When I moved from scene 1 to say scene 3, the SSS doesn’t seem to work. There seems to be no SSS between any scenes.

That is correct. You are getting the new terms mixed up. There are different terms to describe different functions.

SSS is Seamless Sound Switching and describes the ability to move from one KBD CTRL (maximum 8 Part) Performance to another without interruption in the sound or effects.
A Scene, on the other hand, is an entirely different function that has nothing to do with switching Performances. Parameters that are memorized in the Scene “snapshot” memory are mainly mixer settings and some synth Part offset parameters all within the current Performance.

Don’t mix up features, if you are using the Scenes to “Mute” and “unMute” Parts, that is quite a different thing from switching between Performances.
A MUTE is a connection between a Mixer channel and the audio Output. If you’ve ever worked with any kind of mixer, you’re familiar with a channel mute. This is entirely different from switching instrument programs. The MUTE simply turns Off the audio prevent that channel from being hear. When you unMUTE you are reconnecting to the audio Output.

There are many ways to switch between sounds without cutting off the sound... using MUTEs, however, is NOT one of them!

They include, XA CONTROL to change which Elements are sounding within a Part (sonically invisible), the Super Knob can be used to morph one sound into another, [PART SELECT] buttons allow seamless switching between sounding PARTs that are not linked by KBD CTRL.

The Scene buttons do not seamlessly switch sounds if your are using the MUTE feature. A Scene can recall a volume setting, a Pan position, a Send amount, offsets to the filter, the amplitude Envelope, swing quantize of Arps and MS, and MUTE status... there is nothing wrong, that’s simply not what it is designed to do. It does exactly what it is supposed to do. To think it does seamless switching between instruments is like discovering the windshield wiper switch in the car and wondering why it doesn’t roll up the windows. That’s not what it does. That’s not what it’s for...

What are you attempting to do? Maybe we can offer you a solution.

Hi Adrian, Bad Mister is of course correct. That’s just the way it is. However I also initially assumed / hoped that scenes would support seamless switching between sounds. Making it work is my #1 feature request for the Montage, and I think I am not alone. Without that scenes are much less useful than they could be.

For now I use the assignable buttons for that, it’s a bit tedious programming them and there are only two, but it works: you can select elements which only sound when the AF1 button is pressed, others which sound only when AF2 is pressed and elements which only sound when neither is pressed. When pressing the buttons already sounding elements are not muted, they continue until you let go of the keys and/or the sustain pedal.

Hope that helps!

That’s just the way it is. However I also initially assumed / hoped that scenes would support seamless switching between sounds. Making it work is my #1 feature request for the Montage, and I think I am not alone. Without that scenes are much less useful than they could be.

Making the windshield wiper your #1 request to open the windows is still just an empty request, no matter how many people discover it and conclude wrongly about how it works... it ain’t gonna happen.

Please, wake up. It’s like FN2187 thinking that in shutting down shield “we’ll use the force!” And Han has to tell him “it doesn’t work like that”. You learned the light switch turns the light on and off and you think it’s for running a bath, ... it’s simply a wrong guess, please, guess again. You’re simply, way off.

Try this... the MONTAGE combines a musical instrument and a digital mixer. Some functions you perform are musical. Some are technical.
You probably heard me use the analogy of wearing two hats... your musician hat and your technician hat.

Changing volume using velocity is a thoroughly musical skill - you have your musician hat on when manipulating velocity. Don’t let an engineer mess with your note-on velocities.
Changing volume using a Fader is a thoroughly technical skill -you have your recording engineer hat on when mixing... fade outs are technical, most musicians don’t fade out as they play... totally a technician invention.

Changing which Elements are sounding via XA CONTROL is a musical control - musician hat determines appropriate articulation. XA CONTROL is a function that with your musical skill you can, on demand, recall a specific musical articulation, be it a growl on a sax, or a pitch slide on a guitar.
Changing which Elements are sounding via MUTE buttons (linked to the Scene buttons) is a technical control - mix engineer decides which instruments are connected to audio Outputs.

