I assume that some of these are single-part performances while others are multi-part. With 10,000 arps, 1,900 seems really low but we don’t yet know how many are only a single part vs a multi-part.

We do know that only a max of 8 parts existing in the preset performances, although I think there can be up to 16 from what I’ve gathered. For the multi-part performances, many of the parts (similar to voices I think) could be redundant within other performances not to mention that a “part” can also be something besides an internal AWM2 or FMX voice.

So we don’t really know how many voice-parts are contained inside the Montage. If every preset performance had one voice-part then we have at least 1,900 non-redundant voice-parts which seems low in quantity. The FS1r had something like 1,500 voices back in 2001.

If 50% of the 1,900 have at least (4) non-redundant voice-parts, then that would equal 3,800 voice-parts. It’s hard to determine from what we know currently how many individual voice-parts actually exist inside the Montage.

Perhaps the category function sums up the total number of voice-parts per category for AWM2 and I guess it also includes FMX voice-parts as well. I assume that since voices don’t exist as we once knew them but are now defined as “parts”, that this building block of a performance is the smallest reducible element and is either a AWM2 or an FMX element. In other words, there is no AWM2/FMX combination voice part (like in the SY77/99) because it’s now called a layered performance if it contains more than one element or part. 

Someone please correct me if I’m misunderstanding the Montage “voice/part/performance” architecture.

Additionally, I suppose that many voice-parts in Montage are not included in any performances so until we get a data list it will be impossible to know.

There is only Performance Mode in the Montage, so all existing Parts exist in the performances.

What we know from the specs:
the MONTAGE comes with all of the MOTIF XF’s 1,353 voices, 3,977 waveforms, and 7,881 arpeggio patterns.

1353 voices from XF we will find as Single Part Performaces in the Montage.

and we know also about Montage generally:
• Waveform-ROM expanded from 741 MB to 5.67 GB!
• 2,370 new Waveforms, 6,347 total in ROM
• 1.75 GB Flash-ROM built-in
• 1,920 Preset Performances
• 640 User Performances
• 5,120 Library Performances
• Seamless Sound Switching „SSS“ for Performances with
up to eight Parts
• Dual Insert Effects for all 16 Parts (no
SSS then)
• 192 Preset Live Set Slots, 2,048 User & 16,384 Library
Live Set Slots
• 10,000 new Arpeggios
• Eight Arpeggio Parts simultaneously
• Eight Scenes per Performance
• Masterkeyboard functions for each Performance