Re: Memresistor & Audio

On Thu, 17 Jul 2008 23:51:05 -0400, Ben Bradley
<[email protected]> wrote:

> I always found this article interesting and informative (and check
>out the dates of the references, the idea was around long before this
>app note):
>http://www.national.com/an/AN/AN-804.pdf

Very cool stuff. Hadn't heard about Bart (Mr. Golden Age JBL)
Locanthi's work in this field before. Giants walked the Earth
back then.

I could quibble about the bottom of page 3 and into the top of page 4,
where it seems that "here a miracle occurs" is implicit, but maybe
I'm just not a good enough reader. That sure happens.

To my small understanding, the really cool bit couldn't be
shown on their diagram - it would only happen averaged over
very many samples. I'm bitchy about this because it's an issue
that got me down a lot of wrong paths personally. And it's
something that so many folks get so wrong so often, even
in 2008.

Dither doesn't actually "remove" *any* quantization errors.
It only spreads them around in time. Dither isn't added to
signal (well, yes, of course it is...) - rather, it's added to
time. Or, *over* time - dither applied to a single sample
wouldn't help at all. It would just be a noisy sample.

Obviously this isn't Gospel; it's just a way of approaching
the subject.



> As for this "memresistor," I'm not convinced that it's the "fourth
>fundamental component of electronics"

If Maxwell somehow missed it, I'd be surprised. And deserve
the Silver Hammer. But somehow I feel confident that nothing
that fundamental lurked in the bowels of physics long enough
to be named a "memresistor". Yikes.


Much thanks, as always,
Chris Hornbeck