Slashdot Mirror
50 Year Old Quantum Physics Problem Solved
notsosilentbob writes "This story about a 50 year old unsolved Quantum Physics problem at Eurekalert.org is interesting, if just for the discussion about the computing power required (SGI/Cray machines).
Unlike the blowhard from BlacklightPower, this sounds like an important breakthrough. " The problem solved is that of the scattering effects of three charged particles. This is important, as this event occurs in everything from fluorescent lights to the ion etching of silicon chips.
This is the kind of story that I like to see. Just when we think there is nothing new to know and that unsolved equals unsolvable, someone cracks an enigma like this and shows that a new perspective is often the only thing required to make significant breakthroughs.
It's just amazing the amount of computing power it takes to solve some of these problems...
I wonder when games with physics engines are going to be able to simulate the universe to this detail?
100 years? 1000 years?
It's just amazing how far we have come since the dawn of the information age!
here and here.
_________________________
Plenty of people out there would cheer this breakthrough, not for its obvious worth as a furthering of scientific thought, but as a further entrenchment of quantum physics as a dominant theory for the mechanations of the universe, because frankly, it suits their personal philosophies of how the universe should remain somehow mystical.
Newtonian physics and its euclidean geometries is far too cold, too exact, too exacting. Bring on the theories that tell us we live in worlds of probabilities: I want to win the lottery, dammit. My ancestors read the tea leaves before me, and soon I'll have a nice quantum computer in a cup of coffee. How much can anyone truly know for sure? Certainly I don't know much, so give me a theory that says no one else can be much more certain. Now that appeals to my insecurities and warms my cockles.
It's quite fitting that such breakthroughs be made on the threshhold of a new era of unprecedented cultural return to mysticism. I'm still betting in science's corner, myself.
"If one is really a superior person, the fact is likely to leak out without too much assistance" -- John Andrew Holmes
http://www.sciam.com/1998/ 0698issue/0698gershenfeld.html a good scientific america article, basically a quantum particle can exist in more than one state at once (based on its probability of being in a certain state). the state is a way of saying that there are finite energy levels a particle (electron) can have, these can each represent a state. it also talks about action at a distance. all of this is interesting but unfortunately very hard to understand. its based on wacky math and probability functions. the ones and zeros are basically the same thing...a high level and low level of energy. this is just on a smaller level
I am trying to understand the importance of this discovery. Although the article mention the ionization process that lead to the grow of the flourenscent tubes, to the engraving of silicon chips, we have done all that WITHOUT understanding exactly how these things are done.
Can anyone tell me what this discovery for the "scattering problem" may yield, that is, apart from the Quantum Physics discipline?
Thanks in advance for any pointer.
Merry Christmas !!
Muchas Gracias, Señor Edward Snowden !
Unlike the blowhard from BlacklightPower, this sounds like an important breakthrough
You're editorialising again, Hemos! An assessment of the majority reaction to the Blacklight Power story might make it seem safe to do so on this occasion, but public opinion would change pretty quickly if Randall Mills was vindicated.
Mills' claims are certainly outrageous but he's only raised enough capital from hardened venture capitalists to fund his research, and is turning would-be investors away in droves. He's obviously not a fraud. Even his critics in the physics community don't deny he is at least sincere. And don't forget he appears to have a better grasp of maths, chemistry and physics than most people - he's not ignorant or even unqualified.
His enthusiasm for his own theory isn't really enough to warrant labelling him a blowhard. It's not as if he's gone around badmouthing everybody who disagrees with him. If you believed you'd made a breakthrough that would turn science on its head, wouldn't you have something to say about it? Would that make you a blowhard?
Don't get me wrong, I'm not jumping on Mills' bandwagon either (yet). But if his ideas were completely without credibility then he'd surely have been forced out of business by now. I think we ought to give him the benefit of the doubt until his work has been properly peer reviewed by people who are qualified to assess it.
Consciousness is not what it thinks it is
Thought exists only as an abstraction
I don't really see a problem with /. editorializing. I mean, I come here expecting a news service with some sort of humans behind it, and I get it. If I don't agree with it, I say so in the comments. Leave the plain facts to the news services /. links to - here, I want opinions to knock down!
Windows 2000: Designed for the Internet. The Internet: Designed for UNIX.
Quantum particles used to store information ("Qubits") can be either on, off, or they can be in superposition between on and off. It's sort of hard to explain, but what it basically means is it's BOTH on AND off at the same time. Not sorta anything. Very weird stuff. I like to believe the Universe is a little more organized than that, but who knows...
The streets shall flow with the blood of the Guberminky.
Is this the sort of thing which could benefit from distributed computing? Or is it one of those things, like protein folding computations, which have to be done on special, ultra-powerful machines?
