Thanks for all the effort James, I am sorry that what I have posted has been so misconstrued, I have the utmost respect for you as a programmer.

Unfortunately, the MP3 example you posted is kind of hard to really distinguish the sounds, as you already used section brass sounds in the first place. It would be fascinating to hear how this works if you replace the section brass with SOLO instruments.

You see, in small section work (think Chicago, or EWF, TOP, that sort of thing) you are really talking about 3 or 4 SOLO parts, sometimes playing in unison (or octaves), sometimes spreading out for block chording, sometimes two or three holding, while one solos, that sort of thing. When all playing the same note, yes, you get a section. But the minute they divisi, you are really listening to four solo sounds. I would be fascinated (and a lot more enlightened) if perhaps you could run the same PCG but with bari, bone, alto and tpt solo sounds, and we might get a clearer indication of what is happening.

I understand your preference for sequencing all this, and in a sequenced environment, this is ideal. But we are ARRANGER players. We are looking for things that we can do LIVE, without prearranging it. And a horn section that automatically splits out to the solo parts when you chord is EXACTLY what you need for doing live horn parts in R&B and jazz, etc. Admittedly, not so much for Enya (but again, synth sounds opens many possibilities) but acoustic music revolves around divisi.

By the way, I am not touting the ARX board as the be all and end all. It is a good start, but needs more complex rules than simply additional voices from the bottom up as you add more notes. I would also like to see modes where, until you reach the section's true size, if you only play two note, they split down the middle (or bracket), if you play three, they go to 2 voices (if a six man section) etc., think basic rules govened by the size of your ensemble. I would like to see hold parameters, so if you play three notes and hold pedal them, the remaining voices either solo up or unison up and don't divisi. I'd like to see modes that automatically drop octaves of each of the parts as they climb outside their range.

The surface has only been scratched with what divisi and alternative note (or oscillator) assignments can do for us musically. And the beauty of it all is, it is simply code. To be honest, it isn't really very much more processor taxing than arranger 'Melody Intelligence' block chording tricks they already do. All it takes is for some of us to see the potential. After all, that's what got us Melody Intelligence in the first place.