Grouping Identical Instruments in NotePerformer – Mixing and Mastering in Dorico

(This is a translation of an original post in the German forum)

NotePerformer is unbeatable when you need realistic sound for composing, but extremely uncooperative if you wish a more refined sound for all these sophisticated interpretations in your digital demos.
An export of every individual instrument from a Dorico project to process it’s track further in a third-party program, to balance and remount them all is a very long-winded process. If the tempo, number of bars, or even just the length of a fermata are changed in the score, all steps must be repeated to keep the tracks synchronized.

In his post “Playback Template: Individual Mixer Channels for NotePerformer”, David Tee clarifies how each individual instrument in NP can get its own mixer channel, and consequently in Dorico as well. This article builds on that and suggests how to use his method in a specialcal way to get all mixing jobs economically done without leaving Dorico.

First, we need a look on how NotePerformer usually works

NotePerformer appears in the form of instances that can contain up to 16 instrument channels (slots). Instruments from completely different families regularly appear side by side. For example, all woodwind instruments can be grouped together with two trumpets in the same instance. However, if large ensembles have six horn parts, the two that go beyond the usual horn quartet may appear in a completely different instance. Basically, NotePerformer returns all instruments of an instance always in one single common signal to the mixer channel in Dorico whose instrument is in position 1 in the NP instance. It is this behavior that prevents the targeted sound modification of a certain instrument directly in Dorico without further measures. Stage and room templates also remain permanently ineffective; you basically have to seat the players in NotePerformer. In addition, NP apparently controls the dynamics curve itself; the Dorico setting has no influence whatsoever in tests.

Further considerations about mixer channels

In a wind orchestra, for example, there may be three first, second, and third clarinets in B flat, which nearly never need to differ in timbre. Wouldn’t it be easiest to manipulate all nine tracks at once? This could even be extended to additional soloists, since Dorico only has single players for all wind instruments anyway. If strings are included, it also would save work to combine all the violins of the Dorico section players, regardless of whether they are first or second violins.

The combined output of all instruments of an NP channel as a common signal to Dorico can be used for exactly this purpose. If you get NotePerformer to load identical instruments from different voices into the same instance, they all will end up in the same mixer channel and receive the same treatment. No matter what changes are made to the piece, audio only needs to be exported once for the entire project or each flow.

Sequence of steps

To create endpoint definitions for the NP instances and record the endpoint definitions (NP channels) in a playback template, follow David Tee’s method, but create as many identical instruments in each endpoint definition as there could be parts in a project. For a large symphonic wind orchestra, I used 5+5+5 clarinets in B flat and a solo clarinet in the same key in 16 channels out of a possible 16. Of course, this endpoint definition can also be used for a normal symphony orchestra, which only has two clarinets (explained later).
If you proceed in this way, the number of endpoint definitions in a playback template remains manageable, the use of resources is unproblematic, and therefore you normally do not need to set the master volume or master reverb to zero.
Please note, however, that a clarinet in A (or in any other key) is different from a clarinet in B and requires a separate NP instance in any case.

If you prefer to use certain seating orders in your projects (keyword “German Division”), set all positions to center in each instance via “Settings – Center panning” for the time being.

It is best to create a slightly more comprehensive „main template“ with more instrument types that can be used for several projects.

Preparing and assigning a playback template
Immediately after creating a new empty project is the right time first to copy your „main template“ and reduce it to the instruments that are to appear in the project for the best possible performance: Play – Playback template… – Apply: In the mask on the left below the list, there are square buttons for copying and editing (in another mask). Organize and name your copied playback templates in a meaningful way.
You can also add further instruments to the template and move them in the list. When setting up the instruments in Dorico later, just pay attention to the order in the template.

The customized playback template will now be assigned to the project: again under Play – Playback Templates…, select the correct template and apply it.

Loading the instruments
Set up all players and instruments in the project, preferably in the order in which they appear in the playback template. At the latest after you have played a written note in the new instrument, the relevant NP instance is active; the created instruments can be found starting at channel 1.
For each NP instance, delete any surplus slots at the end before switching to the next instrument (see the clarinet example above). This supports that NP uses the appropriate instance for each subsequent instrument type (e.g., violins → violas). If this still does not work in some cases, the instrument type usually does not match the one in the definition. Unfortunately, there is no way to manually activate previously unused instances of the template.

