I'm not sure what you mean by "double note". Ignoring arpeggios altogether - if I press a single piano key down which is a violin PART - I expect there to be no sound before I press and a single note until I let go. The duration of the note vs. the tempo would determine the length of the note (16th note, 8th, quarter, half, etc). One way I could say there is a "double note" would be if I press a single piano key and hold that key down - but hear two 8th notes one after another even though I did not let go of the key. This is one type of "double note".

Another way I could get a "double note" would be if I press and hold the same key - and I would expect a single tone matching the pitch of the piano key I press. If instead, I hear a harmonic interval (two simultaneous notes sounding to make a harmony) then I could also call this a "double note".

Which kind are you experiencing?

Delays with ARPs are usually because you trigger the note just past when the keyboard is looking to change notes. Say you can change every measure. If you miss the measure (change after the measure arrives) - then you have to wait until the next measure before the ARP will change. There are other time-based markers (other than measure) when this type of behavior can happen. This may either involve changing the unit or changing how you play the ARP so you change just before the beat/measure/etc. instead of coincidental with the beat/measure/etc. Quantization should take care of this and place the chord on the correct beat.

I'm not sure how a "voice" fits in to this. I'm not sure if you mean an arpeggio track or a PART. Usually I can infer what is meant - but this time it's not clear.

If you're using 4 PARTs just for these single-finger chords - the ARP approach makes sense. There's more than one way to accomplish this using ARPs. Then you could also accomplish the same goal by using different microtuning + pitch offsets for each PART. This approach could create diatonic chords that change from minor, major, dominant, etc - with using a single finger. There would be no quantization/delay/etc. using this method since the ARP is not playing the notes - you would be.

Thank you for the response Jason. I guess I probably used the wrong description of the issue I was having with "double note". When I try to trigger the next arp with the single note, sometimes it is delayed by a split second. My best guess is that because for an example, one arp chord uses A3, and the next arp chord also uses A3, the second arp is waiting for the first arp to release the note before the next chord can be played. So perhaps a better explanation would be that the montage will not allow a "double note", I'm trying to make the chord progression as legato as possible. I played around with the tone section and cutoff the sound before the next chord and extended the release and sustain, but that didn't seem to help the triggering issue. Again sometimes it works flawlessly, other times I get that delay and its driving me nuts why the performance is so inconsistent.

I don't know exactly what you've set up there. In general, "timing is everything" with respect to what you put into the ARP (trigger) vs what comes out.

I'm not sure what the expectation of your chord changes are. 16th note triplets? That'd be difficult to trigger. Whole measures? That's easy to trigger. Something between? There's no information here. I wouldn't need an average frequency, I'd need the maximum frequency (meaning minimum duration between one chord and the next chord).

There's probably two other ways to accomplish what you're trying to do.; Not enough specifics here to silver-platter a solution.

This (silver-platter) would take:

1) Rhythmic and note "score" of your expected arpeggio output (the chords you want the strings to play and the rhythm)
2) The amount of PARTs you can dedicate to the strings. This would mean how many PARTs you use for other things which are not the strings. The answer would likely be 8-(other PARTs)

It's assumed that you want to use one finger to make chords happen.

What sound you want for the strings is not important. You can change that if the one picked is wrong.

Hopefully micro-tuning is saved in soundmondo or an X7B or that example would be difficult to transfer. Not a big deal - just would eliminate one of the ways from being easily communicated.