Subscribe to RSS feed
Dave Bacon is a theoretical ski bum who is also a pseudo 
This glib article from the Wired Blog Gadgets Lab discusses some of the "crazy" ideas for building computers. Among them, of course, are quantum computers, which means, of course that a quantum computing bastardization, can't be far from behind.
Let's begin at the beginning:
Quantum ComputersOkay, so we won't nitpick on whether D-Wave really has a "quantum computer" or not, and whether D-wave's demonstration of solving a small Sudoku is at all impressive, but we might like to point out that playing a general version of Sudoku is computationally probably very hard. But as far as things normally go in articles on quantum computers, this isn't such a bad beginning.Sudoku. That's all D-Wave's quantum computer is good for right now, and even then they wouldn't let us hacks see it in the flesh.
By lining up subatomic particles to encode information in a manner similar to the binary data found in conventional computers, such computers create "quantum bits," subject to the strange workings of quantum physics.
Then things get whacked:
The payoff is that calculations get done without any old-school relativistic inefficiencies getting the way: less heat production, less power consumption, more grunt.Maybe I'm not parsing this right: "relativistic inefficiencies?" Heat production, less power consumptions, more grunt has, as far as I know, absolutely no obvious connection to relativity. And certainly the benefit of quantum computers is not less heat production, etc, but that quantum computers can perform some computational tasks more efficiently than classical computers. Of course quantum computers are reversible computers, and there are claims about such computers having less power consumption (less heat production seems to me to be more of a pipe dream: external noise creating entropy in a system would have to be completely eliminated which seems to me to be a pipe dream.) But it certainly isn't the main motivating factor behind the quest to build a quantum computer.
Physical Science
Comments
There is a simple equation:
quantum computer = magical superhypermegadevice = write whatever you want using 'simultaneously', 'parallel universes' and 'brain'
Compare the descriptions in my Quantum Computing Sci-Fi list and some of the recent press articles - aint no difference.
Posted by: nextquant Blog | November 15, 2007 1:16 PM
Add "consciousness" to that list, too, I think.
Posted by: Blake Stacey | November 15, 2007 1:20 PM
"Magical superhypermegadevice" made me laugh.
Posted by: Dave Bacon | November 15, 2007 1:22 PM
Apropos 'consciousness'...
"There's no blood in those quantum veins".
from the movie Stealth (2005), just added to the list
Posted by: nextquant Blog | November 15, 2007 2:08 PM
Quote:
"Maybe I'm not parsing this right: "relativistic inefficiencies?" Heat production, less power consumptions, more grunt has, as far as I know, absolutely no obvious connection to relativity."
It's elementary... see as information gets moved around close to c, the light of knowledge that is released from the computations is red-shifted into the infra-red... generating heat. Furthermore, it's well known that knowledge is an Einsteinian manifold, and the calculations are increasing our knowledge-- thus we have an expanding info-space-time continuum which further imparts a cosmological red-shift to the data... Not coincidentally, this datacalogical red-shift was responsible for a redder vote, thus resulting in a Republican win for the last two elections.
And finally, you shouldn't be surprised by D-wave's reluctance to reveal their quantum computer. Currently it's in a state of superposition, and if someone actually observed its state, there is a non-trivial probability that the construction will cease to be a quantum computer.
Posted by: peter | November 15, 2007 6:30 PM
this datacalogical red-shift was responsible for a redder vote, thus resulting in a Republican win for the last two elections
Which is exactly why I keep telling to people to slow down. If everyone would just chill and relax, the Republicans would surely be out of office in no time at all (according to at least one person's inertial frame.)
Posted by: Dave Bacon | November 15, 2007 6:32 PM
ooph.
The latter section is pretty rough, but I don't quite get why you gave a pass on the opener. It seems just as silly as most other popular press articles.
"By lining up subatomic particles to encode information..." First of all, in the vernacular this immediately makes people (the lay population that is) think of nuclei, quarks, and the like, rather than the electrons that the author is presumably speaking about. But is the claim reasonable in general? In solid-state QC, we do, in fact, deal with single electrons - ok. But what about macroscopic quantum coherence in superconductors? Or trapped atomic ions? Are atoms sub-atomic? I just don't see how the generalization works.
Dave, your outrage seems to be diminishing - has D-Wave burned you out?
Posted by: Michael J. Biercuk | November 16, 2007 8:35 AM
Doh, I didn't even notice the "subatomic particles." My mind parsed "atomic" since no one in their right mind would write "subatomic."
My outrage grows everyday. But it started at a low spot because I'm a ski bum.
Posted by: Dave Bacon | November 16, 2007 11:17 AM
If only I were a ski bum...lucky you!
I actually see the "subatomic" references a lot, and have no idea where they originated.
I've also seen descriptions even more infuriating claiming that the quantum mechanical phenomenon of tunneling allows particles to move instantly, hence providing quantum computers with speed advantages over classical microprocessors.
That one almost made me fall out of my chair.
Posted by: Michael J. Biercuk | November 16, 2007 12:40 PM
They lost me at the assumption that the same number can be either a prime or a composite, depending on the velocity of the observer.
Bad enough that two particles can be either entangled or not entangled, depending on the velocity of the observer.
Or that a particle can be entangled with itself in the future or the past.
So, is this another of the weird forms of proof: proof relative to an observer moving at the right velocity for the theorem to be true, even if it is not true relative to the referees?
Posted by: Jonathan Vos Post | November 16, 2007 2:26 PM