It's interesting why they bumped up AWM2 PART Elements from 8 to 128. I would have thought from 8 to 16 or 32 at the most..!?

I think the way the 128 elements per PART works is that you can have 128 on each PART, but if you set all 128 Elements to Normal XA Control, layering all the Elements, and use them all at once, you will likely be down to 1 or 2 note polyphony of a single PART. Then you will really experience polyphony issues when you do the same to a second AWM2 PART (unless you use all Preset waveforms on the first PART and User waveforms on the second PART). Then you would have a mass 1 note polyphony sound :p
But if you then added a 3rd PART full of 128 Element waveforms, you would have major polyphony issues. Hence why it seems odd that they would go so drastic with the number of Elements per AWM2 PART..!?

It would come in real handy if like Jason mentioned in another post about setting the XA control to cycle (or random) on many of the Elements. Then you are not using nearly as much of the polyphony off that PART. There's also splits, velocity ranges, etc. that can also make use of the AWM2 PART and manage polyphony more reasonably... 😉

You can always do "silly" things that kill polyphony. However, there are many valid ways to use 128 elements in a single Part without using much polyphony at all.

It's a great building block for new ways of solving old problems.

8 Parts ...

A bummer that those representing Yamaha in one capacity or another that said all 16 Parts could be under keyboard control were wrong. However, not surprising. I mentioned before I thought the source of 8-Part only (Blake) was the most credible. I guess a nice thing is that an external controller can now target any Parts you want by assigning the MIDI TX/RX channels to the same value. Utilizing 9-16 is easier now.

[quotePost id=123668]I guess a nice thing is that an external controller can now target any Parts you want by assigning the MIDI TX/RX channels to the same value. Utilizing 9-16 is easier now. [/quotePost]

I think that would be awesome for a non-expert like me since I still have not managed to play/target only the parts 1-8 (or 9-16) simultaneously via Cubase Pro, not at least without creating 8 separate midi tracks for that purpose. I mean, I have earlier managed to convert some fine 8-part TX816 factory performances into Montage format with the FM Converter. But if I e.g. wanted to play and record on two separate midi tracks in Cubase so that the first track has the TX performance "Electric Piano" (parts 1-8 in a Montage performance) and the second track has the TX performance "Strings" (parts 9-16 in the same Montage performance), I cannot find a way to do it, that is, I don't know how to make one track in Cubase to send the midi notes only on midi channels 1-8 (and another track to send on midi channels 9-16 respectively) simultaneously, when using the Montage in MULTI midi mode. I'm not an expert and I'm sure there is a way to do it (maybe via the Input Transformer of Logical Editor, or some kind a midi remapper, or maybe using the HYBRID midi mode), but I haven't found a way yet.

This will be easier too as you can take all multi-Part sounds (I actually think they're using the term "voice" again - so maybe remove "voice(s)" from the dirty word list) and assign all Parts in a multi-Part sound to the same MIDI channel. Of course you could still work the old way - but there's no longer a need to have a new track for each Part in a multi-Part Performance.

In practice, it's probably "harder" to setup since there are slightly more steps that change depending on your Performance configuration. I think this is something that, should MIDI 2.0 be implemented, that could be covered automagically eventually. I still don't know the MIDI 2.0 story present or future.

I think the way the 128 elements per PART works is that you can have 128 on each PART, but if you set all 128 Elements to Normal XA Control, layering all the Elements, and use them all at once, you will likely be down to 1 or 2 note polyphony of a single PART.

Thanks - that is a great point to make.

Testing things like that are gonna take a LOT of effort to make sure the test is really testing what you think it is.

That is why I added that CAVEAT at the end that I had not actually tested using 128 on multiple parts.

Element definition only consumes memory - and now there is plenty of memory to allow all of the definitions.

Actually USING an element brings all of that polyphony stuff into play.

For the short term I'm just trying to spot things that are either different or have things worth looking into.

The difference in how you navigate means you can't even do simple things like performance search without learning a new way to do it. The new docs don't have the 'link menu' on the right side that lets you quickly go to a section of interest. The new docs have underlining for topics but they aren't actually links - so you need to manually scroll down to a topic.

It's kinda funny when you think about it. If someone told you 30 years ago (or even last week for that matter), that someday there would be a Synthesizer to rule them all, which would have the power to play 128 different synthesizers at once and be configurable to have this ability 16 times over, I think most people would have laughed thinking it was ridiculous and would never happen!! lol :p

I don't care what it's called, I've used a DAW for almost everything, done all the VST instruments and softsynths, and I think the Montage M's AWM2 functionality is pretty awesome!

I also think the Montage is still an awesome Monster Synth with more than enough power that I'll ever need...

