Comments by "Dr Gamma D" (@DrDeuteron) on "3Blue1Brown" channel.

  1. FYI: this effect is used in GPS. The ionosphere rotates the polarization of the L-band signal (Faraday rotation). By transmitting circular polarized radio at 2 frequencies, L1 and L2, the difference in the delay can be used to find the total delay, and since GPS is based on timing, you need to know the travel time btw satellite and receiver.
    556
  2.  @InXLsisDeo  His visualizations are top notch. Back in the day, you had to come up with them in your head to be a successful student.
    12
  3.  @joshyoung1440 I did meet him, I was in his last lecture, and I can read, and I have no idea what your point is.
    12
  4.  @TheToric  idk, they keep changing it. The GPS signal is sophisticated.
    10
  5. 0:18 Your linear polarizing filter has a little arrow (vector: direction, Y_(1, m )) showing the filter's alignment. It should be a double headed arrow b/c linear polarization is a tensor alignment (no direction, Y_(2, +/-2))
    7
  6.  @Megaemce  now that is the mind of an experimentalist....or maybe a theorist. idk.
    7
  7. A water/lucite/air boundary is a partial polarizing filter with a dependency on the tangent plane normal (n) and incoming/outgoing wave vectors (k, k'). Should be easy to find a scalar from the 3 vectors, idk, (n x k) dot (n x k') maybe, that gives the degree of polarization. I liked n dot (k x k'), but that's a pseudo-scalar, and Maxwell respects parity symmetry. btw: this is why the light is white at then end of the tube: you can't define a plane of polarization if k' is parallel to k, tldr; k x k' = 0.
    7
  8. Laser ppl have frequency doublers and triplers.. But they’re nonlinear optics things. Over in microwave stuff, you do get harmonics when you mix frequencies and do basic linear filtering.
    7
  9. Simple, damped, driven, coupled …and then quantized Ftw. Pendulum with large amplitude: Hold my 🍺
    4
  10. I’m going to strongly disagree with that.
    4
  11.  @SodiumNitrateBot  right, because clock drift is one of the variables
    3
  12. There are four kinds: Easy Magic Computer Quantum chromodynamics.
    3
  13.  @SnakeSalmon8izback  that’s is true, bit not enough. watch this. I am thinking about Venus. No Jupiter. No andromeda. The spot I was thinking about moved much faster than light.
    3
  14. This where using wave number (inverse reduced wavelength) is more fruitful.
    2
  15.  @JamesTMonaco  I didn't want to get into different polarization basis states...so if the |+>, |-> propagate at different speeds, then linear polarization, which is a superposition, will rotate. Seemed like a bit much. But, yeah it is misleading.
    2
  16. |H> = (|+> + i|->) / sqrt(2) |V> = (|+> - i|->) / sqrt(2) are just different basis for 2 degrees of freedom. Linear (H/V) or circular (+/-) polarization depends on the problem you're solving. Ofc there are an infinite number of real combinations of H and V, which just rotate the definitions of "horizontal" and "vertical". For circular light, that degree-of-freedom is just a global phase factor...we chose exp(i 0) = 1 for simplicity. The reason we use handedness vs clockwise is because there is a direction of propagation, which tells you which way to point your thumb. That is ambiguous with (count)clockwise.
    2
  17. I got nervous af.
    2
  18. true. It's definitely on the B team.
    2
  19. it breaks down when you're looking at photons, which is not the case here. See Glauber State. It really breaks with multi-photon states. See Hong–Ou–Mandel effect
    2
  20. Omg, grant should get paid to do a complete animation of the Feynman lectures.
    2
  21. nothing. The timescale for moving water is infinity compared with the light/molecule interaction time.
    2
  22. You need change "only colors perpendicular are [refracted/transmitted]", to "[refraction/transmission] is weighted by cos(angle)"....or maybe sin(angle) for perpendicular....but cosine is the traditional function since z = r x cos(theta) in spherical coordinates.
