Re: L* versus rods
Re: L* versus rods
- Subject: Re: L* versus rods
- From: Nathan Duran <email@hidden>
- Date: Mon, 05 Dec 2005 16:45:17 -0800
- Thread-topic: L* versus rods
> Then what exactly does the CIE lab model represent?
Mathematical models are not the things they are modeling. While the number
of points on a line is infinite, time and storage space are not. Any time
you're going to break an analog phenomenon down into something quantifiable
you can record and/or reproduce, you have to pick a number of points and
stick with it, otherwise you'll quickly run out of room to writewhether on
paper, magnetic storage media or crystal atomsand you'll run out of
lifespan before you've even made a dent.
The 65,536 figure is an essentially arbitrary artifact of present day
computer architecture, representing the largest value one can store in two
bytes worth of space. This number has nothing whatsoever to do with human
perception or any particular color space, and as drive storage capacities
and processor speeds increase, you can doubtlessly expect to see image file
formats that allow for 32 or even 64 bit images. This does not mean that
your eyes will magically develop the ability to discern 4,294,967,295
discrete lightness values, it just means you'll have more data in your file
to play with, affording you greater adjustment latitude.
So to your original question "Are there a finite number of points along the
L* axis that correspond to the eye¹s ability to distinguish variations in
luminance?": Probably. Will those points be identical for every eye on the
planet in perpetuity? Probably not. Should the color model impose any
artificial limitations on acceptable values? Absolutely not.
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Colorsync-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden