I've pretty much completed orchestrating my score for a musical. As soon as I'm sure that I don't want to make any more changes to the parts, I will do a bit of polishing (making sure that I am using the best virtual instrument, from each of the libraries that I own, per part, per number) and then, hopefully, begin trying to record the vocals.
For the full score, I set the orchestra size at the current musician's uninon minimum for the largest Broadway house (18 players + a conductor, or 19 players with the conductor playing an instrument). In my case, that translated to 19 with the conductor playing the second keyboard.
Since the story is set in the late 19th century, I didn't want it to sound like I was using any instruments that weren't available in the period. I didn't use electric guitar, bass or synth sounds. To make the small size of the pit work, I did orchestrate for two MIDI keyboards, however. They play piano, harpsichord, organ, harp, acoustic guitar and banjo, accordian and pitched percussion sounds. I also have a drummer, a percussionist (mostly timpani), a four piece brass section, and four reeds.
Which brings me to the strings ... Everything I've read about orchestrating for the theater today says this is the most difficult area. Because a string section of 5 or 6 is so easily overpowered by the other instruments. I've tried to be clever about how I used them. In some numbers, I save the strings for quiet sections with just one of the keyboards (usually playing a harp or harpsichord patch) and bass accompaniment and a solo reed for the instrumental breaks. But there are times when I need more. Especially in the big tuttis, where it really should sound like a full orchestra is playing.
I tried several different ways of scoring the strings, eventually settling on 3 violins, 2 violas and 1 cello (with the string bass sometimes added). This allows several ways to voice them, where more than one instrument is playing on each part, to make it sound fuller. But what the orchestrators who do this kind of thing professionally are telling me is that 3 violins don't sound noticably fuller than 1, 2 violas don't sound noticably fuller than 1, and a solo cello sounds weak unless it is doubled with another cello (or the violas, or the arco bass, if the ranges make sense.)
[NOTE: I am not talking about simply being heard. Strings are amplified in the theater these days. Increasing the number of instruments on a part is done to make it sound fuller, rather than louder.]
So, now I am contemplating doing what I tried to avoid: going with a string quartet (Violin I, Violin II, Viola, Cello), using this, with or without the string bass, in the quieter, more intimate sections. For the sections that need to be fuller, I'd use a third keyboard, playing a string section patch.
So, my first question, to those who have orchestrated for the theater, or faced this problem eslewhere, is: Do you share the view that this is the better way to go? Secondly, in the sections where I use MIDI strings, should I add the live strings? And, if so, should I divide them (i.e. voice them so that the same chords are played by the quartet)? Or play the upper voice with the violins and viola, where possible, and the other voices on the keyboard strings? If you use both techniques, what is your criteria for deciding how to use the real strings with the sampled strings?
Any input would be appreciated, as very little has been written about it. Thanks.