This being Xmas, I was home having an argument with my father today about Canada's adoption of metric 30 years ago. He (age 53) is rather offended by this still today. I was trying to come up with ways to convince him that his personal discomfort was not enough reason to stay Imperial -- and now I've found one.
This discovery has nothing to do with metric specifically, and (rather amusingly) happened in an Imperial (and imperious, sometimes) country. But it's still representative: countries using common systems (metric) allow many to work together, across borders, to solve problems that we could not work out alone. In this case it was three American schools, but in other cases it has been schools or researchers from separate continents.
Yes, NASA messed up the metric thing. But that was based on one country not matching *all* the others, right? So imagine if this discovery is recanted in 3 weeks: "Oops, we were using inches and gallons, not centimeters and litres." This happens too often (even once is too often).
I'm not sure what my point is. I think it's a combination of "cool" and "why isn't everyone metric yet?"
Cam
- Cam MacLeod
Sir, this is in essence, better than a beowulf cluster. I do not wish to waste my time, or anybody else's- but a little explanation may be in order. A quantum processor would work in parallel with itself; checking as many possibilities as it has capacity for all at once- in a really short time. This could find the solution to an equation much faster by merely recognizing which state of the supposition is the correct answer rather than trying them all in sequence. You will not get a faster quantum processor- merely a bigger one. What would a cluster configuration do but split up the task and make the chips to talk to each other unnecesarily?
just my two cents
If you want to make a comment, put it in the forums with everybody else's and let moderation take it's course.
Thanks for the pointer.
Call me dumb as you must, but I do have difficulty connecting Quantum Computer with the solving of the "scattering" problem.
I thought someone have already prototyped some sort of "quantum" computer, before the "scattering" problem was solved.
That goes back to my original question - that we have done things like Flouresence tube and engraving chips with ion beams _before_ anyone have a definite answer to the "scattering" problem, and my original question is - what that discovery will yield for us, apart for making the Quantum Physicians feel much better?
Again, thanks for your pointer.
Muchas Gracias, Señor Edward Snowden !
My point is that perhaps the "discovery" of the "scattering solution" may not be yielding much practical effect, like the onet you have mentioned - distributed computing.
Perhaps the "discovery" itself may be used for predicting when and where the "scattering effect" may occur, and with the ability to predict, new branches of science may finally be able to mushroom.
Muchas Gracias, Señor Edward Snowden !
Beowulf clusters need lots of power. By understanding these particle interactions better, physicists will understand plasma physics better. Greater knowledge of plasma physics will lead to affordable power from fusion. Affordable power will reduce the costs of operating beowulf clusters.
Yeah, it's a stretch, but you asked for it.
seen it? hell I helped write it! ;)
_________________________
And just when I thought the Data Encryption Standard was absolutely uncrackable...
:)
OFTC: By the community, for the community
This result is interesting because previously this problem has been treated by using approximations. The many different "solutions" given by wildly differing methods did not agree - and the errors introduced by the approximations also were impossible to find.
Numerical methods are very good in that you know the degree of error. Increase the number of grid points, and the error will decrease... (but the computation time will increase accordingly.) This one fact means that the results produced are meaningful - they can be compared with experiment.
Now why are these scattering events interesting? Well there is a slightly more complicated collision where the incoming electron knocks out an electron - leaving the atom in an excited state. The excited atom then de-excites itself by emitting yet another electron. (Auger emission.) You can't do this with hydrogen (not enough electrons.) - However, the nobel gases work well...
This second type of collision is very interesting, in that the distribution of outgoing electrons is related to the Fourier transform of the wavefunctions of the electrons in the atom... You can "map" the distribution of an electron in an orbital with this technique. This in turn provides tests on the quantum theory...
This also happens in ionisation events that form Aurora.
"Would you like a cold drink with that Sir? Yes, yes, for the sake, of the future, of all mankind, I will have, a sm
If you feel that God's word is to be found in a book written by men rather than in the study of the miraculous universe around you, I feel sorry for you -- you're missing a lot of spiritual wonder and beauty.
While you might be true in saying that we ought to improve on the "Feel" thing - in sinking the 8-ball or in other endeavors - but please do not disregard the _importance_ of feel.
There are times I have done thing by "feel" alone, and those are the times I could have done extensive calculations and such, but there is always that little voice (call it instinct if you may) that tells me to go by "Feel" - yea, sounds like Obiwan's "Feel the force, Luke" thing, doesn't it?
I have tried to explain what "Feel" is, but I just can't. It's something you gotta have within yourself.
Anyway, Merry Christmas !!
Muchas Gracias, Señor Edward Snowden !
