Synthesis, Sampling, and Plain Old Acoustic
I've been doing a lot of pondering lately on my future with electronic music, spurred by considering Reaktor as a new software instrument (and instrument building tool) of choice. Reaktor is very cool in many ways. I most appreciate its completely open-ended flexibility, letting you build instruments from the ground up in whatever way you want. My frustration with Reason, by comparison, has been that it's flexible in some ways but not others, because the developers foresaw some uses for the instruments, but not all. This becomes especially apparent when playing from MIDI drum controllers.
But it turns out that Reaktor falls on its face with sampling. It's pretty awful. I've tried addressing these issues on NI's web forums, but to no avail. Their implementation of velocity switching is awful, and it appears to be very difficult to build a velocity crossfading sampler that both works and doesn't eat tons of CPU power. This really shouldn't be a difficult problem, especially not for a system as flexible and powerful as Reaktor, but apparently it is.
This whole adventure led me to consider the following: if you look at "return on investment" from my musical instruments -- that is, what I'm getting good use of relative to its price -- the one clear winner is my LP Giovanni djembe. I've played this in a number of settings, and currently play it at church, always with very positive response. Runners-up would be the LP compact conga, which I've used to improve my conga playing immensely, and my Yamaha S90 piano/synth, which I've been re-learning piano on.
The electronic drums and soft-synths, on the other hand, are all competing for last place. I don't blame the MIDI controllers (trapKAT, drumKAT), since these have been doing their job admirably. But the software side has proven to have innumerable limitations or hassles. Choosing and organizing samples that really sound good together is painfully tedious. (That's the nature of sampling, I think.) Getting MIDI control to do exactly what I want either means jumping through lots of hoops or is just impossible. (Mostly the fault of Reason.) Amplification is expensive and a hassle. (Nature of electronic instruments.) But probably the most significant reason it's just not panning out is that people -- band leaders and audience both -- want acoustic drums and percussion. At least with the people (esp. churches) I've been playing with, nobody's interested in electronics.
But the djembe... people just love it. I like it too, but I'm always surprised at how many complements I get after playing it. I'm decent with hand drums, but I'm not that good. Clearly some combination of the drum and my playing is hitting the spot, though.
I think my priorities from here out will continue to shift to low-tech hand drums, regardless of my instinctual draw to electronic percussion and synthesis. Reaktor is way cool despite its sampling limitations, and I'll probably buy it someday just to mess with it, but before that I'll be buying a conga and/or bongos. At the church I'm playing at they already have congas (LP Aspire), but the Giovanni series requinto would make a great primary drum. I'm also considering bongos because, while I like congas better, bongos would diversify my drum collection a tad more.
Eventually I'd love to do a combo electronic and acoustic percussion environment. But that will require a more open-minded audience, more equipment on my part (at least amplification), and also a good sound engineer that can mix electronic and acoustic effectively.