Brett Lorenzen writes:
> Bob Munck wrote:
> > > "A single silicon transistor today can only be seen in a microscope. In
> > > a few years, it will take a microscope to see an entire chip of
> > > transistors. ... In 1950, a transistor
> > > cost $5. Today, it costs one-hundredth of a cent. In 2003, one
> > > transistor will cost a microscopic nanocent.
> >
> > This is typical mass media technical ignorance. Chips are not
> > shrinking in any significant way, and are not at all likely to
> > become microscopic. Where would you attach the pins?
I don't read the above as saying that CHIPS are shrinking, but
that transistors are; allowing for the usual technical vagueness of
mass-market writing, you can translate this to "chips are getting more
and more function packed into them for less and less price, and in the
future they'll have even MORE and more function, for even LESS and
less price."
> > If transistors cost one-hundredth of a cent, a 64Mb RAM would
> > cost $6,700.00; on the other hand, if single transistors cost
How many transistors will $5 buy you, on chips purchased in bulk
(i.e. by the millions to build into consumer electronics)? Over 5000?
I'm certainly not interested in defending this guy, nor even in
defending mass-market writing, but try and think about what the writer
is trying to convey before miscontruing it. The writer might have
been better off writing this as "In 1950 a single transistor cost $5,
today $5 will get you XXXXXX transistors on a single chip." It might
have had the same impact or it might not.
> Not only arithmetic, but problems with physics and economics too. If we
> make chips so small that they approach particle size, they will be
> incredibly vulnerable to failure when struck by alpha particles and
> other stray oddities of physics.
In point of fact, leading edge chip design today now tries to
*take advantage* of quantum effects, instead of just trying to work
around them.
> Chips have to reach a physical peak in size for this reason alone, not
> to mention heat limitations and a host of other things.
Certainly true, although it's important to note that every time
in the last 10 years somebody said we were about to hit the wall
because of some physical limitation or another, chip designers have
always found a way around the anticipated limitation. I'm not saying
they always will, but it certainly makes it more of a possibility
that they'll find a way around the next obstacle.
> Cost curves don't drop exponentially over time, they peak as they
> bruch up against the walls of physical realities and production
> processes address new challenges.
But they also drop as the production scales up; or as one VP of
Mitsubishi Heavy Industries put it, "Tell me what you want it to cost
and I'll tell you how many you have to order." Certainly there is a
hard limit, somewhere; but I doubt we'll hit it soon. Maybe when:
we have chips that have pushed to ALL of the hard boundaries of
both physics and advanced algorithm design (which will not be for some
time),
AND those chips are being mass-produced as much as possible,
AND we've explored all of the possible options for
mass-production mechanisms,
AND we've built the most scaled up production plants possible (*)
Or maybe not.
(* Note that after many years, LCD screens started getting larger
- and cheaper in the last three or four years, after the various
efforts to build more advanced, larger scale production facilities
finally came to completion. And the bleeding edge of such things is
already much farther along since the foundations were laid for those
plants.)
Steven J. Owens
[EMAIL PROTECTED]
____________________________________________________________________
--------------------------------------------------------------------
Join The Web Consultants Association : Register on our web site Now
Web Consultants Web Site : http://just4u.com/webconsultants
If you lose the instructions All subscription/unsubscribing can be done
directly from our website for all our lists.
---------------------------------------------------------------------
