Whoops, we're drifting off-list. I've included the relevant parts (and committed a stereo fix to the main repo that fixes the bug from my second mail).
On Mon, Mar 4, 2013 at 6:46 PM, Kurtiss Hare <[email protected]> wrote: >> Also, couldn't having a large frequency span be challenging to carry >> over FM? I should probably grok >> http://en.wikipedia.org/wiki/Frequency_modulation before discussing >> that detail further. > > My naive model of the idea here is to consider the number of audibly > distinct (and computationally discernible) instruments capable of delivery > through FM. For a given note, each one has a distinct overtone series which > lends it a unique timbre. Depending on the sophistication of your decoder, I > would think the number of euphonic sounding configurations to be quite high. Well, instruments get their timbre from the *weighting* of the overtones, not their existence or non-existence. I'm concerned that if we go into the game of actually weighting these overtones (rather than just choosing whether they should be there or not), decoding becomes too difficult. We're already spanning three octaves as it is, so they'd definitely need to have less amplitude than the fundamentals. I'll do some experiments when I get the time... JC -- Too many emails? Unsubscribe, change to digest, or change password by emailing moderator at [email protected] or changing your settings at https://mailman.stanford.edu/mailman/listinfo/liberationtech
