G#4 sounds like about the limit of useful mapping. I haven't looked at the waveforms yet - but I would bet this is near the end of the useful range of the actual samples. I do not hear any sample that are not "Hah". It's just that the last octave between G4 and G5 is all pitched stretched from the sound of it (and even before then) so when you pitch stretch something - you loose some of the "core" of the sound. Chipmunking.

G#4 sounds like about the limit of useful mapping. I haven't looked at the waveforms yet - but I would bet this is near the end of the useful range of the actual samples. I do not hear any sample that are not "Hah". It's just that the last octave between G4 and G5 is all pitched stretched from the sound of it (and even before then) so when you pitch stretch something - you loose some of the "core" of the sound. Chipmunking.

I find it useful further up than most might. Using AEG Scaling to bring the volume increasingly down to c7 note and layering other strings on top of it, I found it blends and sounds nice.

Some of the specific notes that are off are c5, c#5, and it more of an Ooo sound than an Ooh sound, but still definitely not an ah or hah sound.

I'm saying I don't think it's a mismatch as much as it is just there's no samples up there. So you deal with any aliasing or other effects of the algorithm to pitch-stretch a sample that was much below the target note in the 5th (last) octave of MIDI notes.

If one or two sound more "oo" to you in the stratosphere - then that's sort of lucky if the rest don't sound like that. I'm sure C#5 and C5 are not different samples than notes that precede or follow those (until you reach somewhere in the 4th octave - one before the source "root" note for all of these). I'm still guessing though.

Note you may be able to take a "Hah" from C4 (or maybe lower if lower notes have a more pronounced attack/delay/release), capture it in Cubase (audio) or Samplerobot (although samplerobot is overkill for this step). Edit the captured audio data to clip anything before the note starts and anything past the end. Save that off as a WAV (or your favorite audio format). Then import into some audio processing tools like Audacity (if taking the "free route" on a PC) - or something else you're more familiar with (either commercial tools you own or some other audio tools). Use that to pitch shift without stretching (so attack, delay, and release areas are not shortened). And see if you can end up with your own processed audio samples to cover that 5th octave (and beyond).

I haven't hunted for other waveforms that may already be specifically for covering soprano (typically ends around C5). But I'd agree that "Mix Hah Soft" isn't tuned for the upper range of soprano. What I haven't done is make sure the piano notes match the actual voice pitch (or if the range is somehow octave shifted to a different range) - so I'm assuming the pitches are "true" to the piano keys and basing the MIDI notes on Yamaha's convention.

I didn't check out the "F ..." waveform variation (assuming this means female only) and "mixed" is male and female choir. I only previewed the mixed waveform.

Another idea is to use the Talking Modulator as an insertion effect and use the "a" vowel. Adjust the various parameters to your liking.

I'm saying I don't think it's a mismatch as much as it is just there's no samples up there. So you deal with any aliasing or other effects of the algorithm to pitch-stretch a sample that was much below the target note in the 5th (last) octave of MIDI notes.

Couldn't they have hired Mariah Carey to sing those higher notes? :p
No worries, I have worked with what I've got and made due!

What I am doing is taking a Grand piano, layering some FM-X voice strings, with some Choir voice strings, and Orchestral violins/cellos. I thought the Boys Choir voice alone was awesome enough when I first got the Montage. After layering the following waveforms, all I can say is WOW!:
Hah Choir St
Boys Choir Ah p
Boys Choir Ah mf
M Hah Soft
F Hah Soft

Plus I took the best Elements for Orchestral violins/cellos from 3 different Performances, added them to 1 PART...also a big Wow!
And my FM-X voice strings complement the Choir voices perfectly!

