Forgot your password?
typodupeerror
Supercomputing Science

Supercomputer Adds Credence to Standard Model 120

Posted by ScuttleMonkey
from the start-watching-for-vogons dept.
ScienceDaily is reporting that researchers at the University of Edinburgh and Southampton in cooperation with partners from Japan and the US have shed some light on the Standard Model of physics using a new computer model. "The project's enormously complex calculations relate to the behavior of tiny particles found in the nuclei of atoms, known as quarks. In order to carry out these calculations, the researchers first designed and built a supercomputer that was among the fastest in the world, capable of tens of trillions of calculations per second. The computations themselves have taken a further three years to complete. Their result shows that the Standard Model's claim to be the best theory invented holds firm. It raises the stakes for the riddle to be solved by experiments at the Large Hadron Collider at CERN, which will switch on later this year. Physicists' efforts to confront Standard Model predictions using the most powerful computers available with the most precise experiments offer no clues about what to expect."
This discussion has been archived. No new comments can be posted.

Supercomputer Adds Credence to Standard Model

Comments Filter:
  • Boycott ScienceDaily (Score:5, Informative)

    by jnana (519059) on Friday February 29, 2008 @09:33PM (#22607258) Journal

    I wish people would stop posting crappy science articles from ScienceDaily and related sites.

    From this article, we learn that computer modeling confirmed something "about the behavior of quarks". That's it. There is nothing of substance in the article other than this and that the computation took three years.

  • Higgs (Score:5, Informative)

    by Lord Byron II (671689) on Friday February 29, 2008 @09:44PM (#22607310)
    Before we claim that the Standard Model is the end all of particle physics, lets see if we can find the Higgs Boson. Afterall, Fermilab has come very, very close, so the LHC should be able to seal the deal.
  • by Vellmont (569020) on Friday February 29, 2008 @10:05PM (#22607440)
    Rather than "they used a supercomputer to do physics"

    http://www.bnl.gov/bnlweb/pubaf/pr/PR_display.asp?prID=08-x5 [bnl.gov]
  • by wizardforce (1005805) on Friday February 29, 2008 @10:08PM (#22607454) Journal
    it's much worse than that, the article was pretty much mirrored from the source university of south hampton article here: http://www.soton.ac.uk/mediacentre/news/2008/feb/08_31.shtml [soton.ac.uk] which has absolutely nothing to add on the subject. three years of work and they don't even say what it is that they were modeling... what exactly was the point? perhaps a better article is required like the one here: http://www.physorg.com/news121963192.html [physorg.com]
  • TFA - enjoy... :) (Score:2, Informative)

    by djupedal (584558) on Friday February 29, 2008 @10:11PM (#22607472)
    Supercomputer Confirms Standard Model Theory Of The Universe, Deepens Puzzle

    ScienceDaily (Feb. 29, 2008) -- Scientists have used a supercomputer to shed new light on one of the most important theories of physics, the Standard Model, which encapsulates understanding of all the material that makes up the universe. This 30-year-old theory explains all the known elementary particles and three of the four forces acting upon them - however, it excludes the force of gravity, which is its shortcoming.

    Physicists have been trying to find the missing pieces in the jigsaw that would extend the Standard Model into a complete theory of all the forces of nature. However, the landmark findings by researchers at the Universities of Edinburgh and Southampton, and their partners in Japan and the US, confirm the Standard Model to even greater precision than before, deepening the puzzle.

    The project's enormously complex calculations relate to the behaviour of tiny particles found in the nuclei of atoms, known as quarks. In order to carry out these calculations, the researchers first designed and built a supercomputer that was among the fastest in the world, capable of tens of trillions of calculations per second. The computations themselves have taken a further three years to complete.

    Their result shows that the Standard Model's claim to be the best theory invented holds firm. It raises the stakes for the riddle to be solved by experiments at the Large Hadron Collider at CERN, which will switch on later this year. Physicists' efforts to confront Standard Model predictions using the most powerful computers available with the most precise experiments offer no clues about what to expect.

    Professor Chris Sachrajda of the University of Southampton's School of Physics and Astronomy said: 'Modern supercomputers and improved theoretical techniques are allowing us to explore the limits of the Standard Model to an unprecedented precision. The next stage will be to combine such computations with new experimental results expected from the Large Hadron Collider to unravel the next level of fundamental physics.'

    Professor Richard Kenway of the University of Edinburgh's School of Physics added: 'Although the Standard Model has been a fantastic success, there were one or two dark corners where experimental tests had been inconclusive, because vital calculations were not accurate enough. We shone a light on one of these, but to our enormous frustration, nothing was lurking there.'

