the musical uncanny valley of emulation
(A reflection on synthesized sound)
The closer to emulating the sound of acoustic instruments by electronic means one gets, the more artificial it tends to sound. As you approach realism the connotations of the intended emulated instrument come into the foreground, making you listen extra closely for the difference, making you notice how off it really is. I tend to find this distracting for the kind of musical expression I’m after. (I find that it works slightly different for non-instrumental yet naturally occurring sounds such as rain / thunder etc. but that's a topic for another text.)
Modal synthesis, physical modeling and sample-based instruments (such as modern digital pianos) all share this issue, to my ears. However, I also encounter a similar distracting phenomenon when using typical synth sounds such as sweeping resonant low pass filters over a saw/pulse wave etc. Then I find that the connotations of the Synth (with capital S) draw too much attention to itself and distract from the melody.
For this reason, I have so far opted to use more of what I experience as plain sounds such as sine waves, wavefolded sinewaves, or sinewaves mixed with other basic waveforms. I have also been modest in the use of modulation of these sounds, keeping them rather static or changing over longer timeframes. By leaving some expressiveness out you can also gain focus for other parts of the music, has been my thought.
However, in this expressiveness of sound (such as changing bow pressure emphasizing the groove etc.) I find some of the most interesting details in the source material. (Again ‘,which’ instead of starting a sentence with ‘Which’) Which I have as a goal to highlight in different ways! So this poses for me a great challenge: How to play as expressive as acoustic instruments without getting stuck in the musical uncanny valley of emulation?
(next: breaking my rules)