More news on soon-to-be-released ‘Dorico’ music notation software
CHICAGO, July 22, 2016 — We continue today with the second and final installment of our exclusive interview with Daniel Spreadbury, the well-known and well-regarded figure in the demanding field of music notation software who currently serves as designer and marketing manager for Steinberg Media Technologies’ “Dorico.” This promising new music notation package is currently expected to debut in Q4 2016.
In this installment, we focus on the Dorico feature set and explore the potential future path of music notation software in today’s rapidly evolving technology environment.
Mark Nowakowski: Many composers have long dreamed of having a single “in the box” experience, where the sequencing process to create a mock-up or final audio product can occur concurrently with the notation and composition process. Will Dorico move us more in this direction?
Daniel Spreadbury: Certainly the aim of building this kind of direct control over both how virtual instruments are set up and the specific nuances of musical phrasing and expression are intended to at least delay the point at which a composer needs to move from his scoring program of choice into his sequencer to complete work on a mock-up. At the same time, it’s not our intention to duplicate the DAW [Digital Audio Workstation] functionality of Cubase in Dorico. Each application has its own identity, and we have no plans to turn Dorico into a fully-fledged DAW. Over time, however, we hope to be able to bring Cubase and Dorico closer together and to build powerful interoperability workflows between them.
MN: The audio engines for the other major notation packages are mediocre at best, making it difficult to bounce a truly quality mix for mock-up or final scoring purposes. There have been whispers that Dorico is using the robust Cubase audio engine for such duties. Can you tell us more?
DS: Dorico will indeed use the very same audio engine as Cubase and Nuendo, which is renowned not only for its high performance and efficiency but also for its high audio quality. Dorico will support VST3 compatible plug-ins and certain well-behaved VST2 plug-ins—a list that I am pretty sure will include Kontakt, EW Play, and other important sample players that have not yet made the move to the superior VST3 format—and will be able to produce high-fidelity audio with full floating point accuracy along the signal path. Over time we hope that this link to the Steinberg audio engine will unlock all sorts of powerful functionality within Dorico, and [prove to be] a really unique advantage for the software into the future.
MN: Can you tell us more about the onboard sounds that will come standard with Dorico? Will any of them feature automatically switching techniques – say, a violin changing from normal vibrato to non vib to pizz?
DS: Your readers might have read the description of how users can work with third-party virtual instruments earlier on and [how they may] have concerns that even with the built-in samples they will need to create an expression map, load the samples manually, and so on. That’s not the case. It’s very important to us that we are able to provide the kind of flexibility and direct control that you find in applications like Cubase. But it’s equally important that it should be possible for users who don’t have a great deal of experience and expertise in this field to simply create a project and hit a key to hear reasonable playback without needing to spend any time setting things up.
So to satisfy those users, Dorico comes with the HALion Sonic SE 2 sample player and its library of more than 1500 sounds, plus the complete HALion Symphonic Orchestra library, a high-quality sample library featuring the majority of the standard orchestral instruments – though not an exhaustive array of percussion, and not brass and jazz band instruments – in a variety of detailed patches that include different playing techniques. [These generally include] a consistent set of techniques, such as legato, staccato, half-step and whole-step trills, accents, and so on, along with some more instrument specific techniques, such as pizzicato for strings, flutter-tongue for flutes, and so forth.
When using the HALion sounds, Dorico will be able to load sounds automatically into channels into instances of HALion, including [the ability to add] further instances of HALion as needed; and will automatically switch playing techniques when it encounters the relevant notations in the music.
MN: How about key combinations? Many users have spent years inputting music a certain way and don’t want to spend the time necessary to learn an entirely new system. Will Dorico offer any options in this regard?
DS: We have designed Dorico’s key commands, or keyboard shortcuts, from first principles without specific regard for how other programs do it. For example, our use of the number keys along the top row of the keyboard for durations is based around the assumption that we want the longest note value with a shortcut to be the double whole note (breve), which has the shortcut 9; and this therefore dictates that the shortcut for a quarter (crotchet) is 6, the shortcut for a 16th (semiquaver) is 4, and so on, down to a 128th note with the shortcut 1. In Sibelius, a quarter is 4, and in Finale, a quarter is 5, but in Dorico, it’s 6.
We have also tried to create a consistency in the way shortcuts are assigned, so that it’s easy to learn them. For example, to create a particular kind of notation, you hold Shift and type a mnemonic letter for that kind of notation. Shift-C is clef, Shift-K is key signature, Shift-M is meter, Shift-T is tempo, Shift-O is ornament, and so on.
Even though the shortcuts in general bear little resemblance to those used in other programs, they are hopefully easy to learn and to remember and are logically set out. If, however, you want to customize those shortcuts to match your own preferences, it’s possible to do that from Dorico’s Preferences dialog.