I found out how quick it is to hit the polyphony ceiling too. I can easily do it with just 2 AWM2 PARTs. 8 layers x 2 = noticeable polyphony issues. The piano was at times delayed playing notes from when I pressed some keys, although maybe it makes a difference that the piano I am using is a SampleRobot sampled one that I loaded as a library of .Wav files..!? Once I set ranges on the multiple layered strings for both the choir voice strings & orchestral voice strings, that seemed to fix it up. But even though it's that easy to run out of polyphony, that shouldn't be the reason they have limited the KBD CRTL to PARTs 1-8. For playing entire songs using a single Performance using the Audition button for triggering sequencing and wanting to be able to have numerous options/variety of different sounds to switch to within that Performance using the Scene buttons & over 8+ PARTs, it would be great to have 12 to 16 PARTs that can use KBD CTRL.
However, as always, I'm not holding my breath and can make due with other options within the 8 PARTs using the Assign Swit buttons, Superknob, etc. to make the 8 PARTs more like 16 or more! But still though, I hold out some hope that Yamaha will look at giving us the convenience of more than 8 PARTs via KBD CTRL in a future release ... just not counting on it! 😉

Wow, I just checked and my idea for this on Yamahasynth.ideascale.com has jumped to 61 votes, the most out of any of the ones I posted on there...Even far more than the 'Adding KBD CTRL to Scenes' idea! So I will plug the idea again 🙂

If anyone would like to see this idea as a feature/enhancement in a future OS update, please 'Up Vote' my idea below:
*Expand KBD CTRL to 16 parts on the Montage/MODX (or at least 12 PARTs)
https://yamahasynth.ideascale.com/a/dtd/Expand-KBD-CTRL-to-16-parts-on-the-Montage-MODX/223218-45978

BTW Jason, I just up voted your new idea 'More choices for pitch bend parameters (elements, destinations)' ... that has some promise to do some really interesting and useful stuff! 😉
https://yamahasynth.ideascale.com/a/dtd/More-choices-for-pitch-bend-parameters-elements-destinations/241699-45978

Also, I'll definitely check out your idea of the Talking Modulator as an insertion effect...

I found out how quick it is to hit the polyphony ceiling too.

Welcome to the bigger world of sound design. The difference between “layering” and “programming” is at first a subtle difference, but once you start to discover how quick you can hit the polyphony ceiling, you transition into a bigger, deeper area of sound design.

It take very little imagination to simply layer one cool sound on top of another...cool layers are everywhere, literally everywhere... it takes a deeper look to figure out how to make the best use of the resources. “Programming” gets you into being an efficiency expert, the reward is huge sounds that don’t bump the polyphony ceiling at all.

Yes, you could build a (synth) string orchestra a string instrument at a time, but each string player would cost polyphony. The wise programmer would look for the Waveform that is already a sample of the string orchestra. Of course, it will depend on what else you want to do.

In other words, the lofty goal of building a string orchestra a single violin at a time might look good on paper, but in reality, the ear quickly ‘generalizes’ — a solo violin is interesting, two violins is also interesting, (more so), by the time you get to six violins, your ear/brain is less interested in each individual solo violin, it generalizes and says “violins” — the exact number is of less interest (if you get my meaning).

If you are attempting to build that string sound a solo violin at a time, you might consider the ear/brain and when it begins to “generalize”.
Also since polyphony is the equivalent of money here... what prevents you in the real world from hiring 40 string players is what it would cost in money to pay them, here it’s the cost of ‘polyphony’.

With that in mind, back-in-the-day, in the recording studio, we would turn six Juilliard students into a mighty string orchestra by Multi-tracking to analog tape. Also popular was the use of time delay effects... instead of using it as a multiple repeat sound effect, simply delaying by the slightest bit the start of the signal and use of pan position placement, you were able to create the illusion of larger ensembles.

Good use of effect processors can actually sound as convincing as adding more players. The ear hears the slightest of time delays and the location of the sound goes a long way to supporting the illusion of more players. Of course, there is a limit because “too perfect” is also a giveaway. This is where judicious use of detuning can be helpful (how much is critical, a little dab’ll do ya). Yamaha provides strings that provide just left side, just right side of stereo ensembles, and +/- versions to help randomize the natural vibratos when used in an ensemble. (Some folks don’t listen that closely, Yamaha ‘knows’ some of you do!!!)

Study some of the “bread and butter” Parts and how the different Elements are used. When the Motif series evolved from four to eight Elements, (except for Tone Wheel Organ sounds), the additional Elements were not used to make bigger layered sounds, mostly they were used to expand the nuance that could be evoked from the program. When Motif evolved into the current Motion Control engine, where they’ve blown the lid off of the number of Elements you have access to in a single playable program... wise use of this extra access is reward through judicious use of when these additional Elements are tasked with sounding.

