It seems to me like the two respond to poly aftertouch, but respond as if it was channel aftertouch. Has anyone been able to have true, individal note poly aftertouch?
 
 
I noticed that Kontakt responds to poly aftertouch, but there's an anomaly - poly aftertouch messages are treated as channel aftertouch in a zone, so you have to create one zone per note to have true poly aftertouch (PITA, but it works). I was wondering if there was some kind of similar issue with Dim Pro and Rapture that I don't know about.

I've had the same results in the past, Craig.  The Poly Aftertouch entry in the Matrix seems to ignore that extra data byte for the current MIDI Note.  I want to say that I discussed this with Rene at one point, but I'll have to check my notes.  It's been quite a while now.  I'll revisit the test bed with some capable controllers when I get back home today.

I think I need to revisit this! I've always assumed that my problems with poly aftertouch had more to do with my keyboard chops, but if Dim Pro and Rapture are ignoring the poly part that could also explain things. I'll do some tests tonight and let you know what I find.

what are you trying to control craig? my guess is that this is due to vst 2.4 architecture and why steinberg included per note expression as part of the vst3 spec. their current synths like retrologue and padshop can map poly aftertouch to items like per-note panning and cutoff, etc. i can try one of those synths in sonar later, but i don't have a poly aft keyboard and it might take a bit to figure how to write in the controller data. my guess is that poly aftertouch for note specific parameter modifications is just not possible in vst2.x.

Oddly enough (or maybe not so strange after all), AKAI 'drum' pads often feature selectable channel or poly aftertouch.  Without going vintage Roland, Yamaha, or Ensoniq, the KMI Qunexus can be used as a source (and I believe some CME models).
 
On the hardware end, a good test receiver might be an Alesis Ion or Micron.  But I never trust the ad blurbs or user manuals.  Use a MIDI monitor.  Status byte w/ 1 data byte: channel AT.  Status byte w/ 2 data bytes: poly AT.

OK, Craig, here's what I came up with.  Simplest configuration:  Rapture or Dimension Pro microhost; Initialize Program; sine.wav Multisample; one Matrix line using Polyphonic Aftertouch as a Source; appropriate Depth for a select Destination.  Polyphonic aftertouch transmission confirmed in Bome's SendSX before testing.
 
In either synth (untested in Rapture Pro), a Pitch destination to a single Element was easy to discriminate.  Two notes with varying PAT pressure seemed to 'fight' each other for control.  Lots of warbling before the firmest pressure won out (set to an octave up).
 
With a Pitch LFO Freq destination, the first note triggered seemed to take precedence.  That is, the first note would have to be released before the 2nd note (with additional pressure) took over.  So, it didn't really accumulate like Channel Aftertouch would.  It was more of a tradeoff between notes and current polyphonic pressure.
 
I should mention here that the AKAI MAX49 sends a 'true' Note Off message with a Release Velocity of 0.
 
Rapture (but not Dimension Pro) got a little stranger.  There were occasions where changing Destinations would 'stick' on the original Destination, until - oddly - another wavetable oscillator was selected.  Very preliminary testing here, and hardly conclusive.  But thanks for reviving the questions that I've had on the "how's & why's" this should work.

this is pretty much in line with how i think the vst2.x sdk would behave. I'm just about to sit down with this and see what i can come up with. brock - did you happen to record any midi into sonar to see if the polyat is being captured correctly?

Arturia's VST2 version of the CS-80V works with Poly Aftertouch but that's the only soft synth I've found so far that does, and it doesn't even have a "poly aftertouch" option - just aftertouch. Weird.
 
Anyway I'm going to be gone for a while, but just wanted to close the loop on this before taking off. Thanks for looking into this!

In Sonar Platinum (Cambridge), I created a small MIDI file with poly aftertouch events to shift the top note of a 3 note chord by a semi tone.  I tested this file with a Yamaha SW1000XG (with the appropriate sysex to turn on Poly Aftertouch to effect pitch), dimension pro and rapture Pro (with Sysex removed and the Mod matrix mapped to element pitch).  I used a simple sawtooth patch on all synths so I can easily hear the effect

On the Yamaha SW1000XG, only the top note shifted by the 2 semitones as expected.  on Dimension pro and Rapture Pro, all 3 notes of the chord shifted.  So it appears that both Dimension pro and Rapture Pro do indeed ignore the note data of the poly aftertouch MIDI string.

the reason iirc that kontakt gets away with this in zones is that each zone that's defined is instantiated within the plugin as a separate instance of the synth. i would guess the cs-80v might have the same sort of architecture.