Re: [oc] Beyond Transmeta...

[Lars Segerlund <lars.segerlund@comsys.se>]

  I just had to comment :-) ..... read with toungue in cheek, ( if you 
can do that and still read ;-) ).


Niclas Hedhman wrote:
> I can't help myself....
> 
> On Thursday 13 February 2003 16:55, Lars Segerlund wrote:
> 
>>  Brains consume about 300 Watt's of power on standby, and somwhere
>>around 1500+ Watt's at full 'power'.
> 
> 
> But I heard the energy in peanut can keep the brain going for 2 hours.
> But considering the 100Hz clock, even that it sounds a bit high ;o)
> 
> Also tried running the brain on AAA batteries, but DC-Glucose converter still 
> has some quirks and doesn't function.
> 
> ;o)
> 

  Figure out the thermal efficiency of the process :-) .... brain sucks !

  However the peanut thingy seems to be correct, I ate two peanuts, and 
my brain continued to work for two more hours ! ( seriously, this info 
is plain wrong ).

> 
>>, the wrinkels on the human brain are
>>cooling fins, and the brain has a rather elaborate cooling system, also
> 
> 
> Thought the wrinkles on the brain was related to the wrinkles in the 
> forhead/face, but more distinct, due to its proximity. Theory; 
> 
> frequency * wavelength = speed in medium,
> 
> hence if wavelength (standing wave) ~1cm and frequency 100Hz, the speed in the 
> brain  would be around 1 meter per second, compared to the average computer 
> at  200-300 million meters per second.
>

  This is roughly correct, the responce time of a brain is around 0.5 s 
in the best case,
  however it is good at hiding it's own fallancies.

> 
>>it uses more than 7 'neurons' for each bit it represents and more in the
>>interconnect, I don't think it has got a low power consumtion at it's
>>rated operating frequency of around 100 Hz. Furthermore, it's stressed
>>to the limit most of the time it's operating, it does seem to have
>>someting better than a alfa-beta prunin though as it still manages to
>>function :-).
> 
> 
> "Stressed to the limit", hmmm?? I thought it is mainly inoperable, idle and 
> "rusty".
>
  No, now youre mixing up cause with effect ;-)
> 
> Niclas
> --
> To unsubscribe from cores mailing list please visit http://www.opencores.org/mailinglists.shtml
> 

--
To unsubscribe from cores mailing list please visit http://www.opencores.org/mailinglists.shtml