MN: Now it’s time to sell us: Aside from the clearly improved visual appearance of the scores in Dorico, what do you think are the best one or two new ideas that Dorico brings to the software notation table?
DS: Firstly, I think Dorico represents the application of our many combined years of experience of thinking about issues of music representation, application architecture and design, user interface refinement, and a deep understanding of customers’ needs. We have been remarkably lucky to have been given the opportunity to write a brand new scoring program from a position of considerable experience and expertise and to have been given the time to address lots of the fundamental design issues that would hamper future development of the application’s functionality in future versions. So we have brand new code, written to take advantage of today’s highly parallel CPUs, but written by an experienced team [that] really understands the domain we’re working on.
Secondly, Dorico’s entire model of separating the workflow of producing a score into discrete phases by way of its five modes—albeit modes that you can freely switch between at any time during your use of the program—has enabled us to think much more holistically about the approach to each set of tasks, [while] not being forced to compromise how a particular aspect should work with relation to another nominally unrelated area because they have to share the same screen or shortcuts.
It might at first seem unnecessarily restrictive that you cannot, for example, add a new movement or change the instruments while you’re in Write mode, or that you can’t edit the pitches of notes when you’re in Engrave mode, or whatever. But actually, once you try it out you start to appreciate the cleanliness of this separation of concerns.
Thirdly, Dorico’s approach to inputting and editing music is by design a great deal more flexible and closer to the way MIDI editing works in a sequencer than the way other scoring programs tend to think about music. Once notes have been input into Finale and Sibelius, they become totally concrete, fixed both in their position in time and in the way in which they are displayed.
A note that is a quarter in played duration that ends up notated as a 16th note tied to an eighth is fundamentally two or more discrete events in those applications rather than a single event as it is in Dorico and as it is in a sequencer. This leads to all sorts of small frictions in the way input and editing works in other programs; as when you add an articulation to a tied note, it might be added to both notes in the tie; or when you re-pitch a tied note, only the note at the start or the end of the tie is actually edited, leaving a dangling tie behind. These are small problems that you learn to work around as you become experienced with your tool of choice, but they are completely unnecessary problems that emerge from an inappropriate approach to music representation, and in Dorico they simply don’t happen.
Dorico allows you to insert notes into the middle of an existing voice, shuffling all of the rest along or overwriting the subsequent notes in a more intelligent and less destructive fashion than existing programs. I don’t think it’s an exaggeration to say that the way Dorico handles note input will allow composers and arrangers to approach working with scoring software in a fundamentally different way than they have been able to in the past.
MN: Where do you see the future of music notation software going? Are any of the forthcoming technologies – like VR – on your development radar for integration with current creative options? What about stand-alone software “wrappers” or integration with programs like PD or Max?
DS: My colleague Ben likes to tease me with visions of orchestral players wearing virtual reality headsets with front-mounted cameras so that they can see both their music in virtual space and the conductor in real space in a kind of augmented reality performance space. I have to say I can’t see this happening in the short to medium term, myself. But at the same time I think it would be a mistake to posit that these emerging technologies such as VR, or speech-driven interfaces, or machine learning from big data sets and so on, will not in time change the way we think about creating and interacting with music notation in potentially quite profound ways.
I don’t claim to be a great prognosticator or futurist [but] I think that there is a huge amount more we can with the technologies that are in the mainstream today. If we never had to even consider how we might make Dorico work in VR, say, or in a speech-driven fashion, we would have more than enough interesting and challenging work to do with the boring old keyboard, mouse, and MIDI keyboard input methods we’ve been using for the past two decades to keep us busy for another decade or more.
Technology for its own sake is in some ways not that interesting. It’s all about how we can apply technology to the needs of musicians to help them achieve their creative goals more quickly, more directly, and more simply than ever before. If we can keep inching along that continuum at least a little more every week, every month, every year, we’ll be doing just fine.
MN: Many thanks, Daniel, for your illuminating and detailed answers. We look forward to being able to review Dorico in the near future!
Note: This interview would not have been possible without having had an earlier conversation with a number of composers whose varied perspectives and expertise informed the questions ultimately posed to Daniel. These composers deserve to be mentioned here.
A special thanks to the following composers:
- Stephen Andrew Taylor (http://www.stephenandrewtaylor.net)
- Daniel Kellogg (http://www.danielkellogg.com)
- Jeffrey Quick (http://jeffreyquick.com)
- Jennifer Jolley (http://www.jenniferjolley.com)
- Michael Bratt (http://www.composerbratt.com/)
- Roy Magnuson (https://roymagnuson.net)
- Nebojsa Macura (http://www.societyofcomposers.org/members/NebojsaMacura/)
All the expert input from these composers helped this columnist to craft the questions for this interview.