In today's - erm - wonderful world of universal communication, so-called "moderation" all too often tends either to run amok moronically and without efficacy, or else never get a proper look in at all. Nevertheless, here and now, I'm going to urge caution, self-restraint and moderation for Duality library owners when it comes to considering fabricating a MIDI mockup of Vaughan Williams' Tallis Fantasia. My grounds for nagging? Respect for the aesthetic art and techniques of live orchestral performance. Your motive for paying heed? Perhaps none needed, or perhaps an eye to keeping intact your eminence as a skilled midistrator.
Orchestration for the Tallis Fantasia requires a double string orchestra and a string quartet; Orchestra II being 1 desk each of Vn1, Vn2, Va and Vc, and one player on CB. Divisi and double-stopping are used extensively. So it appears Duality could serve well here as Orchestras I and II, whilst the VI or Synchronized Solo Strings library is an obvious choice for the quartet. Also, there's masses of scope for improvising spatialisation and ambience as desired.
The piece premiered in 1910 in Gloucester Cathedral, conducted by Vaughan Williams; and was also recorded there about 10 years ago by Andrew Davis. (Personally I'm not altogether convinced the piece is musically optimal in such lush ambience, though it's surely a viable option - but that's not the point here.)
The thing the digital midistration community always seem to forget or overlook so readily is intonation. You expect to render Orchestral Intonation always entirely in Equal Temperament? Well you can get away with it easily sometimes but there are cases where ET's 'cartoonified' intonation becomes more of an artistic liability than an asset of convenience. Would you go copy the Vatican's Sistine Chapel ceiling using a cartoon graphics app?
To treat the Tallis Fantasia as if it's merely some melodic lines and some harmonic accompaniment - much like any facile modern piece one might play on a piano - would be a mistake. Why? In a word:- counterpoint. Vaughan Williams made a point of counterpoint in this piece which, after all, is a highly artistic yet essentially faithful elaboration of a 16th century polyphonic psalm chant for 4 voices, originally named "Third Mode Melody"; one of 9 pieces written by Thomas Tallis published in 1567.
Whether in vocal harmony or in a string quartet, counterpoint typically involves Just Intonation - usually worked out between the performers as an essential part of rehearsal. In this piece the main strings are often highly responsive to the quartet and hence issues of intonation can tend to be orchestra-wide. Indeed this may explain some (slightly) noticeable intonation differences at certain moments, between various recorded performances of this piece. But when it comes to attempting a MIDI mockup, ET can make a terrible mess of trying to render delicate and complex matters of intonation such as these.
There's more. In a few places Vaughan Williams deliberately exposes traversal of the Pythagorean comma. As Hindemith has pointed out, usually the comma is fastidiously hidden as far as possible in composition and orchestration. But here, these particular comma-traversals can be heard clearly and with exquisite, rapturous effect in all the great recordings of this piece. (Special praise to the recordings of Barbirolli, Silvestri and Bernstein). The 'great advantage' of ET is of course that it does away with the comma; hence in an ordinary MIDI mockup this sublime aspect of the Tallis Fantasia won't be expressed at all.
Naturally the choice is yours. Perhaps at least now, if you do succumb to this temptation, you might have a better idea of why your mockup doesn't quite measure up to various stunningly beautiful orchestral recordings of this exquisite piece.
The solution? It's a long way away. There'd need to be a critical mass of MIDI mockup creators who feel the need for some technical means of applying orchestral intonation accurately and reasonably conveniently to their MIDI mockups. (Personally I'm not hopeful that MIDI 2 will be a catalyst for this future revolution, despite MIDI 2's elaborate new specifications for fine-tuning operations.) And then we'd need to address the political problem of highly accomplished orchestra musicians no longer being willing to participate in recording sample libraries – because who in their right mind would want thereby to give away what has previously been an exclusive strategic advantage of their own and an important part of their livelihood, i.e. recording bespoke scores for well-heeled media productions?