If you the engineer wants to stop the drummer (since they don’t make musical decisions) they might disconnect it from the Outputs... that’s the mixer MUTE button. The musical solution for the same result would be to change the drummer’s chart (you’d put a rest on the page) the MUTE 4/4 Arp Type is the musical solution.

These functions have context, which you’re being naive if your not understanding them. Changing instruments by turning one off from the Output is equivalent of flipping a light switch OFF, same is true if unMUTE a channel by turning the audio On. There is no intelligence to this “switch” it is bound to be SLOPPY because the instrument might be sounding in both cases when you hit the button. You can switch instruments with Scene MUTE/unMUTE but only if you respect the playing preferences of where the “musician” placed their notes.

Of course, XA CONTROL is more work (you gett no sympathy here) to setup smooth transition between instruments. Where is your analog to the real world? In the real world how would you switch instruments seamlessly... no cutoff, so sounds overlap. Oh right, you’d get a second player... I recommend you do that if synthesizing the result is too much work (lol) activate the adjacent track so that the transition is seamless.

Otherwise learn to use the XA CONTROL which can be sonically invisible at the transition in that the sound switches to the new instrument only at the start of the next note pressed. So even if you touch the XA CONTROL button during a held note or chord, there will be no cutoff in sound.

Too much work (lol) try seamlessly transitioning a flute to an oboe in the real word.!
There are requests that are just a waste - this is one of them. But thanks for playing.

Practical Solutions (also in plain sight)
Please read about the sixteen [PART CONTROL] buttons, which you cannot outspeed... when you absolutely, positively have to switch between instruments as fast as humanly possible with absolutely no cutoff in sound... you can hold one while you switch between fifteen others all with no cutoff (it will dutifully hold that first throughout) ... read about it, try it - downloadable example, embrace it (and leave the windshield wiper switch for its intended role), thanks.

Performance Basics and the Live Set I and II - see the Performance example called “Favorites”. Let us know.

Thanks for your comments, Bad Mister. I like the analogy with the two hats. That's exactly my situation. The only difference is that my two hats totally agree, here ;).

As a musician I want to be able to switch between sounds. And I'd rather have it happen seamlessly. That's it. I think making an analogy between acoustic instruments and the Montage is pointless. After all I will never be able to play a real piano, some real strings and a real flute at the same time. But with the Montage and can do it with virtual instruments. Yamaha and other companies considered it useful to give me that capability. And with the live situation in my band it makes a lot of sense for me to do so. Also it makes a lot of sense to switch between sounds because there are different parts of a song where different sounds are required. Yamaha and other companies considered it useful to give me that capability as well. The scenes are from my understanding exactly meant for different parts of a song. Yamaha even went to the extent to implement SSS which is most probably not a trivial thing to do and actually requires a lot of resources. Yamaha actually put aside 8 parts for that very purpose, to allow us to switch between performances seamlessly. That's a big investment, because in most live play situations those 8 parts are basically completely useless, unless you switch between performances (or use the part control as you mention). So I think at least some people at Yamaha believe seamless switching inside a song makes sense. I agree with those people! I don't know the internal workings at Yamaha. But why you can say that my request will never happen is a mystery to me...

Note also that I could of course achieve the very thing I want with multiple keyboards. I have done so in the past. Could do it again, but I like to limit what I am hauling around. And the Montage allows me basically the reduce my much larger setup of the past just to the Montage. That's huge for me!!!

