Archive of the former Yahoo!Groups mailing list: MOTM

previous by date index next by date
previous in topic topic list next in topic

Subject: Re: Imitative Synthesis and Implications for Hardware

From: "rogerpellegrini" <rogerpellegrini@...>
Date: 2007-04-20

--- In motm@yahoogroups.com, Mark <yahoogroups@...> wrote:
>
> Well, I'm not sure what Brian Eno has to do with this. I have
dozens
> of his records, and I've never heard him do anything vaguely similar
> to imitative synthesis. Perhaps I'm missing something.

Perhaps. His point, and mine, is that EQ can dramatically shape
sound into what you might like. That could be an imitation of an
existing instrument, or something else. In Brian Eno's case, much
the latter.

> EQ cannot produced inharmonics, it can only change the amplitude of
> frequencies that are already present.

Yes, this is a premise of subtractive synthesis. We start with
almost all the frequencies (a sawtooth) and snip away wearing oven
mitts. I'd like to be able to use a more sophisticated surgical
approach.

> The resonance or frequency response produced by the body of an
> acoustic instrument, as well as the room it is in, while often very
> complex, does not vary with the pitch of the note played.

Yes, that is largely true. So, if that's what you'd like you could
simply not feed the proposed module with a CV. On the other hand,
there are cases where you might like the EQ to track. For example,
to create an instrument that covers a wide range of the keyboard, you
might be shifting from the resonant characteristics of a double bass
to a cello to a viola to a violin, no? Also, imagine a filter sweep
of 8 parallel resonant filters driven by an EG. Probably would be
interesting. Somewhat like flanging, but with much more control.

If the whole filter set was properly resonant, I think we get into
physical modeling territory pretty quickly. For example, one type of
physical modeling just requires an impulse fed into a delay with
feedback. As the delay changes, you get different pitches. This is
accomplished, basically, by a flanger which has a comb filter
response tracking at 1v/oct. The proposed tracking resonant EQ
should be able to emulate this nearly self oscillating "comb filter
response" as well.

I think the proposed module could be EXTREMELY interesting.