It receives on channel 1 and forwards the sustain to channels 1-16 and and all the rest on channel 9-16. Everything works and I only have to program on the Montage. So your idea plays out! However, could the Montage receive channels be freely selected, the setup would be trivial: All parts to be played from the Montage would be on channel 1. All parts to be played from the MODX would be on channel 2. I would have one zone on the MODX for sending the sustain on channel 1. The second zone on the MODX would send all MIDI data on channel 2. Exactly the same effect but without additional hardware. And I would not have to statically assign parts to the MODX / Montage (1-8 for Montage, 9-16 for MODX playing) but could select them freely for each performance.