LA-based composer Mike Verta is an accomplished jazz pianist and session musician. He has written innumerable pieces for commercials and advertising as well as scores for a variety of television serials, short films, theatrical features, theme park attractions, multimedia projects, and two solo jazz albums. He lives in Los Angeles with his wife, actress Danica McKellar, and their son, Draco.
Last January, composer Mike Verta spent his own hard earned money to take a short piece that he had written for an orchestra and record it at the famous 20th Century Fox Scoring Stage with some of L.A.’s finest musicians. And legendary recording engineer Shawn Murphy. He had previously done a very good sample-based version but as an unabashed advocate of live players, his expectations for this version were high but the results had some surprises….
I asked Mike if there was a difference in how he approaches a sample based version if he knows it will ultimately be replaced by a live version – versus a hybrid or purely virtual version.
He responded that in all these scenarios, the common factor for him is that “the sound is secondary to the feel. It has to be as emotionally and dramatically satisfying as possible.”
He then said that as someone whose goal whenever appropriate and possible is to end up with a live version, there is always the issue of not making the mockup “too good” as he may run the risk of the client saying that it sounds so good that a live recording is not necessary, nor “too flawed” and thereby running the risk that the client does not hear the music’s potential. He said that sometimes making the mockup really good has given him a greater credibility in the director’s eyes so that when he then says, “If you think THIS was good, there are orders of higher magnitude waiting for you” he has successfully been given the budget to do so more times than you might think.
Mike also believes, as I do, that while many people cannot necessarily identify a piece as being sample-based when they hear it, when it is A/B’d they hear the difference and more importantly, “feel the difference”.
Because he believes that precision is secondary to feel, Mike avoids quantizing MIDI. Now it needs to be said that Mike is an excellent, trained pianist and so is able to execute what he has conceived well in a reasonable amount of time. And because he believes the way one performs a sampled instrument has such a huge impact on the believability, he recommends that younger composers perhaps spend less time and energy learning to manipulate notes well in a MIDI editor and more time developing keyboard or other controller skills.
So given his belief (and mine) in the greater emotional wallop provided by real players, it I was interesting to me that initially Mike was disappointed with the result that he had achieved in a top studio with a top engineer and top players. For instance, in some areas the dynamic range of the sampled version was more dramatic.
Mike attributes this to several factors:
1.The extremely challenging and complex nature of the material, which players had little time to absorb and perform.
2.The improvement in the quality of the sample libraries and his use of them since the last time he was in this situation with an orchestra of this size.
3.The inordinate amount of time he spent on the sample-based version. He spent approximately two weeks of solid work on six and one half minutes of music, something he obviously could not do on a project with a fixed deadline. He did this because it helped him focus on the dynamics and articulations he wanted to mark on the score and parts for the players, so it got “a lot of love”.
4.Due to some technical issues that arose at the session, he was not as single-mindedly focused on the orchestra’s performance as he should have been.
Ultimately however, after some re-mixing he got a performance he was happier with. He already has had an occasion to play both versions for a client who subsequently hired him to do a score with the London Symphony Orchestra based on the live version of this piece.
But this is, after all, a technology column, so let’s focus on what Mike had to say about his silicon orchestra. I mentioned to Mike that he had posted a version of a famous piece, as had another renowned mockup artist and that while the latter’s arguably sound more like recordings I had heard of the piece (more “real” if you will) I found his more listenable. I have always maintained that it is less important that it sound “real’ than sound ”good” and that the two are not always the same, and that this was illustrative of that tension.
Mike responded that for him, “The intent is to have a convincing virtual orchestra AND an emotional quality;, that is the goal. But there are only so many hours in a day. If you want a truly convincing sound, you’re going to spend most of your hours on the performance of the samples, instead of on the music itself – the counterpoint, the orchestration, etc. So given that conflict I will always err on the side of the music – that’s where the emotional impact lies. A great piece of music can cure a lot of mediocre sample ills, but not vice-versa.”
Mike agreed with me that just as we write to real players strengths and weaknesses, so do we need to write to the samples strengths and weaknesses rather than trying to make them play something we conceive of that they are not well suited to. Personally, I believe that because as film composers we learn early on to be willing to make changes in or music to suit the client’s wishes, we cannot take the traditional concert hall composer’s attitude of, “I hear this in my head and that is what must be played”.
While Mike agreed, he did see a “dark side” to this. “Because the sample libraries are ultimately limited in what they do and because they all try to get the same basic articulations, that’s what has led to the great homogenization of music. Because everybody’s instruments can only do the same handful of things, everyone is doing the same kind of things in their writing. It is not an accident that the big epic drums are so omnipresent that it is bordering on self-parody, because if that is what is in your arsenal and it works, that is what you are going to use. And if among your string samples, the spiccatos are the ones that sound really good and your Gothic choir sounds really good, well that is what you are going to use. So the music has become homogenized because everyone is using the same stuff”.
I pointed out that the good news is that many of the newer libraries do more things well than their predecessors so that there is room for the sample-based or hybrid score composer to grow. The degree of compromise is less.
Mike agreed and also pointed out that the mockups he and I admire most are done by guys who have “ chops with the live thing” because we know the ways the real thing would work and massage the virtual stuff to get it there. Skill and craft are always useful. I think Jerry Goldsmith would have more to say with a virtual instrument perhaps, than someone with more libraries but less musical training. And I think that sometimes the best use of these libraries is to do things that the real guys cannot do well, or reliably. With very “conservative” writing or effects, like a big blatty C minor chord in a difficult range- it is possible that the samples will sound as good or even better than a live attempt. Conversely, it is therefore maybe best to use the real orchestra in the areas where it is doing things that cannot be done well by samples. That is where it is sexiest and really shines. I can’t tell the difference for instance of a single note played by a real French horn player from one played with Hollywood Brass because the context is too conservative to reveal the limitations of the samples. So there’s a time and place for both. Horses for courses.”
We then turned to the issue of templates. Mike runs 2 Macs with ProTools as his main DAW and Vienna Ensemble on his slave, but it only runs a handful of things. He has always felt that his orchestra template is perhaps bloated with unnecessary options. He says, “My template has to be fast. I will put up with keyswitching but I am not willing to say about a given part, ‘well, half of this will be one track, and then I’ll cut to a different track for the second half, etc.’ I have no patience for that; takes too much time and kills the vibe. I have no patience for that. Also, for me, there is a real correlation between track count and score pad staves. I think like a score pad and I see the layout in my head. I can’t have 900 tracks to scroll through. It is too slow and throws my orchestration sense off. But I will acknowledge that that is one of the reasons I don’t think my mockups sound as good as some others who will do that because I need to enjoy my work and I would not enjoy working that way. Also I spend so much time on the working out of my melodies, counterpoint, harmony, orchestration, I don’t think in terms of just holding a key to play an ostinato for 30 seconds, so to be able to go through everyone of those moments and try to make everyone of them sound real, it would take me 10 years to score a film. So for me there is often a difficult choice to make, you can either have complex music or complex sound. I think the result of our often obsessive pursuit of getting the most realistic sounding mock-ups is that we hear a lot more producer-based rather than purely composer-based music, now. And I’m not sure that we’re better off for it, in the end. Still, it’s a wonderful tool to have.”