Re: oversampling
- To: KC5TJA <kc5tja@xxxxxxxxxxxxxxxxxxxxxx>
- Subject: Re: oversampling
- From: Eugene Leitl <eugene@xxxxxxxxxxxxxxxxxxxxxxx>
- Date: Wed, 7 Jan 1998 12:08:59 -0400
- cc: Penio Penev <penev@xxxxxxx>, MISC
- In-Reply-To: <Pine.LNX.3.96.980106104413.14993A-100000@topaz.axisinternet.com>
- Old-Date: Wed, 7 Jan 1998 19:03:41 +0300 (MSK)
- ReSent-Date: Wed, 7 Jan 1998 11:29:08 -0500 (EST)
- ReSent-From: Penio Penev <penev@xxxxxxx>
- ReSent-Message-ID: <Pine.SGI.3.96.980107112908.25969W@venezia.rockefeller.edu>
- ReSent-To: MISC
On Tue, 6 Jan 1998, KC5TJA wrote:
> [...]
> Imagine how much more efficient our nervous system would be if we used
> soliton waves instead of simple spikes... :)
Be careful with assumptions about efficiency of biological systems. We
have been underestimating them for centuries. Can _you_ make
deep-submicron liter-sized defect-tolerant nonbrittle systems which work
for many decades, delivering at least 10^18 Ops @ 20 W?
Seems quite a lot of quite unique specs, eh?
> [...]
> Does anyone know what network topology our brain uses? :)
In parts of it a (polarized) statistical hypergrid/hyperlattice, with an
outwards increasing level of defectivity.
ciao,
'gene