[sdiy] Walsh Generators

[Glen]

At 03:25 PM 3/18/2005 , Neil Johnson wrote:

>I think you are confusing sample rate with resolution.  You will see 
>that the sliders (amplitude) can be moved in very small steps.  But 
>there are only 16 samples.  So if you wanted to record this at 44.1kHz 
>sample rate this would produce a sawtooth of 2.75kHz.

Yes, confusion is the key word for me at this point.  :)

I'll have to read through everything a couple more times--at least.
Additive synthesis using sines is much easier for me to understand the
usage of. After reading a few brief articles about Walsh synthesis, I was
intrigued about whether Walsh was considerably easier for digital
*hardware* to handle.

Fourier synthesis with sines does seem easier for an end-user to understand
and manipulate. If Walsh offered a much more efficient hardware synth
engine, I was thinking it might be worth considering using Fourier for
non-realtime analysis (perhaps even using existing Fourier data) and then
converting that Fourier data into a format that a Walsh hardware synth needed.

I was thinking of Walsh synthesis as a replacement for a Fourier synthesis
engine on a hardware platform, but using Fourier-style analysis to produce
the raw data that eventually was "tweaked-to-taste", converted to "Walsh
format", and then fed into the Walsh hardware synth. (I hope that makes
sense.) 

I wasn't thinking of working with a pure Walsh system--from initial sound
analysis, through manipulation, to final sound generation. There might be
some benefits in a system that worked that way, but that is probably
totally uncharted territory at this time.


take care,
Glen