However I achieve seamless sounds switching is a different thing. There are multiple possibilities as you point out:

• Using the part control buttons is one. But that's a bad one for me, because it only allows to play one part at a time, if you select the part. But I want to let the let hand stuff continue and switch only the right hand parts. Also the placement of the buttons is not good for me. So that I never use. I know how it works. It just does not help me.
• SSS has some delay (which is by now thankfully drastically reduced against earlier version). But it is more effort, because I have to maintain two or more performances. If I change something in one performance (e.g. the left hand piano part) I have to apply those changes to the other performances. Also the placement of the buttons for switching is not ideal (for me). In addition I am using the iPad to select the performances through Songbook+. Up to now that had the disadvantage that I could not easily switch between performances of the same song, because the Montage did not give me any control about where the performances where stored. So they would not be adjacent and I could not easily switch between them. That problem is now gone thanks to Johan Melas Total Librarian. There is yet another problem: When switching via SSS to a different performance the sustain pedal is not taken up by the new sound. In some songs I now use SSS for switching, but only in a few because of the mentioned disadvantages.
• The XA controls are nicely placed (for me). They do what I want, but they are limited in some aspects and they are relatively tedious to program for what purpose. Also if I use them for switching sounds I cannot use them for the other things they are useful for: articulations. But I use this in many songs.
• Using the scene buttons. Ideal for me in all aspects except one: no seamless switching.

I think dividing things into what the musician does and what the engineer does is artificial. Yes, muting is what an engineer does. But when I am playing the keyboard I do a lot of things which the engineer could do if I would play multiple keyboards. Also I am not asking to use the mute for that. You are probably quite right that mute is not the right thing for that. But I am convinced it can be done. After all, what is the difference between using SSS to switch between performances to achieve this and switching between different scenes? It's just some internal programming changes in the Montage which are triggered by some user control. Sorry you are not convincing me (but you don't have to, of course - it is me trying to convince you or some product manager possibly reading the forum to pass it to engineering for reconsideration 🙂 ).

Now putting my engineer hat on: Yamaha went through all the trouble to implement the machinery which allows to let some sounds still continue while holding some keys, although at this point pressing the same key would not longer trigger the same sounds. That's something which requires a carefully laid out infrastructure. But it's done as evident from the fact that the XA control can do that, that the SSS can do that, that using the part control can do that, and that changing the transpose using the dedicated buttons can do that. The engineer in me now strongly believes that allowing to for example store the keyboard control with the scene should be not very difficult to achieve. Or something similar. I believe it is a small investment which gives some very nice improvement.

I’m going to guess that Badmister is telling us that the scene switches are outside of and downstream from the digital portion of the architecture where SSS can occur. Not sure - sounds like they are in the A section of D/A conversion that generates audible sounds. He can activate or change analog filters, etc but the SS occurs within the inner digital tone generation section of the architecture?

Beats me - but If correct - wishing scene buttons were internal to the digital SSS-capable tone generation architecture won’t make them move and programming firmware can’t make them move. I suspect this is why he’s adamant its using a wiper blade to turn off an alarm clock (ok, i changed the metaphor for fun).

I agree being able to use scene buttons for SSS is an obvious valuable tool for the musician who didn’t bring his studio tech to stand net to him at a gig. I use XA in my XF and MOXF to do this, really had hoped Montage would make this very obvious and widely usable feature easer. of all the crazy things Montage CAN do ... this is bread and butter for me too

The mechanism for SSS vs scenes is just different. One shuffles Performance memory (SSS) - all of it - including waveforms, arps, etc and one deals with restoring/replacing just parameters without touching waveform memory or arps or many "protected" (non-implemented for scenes) parameters.

Since scenes are a different thing entirely - there's not a "clean" way to wedge in SSS into the scenes. The use of "snapshot" implies it's a rip-and-replace type mechanism. SSS is a slide-and-shuffle mechanism - in order to keep your old sounds playing while new ones are "on deck" both to simultaneously play and to become the new front-and-center performance for the next SSS switch.

"Snapshot" also implies something that is combinatorial. Flipping a light switch - which has been mentioned - is a combinatorial event. SSS is more of a state-based process involving triggers. Something like pressing a "smart switch" ("soft switch") is not combinatorial because this does not directly open and close a circuit but rather tells a microprocessor about what the next state should be.

