Luminance Benchmarks for Human Vision

Source handprint : adaptation, anchoring & contrast

So if 4*10^6cd/m^2 causes "eye damage after brief exposure," and the sun is a thousand times more intense, that means the sun causes instant-blindness.

:thinking:

Bwahaha. Well, I must be blind then, eh?

^ it means "direct exposure".

If you look directly the sun at zenith in a summer day without clouds and without blink, you can bet your vision will develope a "permanent black spot" in the middle in few seconds (where "brief exposure" is intended probably as some seconds, not milliseconds )

This table is confusing to me: it seems to be a mix of the brighness of sources (like the computer screen) and the brightness of things illumninated by a source (like the paper in daylight).

Some very confusing numbers, for example:

7.000 - low beam car headlights
120.000 - 60 watt incandescent light

This can obviously not be comparable, since the optical output of both is almost the same (car headlights are 55 watt halogen usually). Perhaps the 7000 figure is for the beam illuminating something, while the 120.000 figure is for the filament of the lightbulb.

If the efficacy of the lightbulb would be 15 lm/watt (fair value) its total flux would be around 1000 lumens (reasonable). Considering its an omnidirectional source, i'd get something in the order of 80 candela. Perhaps the surface area of the filament is 1/1500th of a square meter? Or is it a frosted bulb? The latter seems somewhat reasonable given the size of lightbulbs, but its just a convenient number, not something that could be derrived from that table.

low beams, perhaps, are assumed to not be focused towards the measuring instrument. i.e. if you were above the cutoff: http://www.fiatforum.com/gallery/data/500/medium/halogen_cutoff.jpg

Its possible that they took that reading from an arbitrary distance and height from an actual car - but without specifying both, it makes little sense to me.

Yeah, you can't be scientific if you're going to report your data that vaguely.