Perhaps the first thing to understand is VOICEs versus VOICEs in the Parts of a Performance or a Song/Pattern Mixing.

When Yamaha hired some of the best synth programmer’s on the planet to create the sounds for the MOXF and Motif-series synthesizers, they were tasked with creating the VOICES that make up the Factory Preset Banks. There are no Preset Performances, technically speaking, all the Performances are in User memory. They are simply different combinations of the Preset Voices. All factory Performance Parts exist as Preset Voices. Some Voices appear in dozens of Performances.

It is the “Voice” that is the backbone of the MOXF. It is in the Voice where controllers are assigned to parameters, also where the dual Insertion Effects are assigned and Elements are routed through one, the other, both or neither of the Insertion Effects. Controller assignments and Insertion Effects follow the Voice when you place it in a Part of a Performance.

Performances in the MOXF, always reference the Voices in Voice Mode. When you want to make a change to a Controller assignment within a
Performance, you must go back to Voice Mode, edit that Voice... then STORE that Voice to internal User memory, then finally, replace the Part of the Performance with your newly edited Voice.

All Part Settings (A Part is where you place a Voice in a Performance or a Song/Pattern Mixing) is a set of offsets that applied to the original VOICE data. A Performance has 4 Parts, a Mixing has 16 Parts.

This arrangement allows you to use the same Voice over and over differently in each Performance you Assign it. You may 6 Performances that contain the same “Full Concert Grand” as a Part. It might be layered across all keys with strings in one Performance and split at middle “C” with a synth bass in another. And split at F#2 with and acoustic bass in another. And shifted down an octave in yet another...

All this is possible because the original “Full Concert Grand” is being “referenced”, but the Performance is a set of parameters that non-destructively allows us to access this Voice differently in each Performance. While the Note Limits, Note Shift, Volume, Pan, Send amounts, etc., etc., may differ in each Performance that the “Full Concert Grand” is used, but the Controller assignments and the assigned Insertion Effects are going to be identical in all the Performances that it is used in.

Application of the Theory
If you want to assign controllers to a program, you must edit the VOICE, in Voice Mode. Because edits to a Preset must be stored to the USER Bank, you must name and store your edits then use that newly created Voice in your Performance. A Voice placed in the Part of a Performance will use the Controller assignment made in Voice Mode.

In the MOXF each Voice has six Control Sets. Each Control Set consists of a “Source” (physical controller) and a “Destination” (target parameter)... the “Depth” determines the amount of change and the direction of application (+/-). Depending on the parameter Destinations, you may have options to activate or deactivate individual Element Switches.

Example, if your parameter is “Reverb Send” then all Elements go together, but if you are targeting something like Element LFOs or Filter Cutoff, for example, since each Element has its own LFO, and its own filter, you will find Element Switches per each Element. This gives you a great deal of control and flexibility.

You must consider your physical Controllers (Source) and the values they send
MW goes minimum to maximum, 0-127; the default is minimum, 0
FC goes minimum to maximum, 0-127; the default is maximum, 127
AsgnKnob goes above or below a center point, -64 ~ 0 ~ +63, default is 0
AF1/2 go Off (0) or On (127); default is Off
PB goes above or below a center point, -8192 ~ 0 ~ +8191
and so on... Each controller has a default it normally resets to when idle.

You must also consider your target parameter (Destination) and the values that control them
Volume is a minimum to maximum 0-127
Reverb Send is minimum to maximum 0-127
Pan is left or right of a center point -64 ~ 0 ~ +63

Choosing a Source that works for the parameter you want to control requires you understand what it sends and how it defaults.

Fading a sound up from zero is better done with a MW than a FC. The MW naturally defaults to 0, while the FC defaults to 127.
You can use either, but just realize it is normal for a sound “biased” to the position of a controller to behave like that controller. If you want to layer piano and a pad, but have the pad begin at 0, assigning its volume to the MW will naturally have it start at minimum.

Extra Credit:
MOXF Physical Controllers and Destination Parameters Tutorial

THanks....have to play with it some more but it sure seemed like some of the assignable functions were affecting an entire performance. Also when I went to the individual voice used in the performance, the AF1 button for example did not create the same effect. I will try to get more detail. Maybe include the voice and performance files and reference some specifics.

You can activate and deactivate Receive Switches within a Performance. The MW or AF1 might be doing something different to each of the four Voices you put in a Performance. Among the Part parameters is a Receive Switch On/Off for each controller... maybe you want the MW to only affect your right hand sound.. and the AF1 button to only work on a single Part... that is where your Receive Switches offset how the program behaves in this combination... Please. read through the tutorial I referenced in the previous post. If you take the time to actually do it (takes about 45-50 minutes) but if you actually press the buttons, hear the results, we think it will clear up much about controllers and how they work.