Divisi playback in Dorico 4


I can’t find a discussion about this issue, so I’ll ask.

When dividing strings, I usually route the full section to an orchestral strings sound; and route divisi to two separate smaller sections.

In Dorico 4, it seems to me that the only way to route things separately is to activate IRV. This creates two separate tracks, each one assignable to a separate sound.

But I need three sounds tath can be assigned, and not two. I need one for the full section, and two for the divisi parts.

Instead, it seems that the first divisi part can only be automatically routed to the same sound as the full section.

Is there a way to deal with this issue?


1 Like

How about a hidden expression to change the channel for the first divisi part?

Maybe duplicate your full section VST in VEPro and assign an unused port/midi channel. Create more voices on the divisi staves and route accordingly. You cannot assign more than one IRV voice to any particular midi channel. It shouldn’t affect the RAM because it’s a duplicate.

Thank you Derrek. It sounds a bit complicate. Wouldn’t this require a separate expression map for the non-divisi and the divisi versions? And then, if the first divisi part is forced on the same non-divisi channel, I don’t see a way to add different expression maps to them.

I was hoping to avoid the ‘virtual voices’ trick with divisi, that would defy the usefulness of Divisi feature. But at the moment it indeed looks like the only workaround.


Assignment of sounds to voices seems to have changed in D4. I’ve tried assigning the notes in the divisi parts a different voice:

Yet, what I can route is the divisi parts with their default Up-stem Voice 1, and not all the individual voices:

I can’t even select the Up-stem Voices 2 in Play mode:

They simply seem to vanish.

I’m obviously doing something wrong, but I can’t understand what it is.


Screen Shot 2022-03-05 at 9.43.40 AM
Try turning the ‘Activate Project’ button off and then back on again and the voices will magically appear. This behavior is known to be not ideal, but it will force create the additional voices and allow for individual routing.


It works! Thank you for the hint!

At this point, divisi will just get a new set of voices. Editing is easy in Play mode, since the new voices will be in a separate track.


May I ask my fellows to try something for me?

I’ve assigned all the notes of the divisi parts to an Up-stem Voice 2. Each of them go to a different MIDI channel and sound. I could have left the bottom divisi part on Up-stem Voice 1, but this would have shown it over the top divisi part in Play mode (assigned to an Up-stem Voice 2).

While the top divisi part plays perfectly fine, the bottom one is not taking playback techniques. It plays everything Natural/Sustain.

If I assign the bottom divisi part notes to Up-stem Voice 1, playback techniques are correctly performed. If I assign them to a newly-created Up-stem Voice 3, playback techniques don’t work. They only work as Up-stem Voice 1.

The correct expression map is assigned to all the MIDI channels corresponding to the various voices.

Is this happening to you too?


Try Alt clicking when you apply the PT to a specific voice…or use the caret and select the appropriate voice by toggling ‘v’ before adding the PT. It should work.

1 Like

I just did a first try with a tremolo on a note. As you know, this one can’t be copied as a separate element, but is integrated in the note.

I removed the tremolo, and then applied it again to the note already in the Voice 2. But it didn’t work: it is still not considered during playback.


Hovering the mouse over the playback technique label, in Play mode, gave me this report on a Legato+Tremolo techniques that is not performed:

And this is an overview of the Violins I part, divided in two, and with two voices for each divisi part:



I had the following parts in Violini I:

  • Staff (a) - Voice 1 - Orchestral Vni MIDI Ch. 01
  • Staff (b) - Voice 1 - Chamber Vni MIDI Ch. 06
  • Staff (a) - Voice 2 - Chamber Vni MIDI Ch. 01
  • Staff (b) - Voice 2 - Chamber Vni MIDI Ch. 06

Staff (b) - Voice 1 was not used. But it was still sending the Natural playback technique to Chamber Vni MIDI Ch. 01.

Staff (b) - Voice 2 was the one that was used. While it was sending the correct playback technique, Voice 1 was winning over its technique.

By assigning Staff (b) - Voice 1 to an empty VI, there is no longer an overlap, and the correct playback technique is received.