[quotePost id=123684]Furthermore, the 16 voices for VA modelling isn't nearly spectacular enough, in this day and age, given the antiquated nature of all that it's working with.
These aren't even Kyra level modelling in terms of envelopes and tonality (128 voice polyphony and 8 part multitimbrality https://youtu.be/_vNP-YGaQIk?t=71 ), and nowhere near the musicality of the Quantum and its analogue filters:
https://youtu.be/6h6_1z4W2sI
[/quotePost]
So I finally had time to listen to your videos above, and I didn't think either synth had as good quality as this synth ... maybe it's the oscillator drift functionality &/or enhanced circuitry that was added, but to me this sounds more like an actual analog synth, whereas those videos sound more like just virtual synths ... and this is just one of many to come from Dom:
https://www.youtube.com/watch?v=g37krx2ZhOQ

I don't know enough about the latest VA synths specs except that I thought the Montage M should have had at least 128 note polyphony for the AN-X, but others who are up on VA's told me that 16 note polyphony is really good for a high end VA synth... Maybe you know something that they don't..!?

I am not sure Montage M's objective was trying to beat the most capable dedicated VA synths on the market in terms of absolute flexibility, polyphony, 1:1 control and so on. I may be wrong.

I personally think it seats comfortably at the top and when you add the rest of the engines into play it becomes a completely different monster. I won't even start on the DAW comparison (loading times, stability, mobility, resource management, you name it).

This isn't sounding that good, and certainly isn't great,

Can't speak for yours but mine is sounding FANTASTIC!

I sold my Montage because it couldn't play 16 parts. Sold my Fantom to get the "M" which can play all 16 parts. I guess I'll be returning the "M" and waiting for the Genos EX.

Two strikes are no PAT on 6/7 and only 8 parts. I only ordered it to check it out but doesn't seem to be meeting the level of hype.

Genos EX might be a better choice for me but we'll see.

[quotePost id=123684]the 16 voices for VA modelling isn't nearly spectacular enough[/quotePost]
[quotePost id=123688]I don't know enough about the latest VA synths specs except that I thought the Montage M should have had at least 128 note polyphony for the AN-X, but others who are up on VA's told me that 16 note polyphony is really good for a high end VA synth... Maybe you know something that they don't..!?[/quotePost]
There was a preview of the M on a Spanish site, which said of the VA, "16 voices become available, a figure that surely speaks for the good of the quality of this modeling" -- I would not assume this is mere puffery, there could be something to this.

VA can be VERY processor intensive. For example, Roland's "better" VA engine (ACB, as used in System 8 and many Boutique modules) has far less polyphony than the versions of the same particular VA emulations they used in the Fantoms (ABM).

Also, I was looking at info on Diva, widely considered among the best VA systems you can put on your computer, and even on the most powerful PCs/Macs, they say that polyphony can be quite low if you choose their "most accurate" analog emulation ("Divine"). They have four options they describe this way:

draft..........CPU-friendly, but FM will sound rough and resonance is primitive (not ‘zero delay’)
fast............Fine for older computers and/or when you need more polyphony in acceptable quality
great ........The best compromise between quality and polyphony on high-power computers
divine.......Top-quality zero delay feedback filters - seldom necessary, but available if required!

So they give you the option of choosing your VA "quality" vs. "polyphony," and the higher quality you want, the less polyphony you get, something they say you have to keep in mind even on the higher powered PCs/Macs.

So hopefully the answer to "why only 16" is "because they want it to be really good." (And remember, the keyboard still has to be able to do a bunch of other things at the same time... up to 384 additional instance of polyphony of other types of sounds while doing the 16 voices of VA.)

[quotePost id=123689]
The Quantum's amazingness, something that's objectively the best VA on the planet (that's in a box), should be obvious. To me, it is. And that's before going nuts with the power of its more complex engines/ways.

There's others to consider that sound vastly better, objectively, too: eg Hydrasynth (another level of flexibility and motion, too) and the pure creme machine that is the Novation Peak/Summit. [/quotePost]
There are certain ways one VA Implementation can be "objectively" better than another... in terms of artifacts like aliasing, or responsiveness, or in terms of how faithful it is to recreating a certain actual analog synth... but I wonder how are you saying these other synths are "objectively" better than Yamaha's. Because short of something "provable" as in those examples, comparing synths is inherently pretty subjective. You can say that one board's violin emulation is "objectively" better than another's, because we know what real violins sound like. But synths largely create sounds with no real world reference. We don't know what "martian landscape" should sound like, so there' no definition there of "objectively" better, any more than banana is objectively better than strawberry or a Les Paul is objectively better than a Strat. Even though it is. 😉

