Just for the record, I picked 8 bands out of the blue, I wasn't
referring to the PAIA vocoder per se. Although didn't someone write that
when put to the test the PAIA only has 3 or so "actual" bands, I seem to
recall reading that somewhere. Although this does help me to understand
the expense and labor required. -Nate
jhaible wrote:
>
> > just a stupid question, is there some internal link between bands on
> a
> > vocoder? i'm asking because i'm wondering if 2 8 band vocoders = a
> 16
> > band vocoder or if it doesn't work that way. -Nate
>
> No, ∗this∗ does not work at all, for obvious reasons.
> But even when you double the number of bands and adjust the filter
> frequencies accordingly, your efforts will have to be way more than
> doubled.
> As a (very rough) rule of thumb, doubling the number of channels means
>
> 3 things:
> (1) Doubling the number of channels (obviously)
> (2) Making the filters more selective / small band -> double the
> amount
> of circuitry for each filter
> (3) More selective -> need lower tolerance components or larger, less
> sensitive filter topologies -> more expensive components, or
> more
> components
> These 3 factors ∗multiply∗. So as a rule of thumb, the effort for the
> filters
> will increase more than quadratic, less than cubic.
> Number of channels x 2 => effort x 4 ... x 8
>
> There ∗are∗ other things to consider, but now you get an idea while
> an 8-channel Paia can be so cheap and a 20-band Sennheiser or
> 22-band EMS 5000 are more than just 3 times as expensive.
>
> JH.
>
>
> -----------------------------------------------------------------------
> [GO.com. Click to find what you're looking for]
> -----------------------------------------------------------------------
>