Having the functions in such disparate domains makes it difficult to support migrating functionality from one place to another. Not that it could not be done (no way, no how) - but that you would be completely changing the intended function. Therefore, scenes would become a heavy lift to re-purpose. Even still - it's iffy if this could even be done at all with the resources available.

For those wishing for SSS under scenes (heck, I'd like SSS to work better between performances - surely scene SSS would be welcome as well) - there are many ways to accomplish what you're after without having scene-SSS.

For me, I map out my charts very carefully. For each chart I have many options of how to tackle managing the multiple "sounds" (e-piano, piano, bass, synth, strings, etc) any given song needs. Although SSS is available to me (switching from one Performance to another) - I have not utilized this feature because I do not like the interface available to me to do the switching and also do not like the small delay which is incurred.

Here are some options:

1) Use different key regions on the piano keys such that two sounds which need to overlap ("SSS") can be played simultaneously
2) Use the same region of the keyboard so that sounds can layer - with the ability to mute one of the sounds. Sometimes strings are layered on top of a synth sound in an intro during the 2nd 8 bars to "build up" the sound. I'll use (un)mute to bring in the strings - and most of the time I have a rest at the end of the phrase (1st 8 bars) to make this sound seamless.
3) Use the same keyboard region but mute switch between two different sounds. Often times I'll have some form of a "solo" sound that is needed in a tune - say a different sounding synth sound. This synth sound does not play at the same time as a different synth sound used elsewhere in the song. Therefore, I'll map these so one scene plays one sound and another scene plays the other sound. There is often "miles" of rest between one synth sound needing to be in the front and the second synth sound to come out.
4) Use SSS and switch between performances. With the previous keyboard I had (MO6) - I would use this method to expand the amount of sounds. MO6 was not seamless, however, so I would plan these switches when sounds did not have to "carry over". To me, this option is best suited for switching between two songs of a medley where the 2nd song has a drum fill at the beginning giving some time to allow for the switch. There are other situations where this option would make sense.
5) Use of a 2nd keyboard and using PARTs 9-16 as a way to expand the sound palette. This would be if I started running out of resources in one Performance and didn't want to switch performances using SSS.
6) Use of XA to mimic SSS in a more limited way between sets of elements within the context of a single Performance.
7) Drastic - two (or more) Montages. Or Montage + any other keyboard or virtual instrument (computer)

The key, for me, to mapping out my sounds is to take a look at the rests within the tune and plan for my engineering work (poking at buttons that do not produce sounds, but change the programming of the keyboard) into those rests. If I'm switching things around during rests then the fact that the switch is or is not seamless becomes irrelevant. Certainly seamless scenes would give me another welcome tool in the engineering toolbox - but there are enough other ways that I have not been painted into a corner yet with no way out.

I, personally, have a different view on engineering vs. musician hat. For me, mashing the piano keys is the musical part. Anything else (mod wheel, pitch bend, ribbon, sliders, scene buttons, A.SW1/2, etc) is engineering. I do understand the lines BM is drawing - they make sense and are not wrong. It's just that, for example, telling a keyboard which articulation to use out of a short-list menu is not a musical thing to me. It's like the producer telling the musician "ok, now play this next passage staccato" in the earphones. Articulation as musical would be something velocity sensitive that I can "play" with mashing the keyboard keys and having to tie up my hand(s) with mashing buttons, to me, is a technical/engineering thing. I do not notate mashing down the A.SW1/2 buttons in musical notation - I use text descriptions written in something other than Italian (above the staff) to tell me to mash those buttons.

Just defining since my definition is different such that my use here of "engineering toolbox" makes sense.

David wrote:

I’m going to guess that Badmister is telling us that the scene switches are outside of and downstream from the digital portion of the architecture where SSS can occur. Not sure - sounds like they are in the A section of D/A conversion that generates audible sounds. He can activate or change analog filters, etc but the SS occurs within the inner digital tone generation section of the architecture?

