Okay, I guess I better write an insulting and condescending post on this
analog computing thread....

So consider yourself insulted. =P

Now for the analog computing business...

Analog computing is interesting cuz it's kinda neet and is much more
intuitive than many of the ideas in signal processing and the nyquest
theorem and all that...

Analog computing hung on for a long time because it was much cheaper and
more practical for a variety of high-performance problems that didn't
need the level of precision and repeatability that can be obtained with
digital solutions.

The brain seems to occupy a knife's edge between analog and digital in
many respects. Neurons seem to use some kind of pulse code modulation.
The spikes are discreet (digital-ish) but they can fire anywhere on a
time continuum (analog-ish). We see rate coded and time coded neurons in
the brain and probably a few other types.

Since 99.9% of everything can be usefully approximated using digital
techniques, what then does analog actually contribute to the mix?

The only thing that appears to truly come out of analog computing is
that a stimulus signal can enter the brain at any time and immediately
trigger a cascade of neural activity. In a digital system, where both
time and space are discreetized down to some specified resolution, any
event  arriving out of sync will have to be sampled at the next update
cycle of the system.

In any event, I see no reason why AGI can't be done, well enough, with a
digital system.

-- 
IQ is a measure of how stupid you feel.

Powers are not rights.



-------------------------------------------
AGI
Archives: https://www.listbox.com/member/archive/303/=now
RSS Feed: https://www.listbox.com/member/archive/rss/303/21088071-f452e424
Modify Your Subscription: 
https://www.listbox.com/member/?member_id=21088071&id_secret=21088071-58d57657
Powered by Listbox: http://www.listbox.com

Reply via email to