    2
  23. I was hoping for the (Romeo +/- i*Juliet)/1.414.. basis, since they're eigenvectors.
    1
  24. Got Lagrangian?
    1
  25. Thinc outside the sinc I always say. (because it's the fourier transform of a box). edit: sorry, I just realized this was a spoiler alert since convolution is fourier multiplication....and rect? no. Box.
    1
  26. because pythagorean's theorem.
    1
  27. His example was a complicated analog to a Moire pattern. You can image those things moving faster than light (in theory)….because the 0 crossing is what is moving, and that’s not a physical thing. It’s a perception. Or just the intersection of the edges of two piece of paper as you slide them over each other. The point where they meet can move at infinity…because it’s not a thing. Phase is very similar. E.g., where/when the field changes sign…which is not information
    1
  28. Hyperfine tat at 25:05
    1
  29. Biggest surprise math application? Lie groups.
    1
  30. Diffraction grating is phased array, Hologram is SAR.
    1
  31.  @joaquins90  well I didn't think to deeply, but if I make my left hand tangent to the tube, and rotate my right hand as I move down the tube (at different radial offsets), the dot product of the normals makes a screw pattern. If I had Grant's skills, I'd just write some python to visualize for me. The point though, is that if your take all the available vectors in the problem, you should be able to construct something simple that describes your observation, and is + under parity and time reversal. It works for a lot things with E (vector) and B (axial vector).
    1
  32. This video shows why eigenstates are used. The color indicates frequency/wavenumber/momentum states, while polarization adds a binary state. Linear vs. circular are just two different bases: H, V vs +/-, with the latter being angular momentum eigenstates.
    1
  33. then real question, then, is not the HUP, but de Broglie....why is p ~ k? (and it's k that's exact, p is scaled by hbar).
    1
  34. 17:40 the charges are bound to a spring....it's called an atom. The charges are not free, but the orbitals develop a dipole moment in resonance to a linear electric field. When the electrons are free to move in a sheet, you get something like silver. Polished silver. Aka: a mirror. Anyway, fantastic video, but I will still say light slows down in a dielectric.
    1
  35. Good question. Is it linear, is it circular? What it truly is is the manifestation of the photon spin and it’s 2 states. And why a vector boson has 2 states and not 3 is because gauge invariance…and all that occurs is because the photon field is a 4 vector affine connection of the manifold…but that’s too much. It’s that all particles transform under some representation of the Lorentz group, (see wiki for that), and the photon has the one we call vector. (Electrons get spinor, gravitons get tensor).
    1
  36. this episode is sooooo... convoluted!
    1
  37. Grant addresses that concern in the pinned comment.
    1
  38. The neutrino needs to be converted into an electron first…..and idk, a math guy switching to physics….usually it’s the other way around (and it’s a cry for help)
    1
  39. Grant is such a great animator and teacher, but if you want to be a pro, you have to have all these animations in your head already.
    1
  40. Math is blobs all the way down.
    1
  41. Coherence means phase stability, which means phase difference stability.
    1
  42. I’m still in the light slows down camp. And I’m not accepting arguments from noobs.
    1
  43. exp(i(wt - kxc/n_left)) vs exp(I(wt - kxc/n_right))
    1
  44.  @thatspsychotic  the reminds me, the eq. should explicitly state the indices of refraction are frequency dependent: n_left(w) and n_right(w), and if you make a linear approximation: n_l,r = n +/- eta(w) you get an exp( -/+ i c eta(w)/n) term for propagation..which is screw shaped.
    1
  45. But the sign of the charge doesn’t matter, since the F -> -F X = F / f^2 -> -x And dipole moment: P = qx -> (-q)(-x) = P Light doesn’t notice the sign change.
    1
  46.  @mp3lwgm  did he say if the prism was made from matter or antimatter, cause it doesn’t matter.
    1
  47. That’s a question for joe Rogan
    1
  48. ideal polarizing filters are idempotent: P^2 = P. So it does nothing.
    1
  49. Perfectly elastic collisions violate special relativity.
    1
  50. I used to work in a digital form of holography called interferometric synthetic aperture radar. There reference beam is just a really stable oscillator on board.
    1