What I mean is, when they say one of the Luxor lights produces a billion candlepower, how did they determine that?
Here’s the puzzle. Build a theoretical searchlight with a perfect parabolic reflector and a point light source at its focus. Assume the reflector’s depth equals its diameter, so it captures 16/17 of the light’s total output and sends it out in a cylinder of parallel light beams (plus the light that didn’t hit the reflector). How do you calculate candlepower when you can’t assume inverse-square falloff in illumination at the target? Or to put it another way, since candlepower is lumens per steradian and the lumens in the cylinder is near constant while steradians are decreasing …
(Preliminary answer; please correct me if this is wrong).
At an approximation, you can work backwards from a lux measurement from a regular photometer if you stand far enough away from the light source such that the beam is more or less “stable” and uniform, and from there assume the inverse-square law.
To be more precise, you use a goniophotometer (a photometer mounted on a rotating stage that can capture all angles) and sample it over multiple points to get a plot like this:
From there you can either choose a defined area and average it out, follow the regulations for a given industry (e.g. regulations for auto headlights), or just fudge it and cite the biggest number you can get.
I suppose the answer will be that the candlepower of the theoretical searchlight is undefinable. With a real searchlight, nothing to do but measure, measure, measure, farther and farther away, until it starts acting fairly inverse-square. Which might mean quite a few miles.
For starters, you don’t have a point light source. You might have something really close to a point, but it won’t be a perfect point. Which means that it won’t all be right on the focal point, which means that it’ll still spread.