Hi David, the Montage is a purely digital synthesizer. Basically everything is done in the digital domain, including filters, effects, tone generation. This is all a bunch of software which creates the resulting waveform data which is in the end converted to analog using the D/A converters. So the switching is certainly doable just by changing the software. The amount of work to do that is speculation, but my educated guess is that it is not that much work.

Chris wrote:I think what you need to do is program part levels up and down to the user knobs then send that to the super knob. The super knob position will control which parts sound or not and the scenes will set the position of the super knob.

Hi Chris, this is certainly doable and not too complicated with OS 2.0. However it does not solve the problem to switch seamlessly: When you level down one sound it will vanish immediately, that's what I want to avoid. The same can actually be achieved with scenes using the mute functionality.

Jason wrote:
The mechanism for SSS vs scenes is just different. One shuffles Performance memory (SSS) - all of it - including waveforms, arps, etc and one deals with restoring/replacing just parameters without touching waveform memory or arps or many "protected" (non-implemented for scenes) parameters.

Since scenes are a different thing entirely - there's not a "clean" way to wedge in SSS into the scenes.

Hi Jason, personally I would think allowing the scenes to store the status of the keyboard control switches should be clean and easy and would achieve what I would like to have. Please make the following experiment: Play a sound with 2 parts - preferably very different parts which have a very long sustain. Enable the keyboard control on the first part, but not on the second. Play something and hold the keys (or use the sustain pedal to hold them). Now disable keyboard control on the first part, enable it on the second. Play further notes. You will hear that the sounds of the first part continue. You will also hear that the sounds of the second will also sound, now. As soon as you let go of the keys holding the first part's sound they will stop sounding. But the others continue. Now you can switch on the keyboard control of the first part and switch off the keyboard control of the second part again and you have the opposite effect. This can be done not only with one part each, but with any combination of parts. So I can seamless switch between parts, proving that the Montage in general can do something SSS like between parts. So the only thing which is still missing is storing the state of the keyboard control buttons with the scene. And I am asking you as a software developer how difficult this can be...

The problem with keyboard control is dealing with the "virus" case of having a part selected which has keyboard control off.

Not only do you have to store the keyboard control state (which may be a "combinatorial" type setting) - but you have to manage the state of currently selected PART (which is not a "snapshot" thing).

Because if you have PART1 selected and you turn off keyboard control for PART1 expecting you're going to seamlessly transition away from PART1 to some other PART(s)>1, then the system will not play the second PART(s) like you want - but will continue to play PART1 since the rule is: if keyboard control=OFF for a given PART and that same PART is selected - then only that single selected PART will sound (and no other PARTs).

Therefore, keyboard control (due to what I have previously labeled as a "quirk") is a messy solution if the rules are stuck to with how keyboard control currently operates.

Neither of us have the firmware source. I'm presenting that I can see a case for "blowing up" Montage by throwing an "SSS" wrench at a glass "scene" house. Yamaha has better insight here and thus far have indicated this is a "cannot get past go" scenario. Given this, I would highly recommend managing the problem with the myriad solutions available and possibly "poll the audience" (post something) if a certain Performance is giving you (that's plural - anyone in the entire forum) difficulties. Responses/discussion will likely get past any roadblock without needing scene-level SSS.

thank you so much everyone for such detailed explanations which i will read shortly.

What I'm trying to achieve is quick PART access from a performance using my left hand with a short movement to the SCENE buttons.

Sorry, I did not mention the problem with "selecting a part". Manually I can still do this by pressing on the Performance button after switching. When the state of the keyboard control switch would be stored with a scene there would not have to be a selected part at all. Like when you freshly switch to a performance. So this is not a problem in general and this does not invalidate the experiment.

