Light Bending · 1919 test 4GM / c²b · falls off as 1/b1.75″ GR · 0.87″ Newton1.98″ ±0.12 · Príncipe 1.61″ ±0.300.02% of GR29 May 1919 · Príncipe & Sobral
Deflection, live
α = 4GM / c²b = 1.75″ / b
…
Exaggeration
shown at ×350
View
Eclipsethe 1919 observation
Raysthe mechanism, side on
Which theory bends the light
Einstein
Newton
Compare
Display
Remove Moon (daytime glare)
Readout · live
Limb deflection
1.75″
Newton's half
0.87″
Nearest ray b
…
Bend at that b
…
Sobral 1919
1.98″ ± 0.12
Príncipe 1919
1.61″ ± 0.30
VLBI today
0.02%
Why exactly twice
falling · time curvature 0.87″ · Newton has this
curved space itself 0.87″ · GR only
1.75″ at the limb
planets barely feel the right half; light at c collects both
Light falls toward the Sun like anything else: that alone gives 0.87″, and Newton's corpuscle theory (Soldner 1801) predicts exactly it. Completed general relativity adds a second, equal contribution from the curvature of space itself. The sum, 1.75″, is exactly double, and that clean factor of two is what the 1919 eclipse was designed to decide.
What is exact vs stylised
Exact: every star's outward shift is computed from the weak-field law α = 4GM/c²b at its true distance from the Sun; the limb value 1.75″, Newton's 0.87″, the factor of two, and the 1919/VLBI numbers are established physics and history. Stylised: the one big liberty is the exaggeration factor, displayed live at all times (the real shift is invisible at screen scale). The corona, chromosphere, glare and the star field are illustration; the ray-view bend angle is exaggerated like everything else, and the b and α it reports are true values.
▶ Story · 8 chapters
⟳ Reset
▶ The story: the eclipse that made Einstein famous
drag the Sun · scroll = exaggeration · click a part drag the Sun · pinch = exaggeration · tap a part
Tip: this instrument is built for desktop browsers; phones may crop some panels Got it
×
The station is free. No ads, no tracking cookies. If the files are worth your time, following is the support.