This is my own opinion - I personally do not think the journalist who wrote the piece actually gets it.
Most things that we have here, today, from gunpowder to electronic wonders, the ideas behind them all originated not from tweaking equations, but from intuition and inspiration.
Sometimes it requires "clicks" in the mind's eye to find a true "EUREKA!". Tweaking equations, IMHO, just doesn't make it.
After all, tweaking equations require _prior_ equations to exist, or there won't be anything to be "tweaked", right? And most of those prior equations owed their existence from the "clicks" of somebody's mind's eye.
Sorry, I've wandered to far out of topic. Gotta stop when I'm still able to.
Merry Christmas !
Muchas Gracias, Señor Edward Snowden !
I'm confused by this, how did they find an exact solution to the scattering problem if they are using a finite version of the wave function? Wouldn't that be an approximation of the true wave function, which extends to infinity?
Mea navis aericumbens anguillis abundat
Which means our intuition works only for day-to-day experiences. When you get to absurdly small or absurdly large things, you have to invent new reasoning methods, like relativity and quantum mechanics. What's really amazing is that mathematics is the tool that makes both possible.
Haven't been to the movies lately, right? Only 007 can beat 666!
..I just want to know if that damn cat is dead or alive.
Except that it tends to break down within schwartzchild singularities, and that problem's being worked on by people much smarter than I. My previous comment was much more facetious than others seem to recognize.
"If one is really a superior person, the fact is likely to leak out without too much assistance" -- John Andrew Holmes
Sorry, QM is successful because of its breathtaking predictive power. And if you're familiar with experimental results like the Aspect Experiment, it should be clear to you that no theory with the "common sense" deterministic appeal of Newtonian Mechanics can correctly mirror reality. Your social scientific explanations for theories confirmed again and again by empirical results are, as US people like to say, way off base.
--
Xenu loves you!
Get serious. No one cares about how many Barbies there are, for crying out loud. But Pokemon are important.
Yours Snorlax - WKiernan@concentric.net
The compute power required was large. They had to use Blue Pacific, probably the unclassfied machine, which has 1344 PowerPC604 CPUs. I wonder how much machine time was required, and how tightly coupled the computation is.
The result of this hack is an arbitrarily close approximation of the actual electron probability functions. In QED, you don't generally look for an exact prediction of the electron's location. Most of the time you are looking for a usefully accurate model. The breakthrough is finding a way to make the problem computationally tractable (which is done by the "large distance" approximation). Finding a way to calculate the large distance in terms of the near distance in all cases is the big deal here.
There are two difficulties with any Newtonian three body model (where gravity is the dominant force). The first is gathering complete information (where are all of the interacting objects). The second is computer errors including the position rounding error (at 32 bits, or whatever) and the sampling error (how often does the computer recalculate *all* of the vectors based on updated positions?). Ballistic models that describe interacting particles in terms of probability functions can be much more successful, but run into difficulty during interpretation (the electron really can be in five different places, the spaceship cannot).
By reducing distant bodies to planar gravity fields (large distance approximation), we end up with spacecraft like the Galileo probe that made it to Jupiter with only a few small course corrections to make up for the slight inaccuracies in the approximated model. But beware, it's still just a useful model. Don't expect to hit the center ring halfway across the solar system with your eyes closed based on any model. You'll need to correct (or update) your model with empirical data to make it actually work.
So, to finally answer your question. The breakthrough is going the other way (from Newtonian three body to QED). Before this, however, the QED models didn't have any way to reduce the large distance wave functions to a useful approximation. Now they do. If the large distance approximation used can be applied or extended to more complex interactions, our models of quantum interactions will be dramatically improved and our ability to describe complex probabilistic events will become correspondingly more confident.
Regards, Ross
If you look again at the article, you'll see that the submitter used the 'blowhard' comment, not Hemos. If anything, Hemos is guilty of not editing out a derogatory comment.
It's been more than 10 years since the original Pons and Fleischmann anouncement of [sic] cold fusion. These guys (Blacklight) are one of the myriad branches from that bizarre root. Personally I find it amusing that they deny a relationship with cold fusion, while the cold fusion advocates point to them as a success story.
Because the Qubit is can be "in between" on and off, it would allow one Qubyte to store every number between 0-255, or just a couple of them. This way, you could multiply all the numbers in one by another qubyte, which on a 32-bit quantum couputer would allow you to multiply up to 4294967296 numbers in a single opperation, although you would have to use every number between one and 4294967296 to get that many. As you can see, a working quantum computer would blow conventional computers out of the water. The only problem is when your working this small, just about anything could interfere with the processor, and as far as I know the most advanced working quantum coputer can and & or two bits together.