Agreed, I don't have the firmware sources (unfortunately 😉 ). But from the above experiment I still assume that it is not too much work. Sure, it could be that the Montage software is badly architected and thus a "glass house" as you write. But from what I see I think it is a very elegant solution in so many aspects that I would be very surprised if that was the case. I hope and think that the Montage firmware was designed by highly skilled engineers, and thus I assume that it is not a "glass house" and thus this request should be easily doable.

Yamaha has better insight here and thus far have indicated this is a "cannot get past go" scenario.

It was Bad Mister who indicated that it cannot be done and he did not say that he actually contacted engineering to propose that and got the feedback that it is not doable. Bad Mister is not an engineer, so his opinion about this is most probably not more valuable than mine (or yours) unless he asked.

Given this, I would highly recommend managing the problem with the myriad solutions available

This is exactly what I do and the list you have given pretty much contains the same things as the list I have given before. Read it again...

So far I have always found a solution which let's me achieve somehow what I want. But I have to remember for each of the > 100 songs which solution I have chosen in the end. Sometimes SSS, sometimes XA, sometimes scenes, sometimes something else. This is highly annoying. If it was possible I would always use the scenes and would not have to experiment so much to find a workaround and have to remember what workaround I have chosen in the specific case. So I can concentrate more on the musician part and have to think less like an engineer.

Hi

I tried (with my PC) to send an exclusive message to change the part parameter called 'keyboard control switch' = 'off' or 'on'.
F0 43 10 7F 1C 02 31 00 XX F7 with XX = 00 or 01
It works.
So i tried to change it on multiple parts. It works too.
By this way, when a performance is selected (title is highlighted, no part selected), i can recognize CC message from scene change and send sysex messages to change keyboard control for each part.
=> No need to store scenes with mutes, but scene continue to provide their functions.
=> I can sustain notes (with arpeggio...) and change active parts with 'SSS'
=> It should be nice if this idea could be implemented in scenes...

@+

Hi Tino, very cool experiment. I wanted to try something similar yesterday. But I first tried to record the according SysEx message with Cubase. But no SysEx was received, not sure why. So I (wrongly) assumed that there is no SysEx for controlling that. Very cool that you found it. If the Montage would send a SysEx when switching the keyboard control buttons, then this might be a solution for me, actually. I have a little self-built MIDI controller (something like an Arduino) which I could attach to the MIDI input and output. That way I could let it record the SysEx and associate it with the active scene. I would record which program was received last, which scene was received last and automatically store the last keyboard control state for that combination. That way it could still work without any UI control... 😉

I'm not sure if ARPs filter out SysEx. Maybe they do - but you can try converting this SysEx to an ARP (say XX=00 as ARP 1 and XX=01 as ARP2) and see if switching back and forth between ARPs will get you where you want indirectly. Scenes can change the ARP.

Source: http://www.moessieurs.com/scene_memory-base-montage.html

Hi

To Stefan:

The Montage sends sysex for scene changes and superKnob, if it does not have a controller message assigned in Utility> Settings> Midi I/O.

Like Jason explain in this thread: How to: send sysex commands to montage ,
you can send messages to get or change parameters in the Montage memory. this is how you can query or change parameters.

There is a dialog to get a parameter value, you have to send a request. Then the Montage responds with the data, if the query was good.
There is no dialog to only send a parameter. Just send it.

With your DIY Midi contrôler, you have to send queries for each part, you will get the values you need.
The difficulty is to recognize the performance we are working on (to store this data for THIS performance).
For that, I use a zone to send a Bank MSB + LSB + prog change different for each performance. (feasible for user perfs)
Then your DIY contrôler can automatically query what you need from the parameters for THIS performance.
It can also recognize scenes change with the midi control message sent by the Montage.
I use this process from my old Motif ES to display pdf files on my PC. I just have to change my 'sound' (can be a voice, a perf, a pattern, a song)
to display the page. With the Montage, it works also in live sets.

@+