[clug] Arlec surge protector
anacrolix at gmail.com
Thu Mar 18 19:36:31 MDT 2010
What about our hardware? I'd say we're about Sentient Primate 2.3 or so,
some of our ancestor species were 2.1, 2.2 etc. Neanderthals were a recent
fork mounted on a heavier chassis. More distant relatives such as the
chimpanzees are early prototypes of 1.0, and have serious performance
issues. Pretty good hardware to work with though, reboots are decades
between, or never happen on the average, etc. etc. =P
I really like the Cluon metric, very amusing and appropriately in favour of
the technically minded.
On Fri, Mar 19, 2010 at 8:00 AM, steve jenkin <sjenkin at canb.auug.org.au>wrote:
> Andrew, an authoritative response. Thanks.
> You've nailed the 'input' side of the equation - but what about 'outputs'?
> I doubt I have the 'intellectual horsepower' to do your work - and most
> definitely not the aptitude.
> I suspect that human brains operate within a reasonably small
> power-consumption range, but their results/outputs (depth, breadth,
> speed, originality) vary wildly, hence the many attempts at "IQ" metrics.
> Is it at all possible to talk about "thunks" as completed tasks, not
> just describing the inputs? It might be that it's impossible.
> Andrew Janke wrote on 18/03/10 9:42 AM:
> >> In the world of Physics, there's Energy/Work and Power (Energy per unit
> >> time), Joules and Watts.
> >> In computing we have MIPS (work per unit time), so what's the equivalent
> >> of Joules??
> > Probably the closest is rate of ATP uptake (for the brain). This is
> > difficult to measure though and instead is generally measured as glucose
> > uptake which is in turn converted to ATP in the mitochondria in the
> > post-synaptic density. This can be measure using PET by tagging some
> > of the this glucose (2-deoxyglucose) and measuring the amount of it
> > you injected as
> > compared to the amount that is still in the persons bloodstream vs the
> > signal from the brain.
> >> And how does that relate to 'thunks' by people??
> > And here is the problem. The brain as a whole has a relatively
> > constant energy metabolism, it goes down slightly when we go to
> > sleep and up a bit when we wake up but beyond that nigh on constant.
> > This is mainly as the brain has a heck of a lot of maintenance work to
> > do just to maintain normal function:
> > * cleaning all the wires
> > * maintaining old memory circuits and keeping them firing
> > * keeping the insulation healthy
> > * keeping voltage potentials good in case they need to fire
> > * doing all the subconscious unimportant things like breathing,
> > heart rate monitoring, temperature regulation, endocrine function
> > * ...
> > The support cells are called Glial tissue and do everything in the
> > brain except conduct sparks.
> >> Is there work done on measuring/reporting the computational 'cycles' of
> >> brains on standard tasks??
> > Absolutely, this is done with either PET or fMRI and are those pretty
> > images you see in journals from time to time with "red blobs on
> > brains". _Quantifying_ the effect is difficult as all you are
> > measuring in the fMRI case is a relative increase in blood flow/volume
> > in the venous capillary bed that is next to a bunch of neurons. But
> > certainly ratios or size of effects can be measured.
> > This method of measuring the amount of work has been been used (for
> > example) in an
> > attempt to differentiate people in early age who will go on to get
> > dementia. The theory is that their brains have to work "harder" to
> > figure problems out early on.
> > So sad to say, no you probably can't loose weight by just "thinking".
> >> We will now resume normal programming... [Silence] :-)
> > Good good.
> > --
> > Andrew Janke
> > (a.janke at gmail.com || http://a.janke.googlepages.com/)
> > Canberra->Australia +61 (402) 700 883
> Steve Jenkin, Info Tech, Systems and Design Specialist.
> 0412 786 915 (+61 412 786 915)
> PO Box 48, Kippax ACT 2615, AUSTRALIA
> sjenkin at canb.auug.org.au http://members.tip.net.au/~sjenkin<http://members.tip.net.au/%7Esjenkin>
> linux mailing list
> linux at lists.samba.org
More information about the linux