    The research, published in Physical Review Letters, was supported by the Science and Technology Facilities Council.

    Adapted from materials provided by University of Southampton [soton.ac.uk] .

  • If it has to do with quarks and it takes a supercomputer, I'd guess a lattice QCD simulation [wikipedia.org].
  • by ortholattice (175065) on Friday February 29, 2008 @10:22PM (#22607514)

    I wish people would stop posting crappy science articles from ScienceDaily and related sites.
    I've found a better site to be http://www.eurekalert.org/ [eurekalert.org] which is run by the AAAS (American Association for the Advancement of Science) and has less annoying ads. A very high percentage of ScienceDaily stories - although oddly not this one - are the same as those on Eurakalert, but Eurakalert seems to have them first (at least based on RSS feed). I think Eurakalert also provides the original press release from the university/organization - not a watered-down, clueless-journalist-rewritten "adapted from materials provided by [university/organization]" - and also gives the link to the actual "materials", usually not provided by ScienceDaily.
  • by l2718 (514756) on Friday February 29, 2008 @10:27PM (#22607538)

    Gravity -- certainly the weakest force -- is completely irrelevant as far as the physics of elementary particles is concerned. In real life there is no way to observe any kind of gravitational interactions on the scales where the other forces are relevant. In particular, if there is physics just beyond the standard model it need not have any connection to gravity. It's true that gravity is relevant on extremely large scales, but for these scales we have perfectly good theories (GR; in fact Newtonian gravity is quite sufficient in almost all cases). You'd have to go to Planck scale before there'll be any guarantee of gravitational effects playing a role.

    This is not to say that a quantum theory including gravity is not an important goal of theoretical physics, it's just to say that so far we have not found any real-life situations where such a theory would be needed, that is when corrections due to quantum gravity would play any role whatsoever. Hopefully the LHC will probe the physics beyond the standard model. The number of orders of magnitude between the energy scales we can actually observe and the quantum gravity energy scale make it extremely unlikely, however, that gravity will be relevant to experimental fundamental physics for many millenia.

  • by jnana (519059) on Saturday March 01, 2008 @12:23AM (#22607988) Journal

    If you can't observe the phenomena in the real world, then how do you know the model has any correspondence?

    The whole point is that many phenomena are observable, and predictions by the model have been verified again and again. Those that cannot currently be verified may be verified in the future, and if they are falsified, that tells us that one of the simplifications that was made in order to create the computer program was not warranted or that there is some factor that our program failed to incorporate that turned out to be important.

    I would suggest you re-read my post and consider this phrase:

    ...cannot expand (or confirm) the frontiers of [scientific] research...

    The more successful predictions about new kinds of phenomena, particle interactions, etc. that we have, the more confidence we have in the theory. Colloquially speaking, those successful predictions help confirm the theory.

    I absolutely agree that you can use computer models in engineering. But the computer model showing a race car has less drag than a Hummer isn't expanding your knowledge of fluid dynamics. It's allowing you to apply what you already have established as the rules to different situations.

    It's only engineering if we actually make the cars. When we run the model, it is just as much science as the experiment under consideration. The racecar/hummer example may not expand our knowledge, but you also said that it can't offer confirmation and has no relation to reality or truth. The important point is that modeling that makes new predictions about unseen phenomena that turn out to be correct is part of the process of moving from untested hypotheses to established science.

    It will not allow you to prove or disprove new rules for fluid dynamics.

    Now you're changing your story from "computer models cannot confirm ..." and "models have no relation to reality or truth" to "cannot prove or disprove new rules", whatever that is supposed to mean.

    I get the feeling that I'm being trolled, since you apparently believe that all computer programs/models relating to the standard model and presumably general relativity and anything else that is not totally-settled-forever-science are absolutely worthless and have no relation to reality.

  • by dario_moreno (263767) * on Saturday March 01, 2008 @09:13AM (#22609408) Homepage Journal
    the point is that 99% of physical equations are non-linear, N-body, ordinary or partial differential equations and thus do not have an analytical solution. So the only approach to check if as you say the hypothesis are correct is numerical. Even the great Fermi had to recur to an army of mechanical calculators staffed by humans to see if his equations had a meaning. Since Galileo mathematized physics there have been 300 years without computers so people (and teaching up to nowadays) are used to approximations and analytical solutions because it gives "beauty" and beautiful exercises ; but numerical analysis if properly taught can also be beautiful and give many insights.

UNIX is many things to many people, but it's never been everything to anybody.

Working...