Yeah it's just the first duplicated pair undupe misses for some reason.

Don't you need to process each key/note value lane and remove duplicates along it rather than just process each selected event?  Doing just the events seems like you will miss a bunch of duplicates since you only remember one last note value.

scook
Here is a different take on removing duplicate notes. Did not test it much. Functionally similar to UNDUPE.CAL

 
Thanks for that, scook. But.....am I right in understanding that this CAL (I tried both the original Undupe and your rmdupes) only removes duplicate notes where no other note events exist (vertically)?
If so, it's not usable for, say, hard quantized drum tracks, is it?
 

That is correct. Neither script handles that case. I made no attempt to improve the performance of the original script. For a script to handle duplicates when notes occur in the same time window at different note values, the script would have to loop through the selection for each MIDI note value processing each separately.

icontakt
this CAL (I tried both the original Undupe and your rmdupes) only removes duplicate notes where no other note events exist (vertically)?
If so, it's not usable for, say, hard quantized drum tracks, is it?

 
no, nor for hard quantized chords.