The 18 Elements of the CFX Concert, never more than a single Element is sounding at any one time.

Long story a bit shorter... Effect processors, used as utilities, can be used to extraordinary effect in creating the illusion about the size of your ensemble. They can actually do a better job of it and have the real world benefit of accomplishing the task without the very real expense of polyphony.

“Talking Modulator” is a great recommendation... less is more. I say that because it can be used to make radical changes, but can also be applied to make subtle changes, as well.

Surprised you have not plugged in a dynamic vocal mic and added your own voice to the mix. Remember you don’t have to dominate the mix with the vocoder, but if you want ‘Mariah Carey-type’ notes (which would be incredibly expensive in the real-world) you can mix in just a little bit, just enough of your vocoded vocal to give intelligibility to the “choir”.... your own Vocoded vocal may, in fact, allow you to reduce the polyphony you are spending by simply layering similar components on top of each other.

Just a thought... when they construct a Super Articulation 2 sound in the Genos, the number of Elements is virtually unlimited... well, they can access scores of different samples from a well-organized pool of Elements available to that sound... they are not layering them on top of each other, each Element accessed plays a significant role, standing alone, to deliver a specific nuance or articulation.

On the MONTAGE play “PopHorns Dyn Shake” for an idea of the concept.... Part 1 offers velocity switching horn section appropriate for Tower of Power, Chicago, EW&F type Horn section work, it’s got the horn “Shake” on velocity, it’s got sforzando on the MW, it’s got the octave trumpets on the Super Knob from Part 2, then with two additional Parts, 3 and 4, it lets you introduce the “Doit” and the “Falloff” via [AsSw1]/[AsSw2]. Each articulation is a fully structured sound... the use of simultaneous polyphony is at a minimum. That is programming with the real-world consciousness that polyphony use is precious - particularly if you want the synth to also do other things, as well.

Welcome to the bigger world of sound design. The difference between “layering” and “programming” is at first a subtle difference, but once you start to discover how quick you can hit the polyphony ceiling, you transition into a bigger, deeper area of sound design.

It take very little imagination to simply layer one cool sound on top of another...cool layers are everywhere, literally everywhere... it takes a deeper look to figure out how to make the best use of the resources. “Programming” gets you into being an efficiency expert, the reward is huge sounds that don’t bump the polyphony ceiling at all.

“Talking Modulator” is a great recommendation... less is more. I say that because it can be used to make radical changes, but can also be applied to make subtle changes, as well.

Yes, I am definitely going past just programming and dipping into sound design now. Definitely need to consider realestate more! I have been adding in as many FM-X sounds as possible in place of AMW2 PARTs, but will need to adjust what waveforms/elements/ranges are used for sure.

I did try the “Talking Modulator” and it is really cool. It gives the existing choir voices more breath and switches any "Ooo's" or "Ooh's" to "Aah's" if you choose the "a" vowel.
The only down side to this effect is that it takes the awesome stereo waveforms and makes them mono, so I couldn't use them for my use case! 🙁
It would be great if they could enhance that particular effect to maintain stereo from the original waveform...

I also found out something unfortunate about ALL the cello waveforms that are single cellos (the ones named with "cello", not the ones that are plural and named "cellos" ) . They all have vibrato! 🙁 I would like the option to have a single cello that doesn't vibrato, but I couldn't find one in all the waveforms..!?

That stringed instruments have so few non-vibrato choices is general feedback given here (for violin):

https://yamahasynth.com/ask-a-question/remove-vibrato

And more feedback of request for non-vibrato for other instruments (other than flute and violin):

https://www.yamahasynth.com/ask-a-question/help-new-montage-owner-where-are-quick-attack-no-vibrato-quartet-or-small-ensemble-strings

... and an ideascale suggestion to make the offerings more balanced (vib vs. non vib):

https://yamahasynth.ideascale.com/a/dtd/More-Non-Vibrato-Preset-Waveforms-for-AWM2-in-Synths/233270-45978

... and an ideascale suggestion to make the offerings more balanced (vib vs. non vib):
https://yamahasynth.ideascale.com/a/dtd/More-Non-Vibrato-Preset-Waveforms-for-AWM2-in-Synths/233270-45978

"Up Voted"! 🙂