TIP:
For ensemble parts in larger orchestras, esp. strings, try using at least two players each. In the NP instance, assign them a neighboring different panning to achieve a much more pronounced spatial effect in the listening experience. Once all instruments have been set up, deactivate the duplicated players for the project. This will prevent empty staves from appearing in the score, which would be distracting when composing. Only after the entire score has been completed does it make sense to reactivate these players and copy the entire content of the original staff into the duplicates. The display of the additional staves in the page layout can be suppressed using “Staff Visibility.” The duplicate parts also do not require an own part layout.
If two identical instruments are to be as loud as one was before, the volume of each must be reduced by 3 dB. Since there may be different instrument groups (e.g., Vl. 1 and Vl. 2 in the same instance), this only makes sense in NP and corresponds to a value of 90 to 91 for each channel; for 3 identical instruments, the value in NP is approximately 86.

Settings and seating order

The following settings can be adjusted in the channels of an NP instance, witch stand for the individual instruments:

  • Volumes relative to each other (see previous tip)
  • Panning (left-right position)
  • Reverb (position in the depth of the room)

Dorico stage and room templates are not applicable. Leave the positions in the Dorico mixer set to “C” and use panning and reverb in the NP channel. In the signal-carrying channels of the Dorico mixer (the first instrument of each NP instance), in addition to balancing their volumes, several plug-in effects can also be loaded and applied to the sound, e.g., equalizer and audio compressor. Changes to the mixer settings naturally affect the entire piece. As before, fine-tuning of individual sections in the piece must be done via the volume or velocity of the relevant instruments in play mode.

Mixer channels that do not carry a signal are “dead tracks”. You can set them to volume 0 for better differentiation.

Before output, the overall signal can be finally mixed in the master channel.

Create additional projects

As mentioned, you should design your playback template(s) in such a way that you can copy them and use them for a whole series of new projects with similar instrumentation by deleting unused endpoint definitions.

If you have already created a project with a suitable instrumentation in the described manner, it may be a good idea to simply use a copy of it. In the score, you then only need to select all notes and other entries with Ctrl+A, delete them, and shorten the project/ flow to a few bars. This way, all instruments will sound the same as in the original project, and they will also be positioned correctly because they and their sound settings are not affected by the deletions. You can also save the emptied project as a template and use only a copy to write your new composition.
If some of the instruments are not needed, simply delete their players from the project and remove any surplus NP slots (from NP V5).

Unfortunately, with older projects, it cannot be expected that assigning the new playback template will always produce the desired result for all instruments. But feel free to try it out on a copy—you might get lucky. If instruments do not end up in the correct NP instance and the layout is already complete, the only solution may be to create a new empty project with the same instruments as a “sound file.” Then all voices and the tempo track will be copied into it. Unfortunately, however, there is no way around having to maintain the musical text twice when making future changes.

1 Like

I haven’t read the whole thing, but two things seem relevant.

  1. You can now route each instrument to a separate channel in Dorico’s mixer.

  2. You can of course create your own complete Endpoints, but I also find it useful to have a few saved ‘presets’ inside the VST. Like any VST, you can save the ‘state’ of NP.

This allows me to load instruments that aren’t in my document.

1 Like

@benwiggy, thank you for your addition!

  1. The focus of the article is on the joint sound improvement of several identical instruments. The seemingly random arrangement of the instruments per instance that NP makes is not usable for this. With a stereo output for each, in any case you would have to edit all tracks individually, even if identical instruments were grouped in instances – and there is the CPU requirement.
  2. This could possibly work in a similar way to the playback templates. Is one preset valid per instance? Only then you could save a group of identical instruments in one instance as a single preset with a meaningful name. Does it work in a new project to load the instruments correctly?
    Someone would have to play through that and report here.
    In contrast, the templates have a corresponding editing mask showing exactly which instruments (endpoints) are included in each of them. Depending on your template, all common instruments are already available at the beginning of the project, you could of course exclude exotic ones from the template and load them via a preset. That would also be worth a reported test.

With all due respect, but your initial post is wayyy too long. Is there som “TLDR” section anywhere? :wink:

  1. As @benwiggy has already said, you can configure NP now to return all the channels separately.

  2. “Wouldn’t it be easiest to manipulate all nine tracks at once?”
    Please have a look at note performer: You can configure each player to sound as “n” players by setting some CC. I have set my clarinets to have 4 players each without duplicating anything.