[quotePost id=123689]
Did you stop to think that those folks are in here (I saw their comments, too) and that they're somewhat sycophantic to the environment and less than objective about the current market for VA synths?[/quotePost]
Maybe they are as you say, but only in a obsequious way. :p
But I'm not just referring to folks on this forum...
Are you on the various Facebook forums and the Yamahamusicians.com forums, etc.? I am hearing the same thing from many different people on numerous forums, who are hugely into VA synths, whereby they are saying 16 notes is on the high end of polyphony for a high quality hardware VA synth.
Also, Paul from sand, software and sound has written articles about the processing & tone generation required for VA synthesis and he says it is very intensive. He is also on this forum commenting about such things. Are you saying Paul is less objective than you are?

The Quantum's amazingness, something that's objectively the best VA on the planet (that's in a box), should be obvious. To me, it is. And that's before going nuts with the power of its more complex engines/ways.

You can't say it's objectively the best VA. You have no way to fully prove that. That's a subjective opinion. I like the colour blue better than red and I think the Fantom would look better if it were blue with black than having the red. But there are certain blues that are "objectively" better than others. Others disagree... :p

There's others to consider that sound vastly better, objectively, too: eg Hydrasynth (another level of flexibility and motion, too) and the pure creme machine that is the Novation Peak/Summit. But its effects... a different level of reverb. Immediately apparent that Montage M has not significantly improved Reverb effects, which are probably the single most important thing for a great in-one-box-VA.

Firstly, when you say "sound vastly better", that doesn't sound objective, as you are injecting emotion & drama into the exaggeration of one VA synth sound over another. "another level" could just be 1mm above, as opposed to 1Km...
But, if "reverb" is THE Single most important thing for a great VA, well that doesn't sound objective to me. And I've already experimented with a few quick tests on the Montage regarding its reverb capabilities. It sounds pretty amazing to me and I can have it last more than long enough with a beautiful highest quality billowing tail.
https://soundcloud.com/dclowe/yamaha-montage-fm-sounds-analog (I had to taper down the middle part because the tail just kept going and going)
https://soundcloud.com/dclowe/montage-shimmer

When it comes to something like Kyra, to my ear, with less complex voices and 8 times the polyphony, it's doing a better job, too. Dom is wiggling things like crazy, to try make some shiny out of the new Poly AT, but even with all that, it's not a near thing to the Kyra, and nowhere near something like the Novation Peak/Summit.
It's a bullet point VA, to my ears.

You don't sound objective about this, but rather subjective in your opinion. Not that I am any less subjective than you, but when I listened to all the various sounds in the two different videos you provided, then compared it to Dom's sound using the AN-X, I wasn't comparing reverbs and I "ignored" all the PolyAT wiggling stuff he was doing, listening only for the 'core' sounds, and the AN-X sounds more true analog to my ears.
https://youtu.be/_vNP-YGaQIk?t=71
https://youtu.be/6h6_1z4W2sI

...such as right here (ignoring that 1 PolyAT note): https://youtu.be/g37krx2ZhOQ?t=40
...and here: https://youtu.be/g37krx2ZhOQ?t=21

When comparing the core AN-X sound in this one Performance Dom programmed, none of your videos VA's sounds had that true analog warmth and thickness IMHO!! They just sounded well modelled...

[quotePost id=123689]
The Quantum's amazingness, something that's objectively the best VA on the planet (that's in a box), should be obvious. To me, it is. And that's before going nuts with the power of its more complex engines/ways.

There's others to consider that sound vastly better, objectively, too: eg Hydrasynth (another level of flexibility and motion, too) and the pure creme machine that is the Novation Peak/Summit. But its effects... a different level of reverb. Immediately apparent that Montage M has not significantly improved Reverb effects, which are probably the single most important thing for a great in-one-box-VA.

When it comes to something like Kyra, to my ear, with less complex voices and 8 times the polyphony, it's doing a better job, too.
[/quotePost]

Is it just me, or you have completely missed the point? The question is not about "the best" VA synth out there, it's if 16-voice VA is adequate to today's standards. I would say that there are very few synths that offer more than 16 voices VA and you proved it with all your examples:

Novation Summit - costs half of the Montage M7 - 16 voices.

Waldorf Quantum MK2 - costs just as much as Montage M8x - 16 voices.

Hydrasynth - not a VA, it's an amazing wavetable synth with innovative effects and sound shaping tools, but not a VA.

Waldorf Kyra is mostly unknown to me, cannot comment on it, but still goes to the "very few" category.

Now, what sounds better to your ears is not part of the question. I am not going to question your preferences and will appreciate it if you don't question mine.