Sorry, I'm not asking about this as to solving this problem, just trying to understand how this is possible, in terms of sample per note, from a purist analogue synth nerd dreams point of view.

Is this done somehow differently to elements having a chosen "sample"?

Where can I read about this, and what terms are used to search for info about it? Or is it merely making an element for each note with a unique sample for each?

Sorry, I should have pointed this out, it's super unintuitive, and something I've internalised, so I forgot about it. SORRY you had to go down the rabbit hole to find this. It should be the other way around, like it is in all other multipliers.

No rabbit hole really. The nature reveals itself when you listen. But a lot of these parameters do seem like a reverse Polish calculator - difficult to relate to for no good reason. It's part of that legacy of having very little between the hardware implementation and user interface. At a low level - this is how the register works - so this is what's presented to the user. With little suggestion that would otherwise make things easier.

However, unlike something like unit multiply (which is nebulous) - at least "Time/Velocity" properly frames what happens. Bigger velocities in this ratio will decrease the time (if seen as Adjusted Time=Programmed Time/Received Velocity). And inverting Velocity (0..127 = 127..0) will make bigger velocities increase time due to this "inversion".

You still have to put your math cap on instead of following an interface that is in as much a musical language as possible.

I feel your pain although I also feel the pain of unlearning all of this stuff that has been part of the parameter sets for decades. I suppose if done right - the new parameters will be self-documenting. There's some comfort in having a collection of known issues rather than potentially inheriting a lot of unknown problems.

Fully agreed.

When I first realised AEG stands for Amp Envelope GENERATOR my eyes crossed as I realised Yamaha promotes their programmers to Product Managers.

To users, there's zero need to know these things are generated, and there's a lot of good reasons to suggest that they shouldn't be, anymore, anyways, now that we have multicore processors for everything that run so fast that there's no need for LUTs for things like this, and they might even be slower, due to the memory access required. etc.

For the sidebar question of generating keybanks - you can see:

https://yamahasynth.com/ask-a-question/mapping-samples

The summary there is short but relatively complete. I can only add that loop points can be added by free software as well.

https://yamahasynth.com/ask-a-question/are-there-any-applications-out-yet-to-create-you-re-own-waves-for-the-montage

There's probably software for Mac available ("Loop Manager" was mentioned elsewhere - maybe this means Cakewalk Loop Manager) but I don't know. I've always used Windows for this task.

Endless Wav also mentioned in the previous threads loads on my Mac but I haven't tested the output.

Sorry, I'm not asking about this as to solving this problem, just trying to understand how this is possible, in terms of sample per note, from a purist analogue synth nerd dreams point of view.

Is this done somehow differently to elements having a chosen "sample"?

Aren't Drum Parts exactly that?

Sorry, I'm not asking about this as to solving this problem, just trying to understand how this is possible, in terms of sample per note, from a purist analogue synth nerd dreams point of view.

Is this done somehow differently to elements having a chosen "sample"?

Aren't Drum Parts exactly that?

"Is it possible to build a sound with a sample per note? eg I sample a lush analogue synth on every note... how do I (without using the drums thing) make a keyboard wide mapping of those sounds? Is this possible for end users to do?"

"Is it possible to build a sound with a sample per note?

Yes, I responded to that. The on-board sample load process for a normal AWM2 allows for you to create 1-note keybanks. So you would use this facility to load a different .WAV or .AIFF on to every note. The previously linked material from BM primarily explains that samples must be "prepared" offline. What this means is that lack of a sampler means you cannot manipulate the sample files after loading. And specifically this means you cannot slice/dice, cross fade, set loop points, set the mode (one shot vs loop), etc. Sample files such as WAV have provisions to store metadata. This metadata includes the loop point information. Unless you want your samples as one-shot - the offline preparing means setting these loop points. The links and discussion have suggestions for Windows and Mac software that allow (for free) to create samples with this meta data embedded.

The rundown would be:

1) For all possible MIDI notes, prepare 127 samples (WAV or AIFF) with embedded meta data for loop points if you want those. Best to name by the MIDI note so you can more easily do the next step
2) Recall "Init Normal (AWM2)", go to element 1 and press "New Waveform"
3) One-by-one load each sample into a new keybank and set the note range of that keybank to a single note matching the sample filename prepared in step 1

I was going to link in some pictures from Paul's site - but it would be better to just link you to an article where this is covered fairly well already:

http://sandsoftwaresound.net/tag/yamaha-expansion-manager/

Sample Robot: set up to make each note at given steps of pitch and velocity and it renders them all and fires them into the Montage/MODX...

Excuse my French... BUT FARK!!!

How did I not know about this????

And... given this ability... why are the pianos so ... thin?

And why aren't their libraries of every super synth ever, for these things?

What is going on/wrong with the lifetime support of the image of this range?

Are Yamaha sleepwalking?