[sdiy] Walsh Generators
Harry Bissell Jr
harrybissell at prodigy.net
Fri Mar 18 18:46:54 CET 2005
even more naive... if you were to make a walsh
generator with equivalent 16 bit accuracy, what
would you produce with it. A sampler is a far more
effective way of specifying the waveform.
Do you really mean to figure out all the coefficients
by hand ? What about time-varying envelopes etc...
walsh functions (to me) scream FPGA, because the
essentially '1 bit' nature of the process is ideal.
The multiplications are very easy as well... but as I
posted earlier (and got no good answer)... what is the
'magic bullet' in the operating system. The one, two,
or ten parameters that define an set of musically
useful sounds ???
(not just being a wise at ss here... but fourier
synthesis
has not worked yet, imho. neither will walsh until
this issue is addressed)
H^) harry
--- Barry Klein <Barry.L.Klein at wdc.com> wrote:
> Naïve question - can't this stuff be accomplished in
> software?
>
> -----Original Message-----
> From: owner-synth-diy at dropmix.xs4all.nl
> [mailto:owner-synth-diy at dropmix.xs4all.nl] On Behalf
> Of Glen
> Sent: Friday, March 18, 2005 9:20 AM
> To: music.maker at gte.net; synth-diy at dropmix.xs4all.nl
> Subject: Re: [sdiy] Walsh Generators
>
> At 12:07 PM 3/18/2005 , Glen wrote:
>
> >I wonder if several of these chips could easily be
> ganged together somehow,
> >for better quality results?
>
> I just did some calculating, and it seems that it
> would cost about $1280 to
> achieve an SNR equivalent to a good cassette tape
> deck (66dB), assuming
> that those chips could be easily ganged together.
> Surely, the situation
> isn't that bleak? Please, show me that I've made a
> terrible mistake in my
> estimate somewhere.
>
> The assumptions that I made for that estimate:
>
> 1) $20 per 31-function Walsh chip
>
> 2) You get a 6dB boost in SNR, for every DOUBLING
> of the number of Walsh
> functions. (I think that was on Neil's site,
> somewhere.)
>
> 3) These Walsh chips could simply be added
> together for higher levels of
> Walsh functions.
>
>
> thanks,
> Glen
>
>
More information about the Synth-diy
mailing list