I'm no expert yet. But these are my thoughts around the subject.

I think first off, having one mode only, which is the performance mode, makes the instrument simpler and more intuitive.

If we compare a Voice of 16 elements to two parts of 8 elements. I think the two parts is more flexible with seperate effects and the ability to replace or remove a part of the sound.

I'm seriously considering buying a new keyboard (not necessarily Yamaha) but firs, I should know what Yamaha have really improved in MONTAGE. I hope I get an answer for these two questions. These are very important questions for me as I'm also considering to go completely software and it really must make sense to buy a keyboard like MONTAGE at this price range.

Since I don't have a MOTIF XF and had MOX (I already sold my MOX), I can do the following in MOX and MOXF for sure:

I can layer up to 4 VOICEs in MOX/F. In MOXF case, if the number of waveforms and memory is within the bounds, I can technically have a VOICE as detailed as MONTAGE's CFX (if memory and number of waveforms are within the range) expanded in two VOICES which I will layer the up.

I never tried the other idea but if one can route all the channels in let's say MIXING mode to one channel it might be possible to trigger all 16 VOICEs together (not sure and I haven't done a research about it).

Now I guess my questions are clearer:
I understand that a single performance mode definitely simplifies the workflow. I also understand all the other new features that has been introduced in MONTAGE. But I want learn about them one by one and my focus is on the one that I know the most the AWM2 synthesis part.

What I want to know are:
1- What really multi-part voice architecture offers which is not achievable with a greater than 8 element VOICE set up (16, 32 etc elements per VOICE) and in other words, why Yamaha is choosing mulit part VOICE architecture over a single VOICE architecture?
2- If we merely talk about AWM2 interface and synthesis engine in MOTIF XF, Is MONTAGE's multi-part interface a reface or re-write of the MOTIF XF's MIXING mode interface or not?

What I want to know are:
1- What really multi-part voice architecture offers which is not achievable with a greater than 8 element VOICE set up (16, 32 etc elements per VOICE) and in other words, why Yamaha is choosing mulit part VOICE architecture over a single VOICE architecture?

I think the answer is, the flexibility. By keeping the structure of a Part to match the Voice archeticture of the Motif XF allows for compatibility with the past three generations of the series. Often that architecture with its 8 Elements per normal program, that's 8 Oscillators, 8 Pitch EGs, 8 Filters, 8 Filter EGs, 8 Amplifiers and 8 Amplitude EGs, etc. is more than enough to create the sound. But in order to create this new multi dimensional engine, the Montage introduces additional kinds of programs:

_ A Single Part, single instrument Performance... Example: "Full Concert Grand" - an 8 Element acoustic piano found in Motif/MOXF
_ A Multi Part, single instrument Performance... Example: "CFX Concert" - an 18 Element acoustic piano found in Montage, Example: "Seattle Sections" - a 22 Element String Orchestra that morphs between large ensemble and solo instruments.
_ A Multi Part, multiple instrument Performance... Example: "Pearly Gates" - utilizes multiple arpeggios to create a backing atmosphere in which you can compose...

2- If we merely talk about AWM2 interface and synthesis engine in MOTIF XF, Is MONTAGE's multi-part interface a reface or re-write of the MOTIF XF's MIXING mode interface or not?

No, not so much... It builds on a strong legacy of the 15 years of Motif, but what is new is how you are able to access more nuance from the sound you are controlling.

This will be immediately apparent when you play and hear Montage. Having many more Elements available while performing means Montage has a far more flexible sound. I can bring in more different articulations, performance gestures when playing. And it's really about how smoothly you are able to do so that makes playing Montage a unique experience.

By allowing the entire 8 Element Structure to function as a basic building block, we can build much more nuance and change into the program. What's new about Montage is the "motion control" the way in which parameters can be automated and the sheer number of parameters that can be set in motion by a single gesture is massive -compared to the Motif. And we are not even considering the FM-X ENGINE in this discussion. Or that even external devices can be used to influence and modulate sounds in Montage.

There are some 251 parameter destinations in the controller matrix. At the Common level of editing, you have three rows of eight Quick Access parameters via the 8 rotary encoders, plus you have eight Assignable Knobs that are automatically linked to the Super Knob. Then you can control eight Parts, and each of these Parts has three rows of eight Quick Access parameters via the 8 rotary encoders plus each of the eight Parts has its own eight Assignable Knobs, which can, if you wish also be linked to a Super Knob control, or not.

This may not mean much on paper, but listening to this in action, will tell you what's different about Montage. There is more interaction within the sound... The new chip has some serious ability to manipulate massive amounts of data simultaneously - this immediately translates in the playing experience. At the different positions of the Super Knob you will find subtle or radical changes (and there are tons of Easter eggs (surprises) built in).

Thanks Phil. I appreciate the patient you have in answering my questions. I'm mostly looking for an insight. I'm looking forward to try MONTAGE and I've already heard some good quality demos. I guess Yamaha has a channel on soundcloud do you? It would be great if you post good quality performances there or any other place you believe it's a great place to have them.