RE: re Re: Headlight Hype? or Facts

["Smith, Kirby A" <kirby.a.smith@lmco.com>]

> ----------
> Robert Houk asked, below, relative to the following:
> 
>    The diff between a 80/100 and 55/60 bulb is only just discernable.  I
> think
>    ...
> --- and ---
> 
>    Huw is right.  The eye is logarithmic in response.  Further, there are
>    ...
> 
> Isnt' the eye progressively more sensitive at low light levels than at
> high light levels? I.e., at low-light-levels ("night"), doubling the
> light energy is substantially more perceivable than at high-light-levels
> ("day")?
	-------------------
Well, with 5XX Phil Payne watts shining forward, one is likely still in the
photopic (daytime) mode.  :-)   More to the question, if one plots the
logarithm of threshold illuminance of the eye versus the logarithm of
background luminance, the curve is not perfectly linear.  However, the
photopic portion has roughly a slope of 1, while the scotopic portion has a
lower slope.  I interpret this to mean that the minimum illuminance I can
detect doesn't increase as quickly as the background luminance at night.
Conversely, as the background luminance decreases (gets darker), the
improvement in sensitivity does not keep up (eye is worse than one-to-one
logarithmic).  

There is another factor when it is extremely dark.  Then the eye is quantum
limited; below about 100 photons (into the eye) the eye will have a low
probability of detecting the source.  (The maximum period for which these
have to be applied is not listed in my reference.)  A small increase in
light level from this threshold will appear to be a significant relative
improvement in seeing.  However, one shouldn't be driving at the quantum
limit; resolution is very poor.  HTH

>                 .... Kirby   (Kirby A. Smith)
>                               2 x 1988 90q
>                           New Hampshire USA