Comments by "Lepi Doptera" (@lepidoptera9337) on "Newton’s three-body problem explained - Fabio Pacucci" video.

  1. That's why the folks in the physics department are looking down on the folks in the philosophy department. ;-)
    2
  2.  @GregValentine  In what way? Please explain in detail. This physicist is always eager to make fun of philosophy. ;-)
    1
  3.  @GregValentine  And the kid takes his ball and runs home to Mommy. ;-)
    1
  4. It's not the simulation that suffers from that, it's the system. As soon as we have an almost periodic system that has bifurcation points this is bound to happen. It's not that much different from a Galton board (or a pinball machine): the trajectory has to pass the equivalent of a nail or obstacle, which creates a hard decision point. The effective "gain" in such a system is infinite and the system can not be stable.
    1
  5. For all the wrong reasons. ;-)
    1
  6.  @yeslinsequeira4612  Because it's entertainment that people are searching for, not science. ;-)
    1
  7.  @yeslinsequeira4612  Because it's entertainment that people are searching for, not science. ;-)
    1
  8. Basically all simulations so far indicate far less stability than has actually been observed. If we take numerical problems out, this indicates that we aren't simulation reality but only an oversimplified version of it. We either have to include additional bodies like asteroids and Oort cloud objects or we have to conclude that there is a dissipative influence, e.g. from dark matter streams within the Milky Way.
    1
  9. Not much. The problem will probably be just as hard on a quantum computer as on a classical one, if not harder. The great myth about quantum computing is that it can speed up all calculations. That's principally false. QC can perform better on some problems, but it can not perform better on all problems and the reverse is also true: there are some problems for which classical computation is actually more powerful. The universe uses both methods.
    1
  10. They tried that. Neural networks seem to make actually pretty good educated guesses, at least statistically speaking. It doesn't solve the predictability problem (that's mathematically impossible) but it might lead to numerical techniques that are more powerful than the conventional brute force algorithms.
    1
  11. I wasted eight episodes of my life on the show. It's horrible.
    1