  3. “An export of every individual instrument from a Dorico project to process it’s track further in a third-party program, to balance and remount them all is a very long-winded process.”
    If you have set it up once, it’s almost a no-brainer.
    I write for wind band and have a Cubse template that is ready to take all the individual players’ .wav files and has some extra lanes to record percussion or a soloist. Exporting stems from Dorico and pulling them into the template is a work of 5 minutes, and when you re-export from Dorico after doing some changes, all you have to do is open the Cubase project, and bounce out the mixdown file from there. All done in 5 minutes.

I do agree that the process might look tedious at first, but after setting it up the first time, doing it a second and third time is really quick.

@Estigy hi, ok, quite a little long … - but what is a “TLDR” section (I’m not an English native speaker)?
Due to the length of the text, everyone apparently missed a few details. In addition, (of course) the relevance for one’s own equipment is reflected, so I did for mine. I will try to give a brief clarification.

For a professional prolific writer, I can definitely imagine using Cubase, but then also with high-quality sample libraries. As long as I’m on the road as a hobby composer, the financial outlay for this is simply too great. So I only work with Dorico and use the time-saving NP.

In this constellation, I do precisely not want every single instrument to get its own channel (CPU!). In the clarinet case instead of the 16 instruments, the CC signal ist grat in principle, but would still leave four independent voices, which could not be processed together in terms of sound.
And what about the spatial distribution with only 4 seating positions? Can the 3x5 tutti clarinets then still be realistically distributed over the entire left side of the stage? This is my main motivation for the doubling. In standard instrumentation, it affects the string section.

“TLDR” is short for “Too long, didn’t read”, a form of short summary for those who don’t have the time to read everything :wink:

I totally understand where you are coming from as I’m in the very same position: Hobbyist composer using NP to get a (relatively) nice result for (relatively) cheap money.

To be honest, I don’t see why the minute details of how all the clarinettists would exactly be spaced along the stereo spectrum would be that important - if we use NP after all. NP is great, don’t get me wrong, but in a wind band setting where I have 25+ staves playing at the same time, I’m not too concerned with the exact position of the 4th of the 2nd clarinetts. (To put it in other words: If I ware that picky about such details, I wouldn’t be using NP in the first place.)

P.S.: I’m using Cubase Artist, as this version has all I need for working with those templates. It’s currently about 330 €, and if you wait for a special sale you can get it even cheaper. The improved audio quality, in my mind, helped me create better score videos, so I consider it a net win.

@Estigy

The seating of such a giant section was to get a playout for a competition sound as realistic as possible - and it’s a tip only. I do expect that very few people will take this effort for an everyday digital rendering.
Where can I find, which CC-Signal is effective for doubling?

Here you find it in the NotePerformer docs: NotePerformer 5 - Realistic Playback for Sibelius, Dorico & Finale

It’s CC 104, it goes up to 8.

1 Like

Hi Manfred - Interesting thread. Any chance you can post some audio examples? Ideally before and after the new setup. Thanks.

Hi, @David_Tee , I’d have to look for a project that I still have both configurations of. But I think, with pure NotePerformer only, there is nothing different to hear whether the instruments are grouped or not. Have a new classic orchestra project made for a friend who works in Finale without any add-on.
Of this I could give you a short example excerpt with and without doubling the strings and having activated the sound-plug-ins. I’m not allowed to post a whole movement.
OK for you?

OK, CC 104 is even a dynamic signal, also usable for passages where à2-notifications occur in staves with explicitly written 2 or more identical Instruments. I presume, it will need independend voice playback?
Thank you!

I have to admit that I don’t know too much about how playback is supposed to be configured when using more than 1 voice per stave as this only rarely is a use case in my music.

For me it’s enough to set the CC 104 to some value at beat 1 of the first bar and be done :wink:

Thanks again. Apologies, I may be misunderstanding your original post - I thought this was about mixing multiple instances of NP instruments to improve the overall sound. What’s the difference between “Pure NotePerformer” and what you’ve set up?

Here it comes:
First, the “pure” NP-Standard as you all know it: all Instruments in ongoing order
(2 instances)

Related Dorico-Mixer

Second, NP instances with grouped instruments

This is the related #9 with the violin section, voices à 2 section players. You can see the panning: 2 left, 2 right, in German Seating Order as common in Vienna Classic era - not extremely spread, we don’t have the size of a Mahler orchestra!

And this it how it looks in the Dorico-Mixer: one active Instrument for each type, with equilizer in all visible cases, cello and bass got a tiny compression in the low bass frequencies for less brutal fortissimo.

1 Like

Thats the sound with the NP-Standard

And this is the improved version of the same passage. Compare them on a good audio system!

1 Like