Archived posting to the Leica Users Group, 2001/07/25
[Author Prev] [Author Next] [Thread Prev] [Thread Next] [Author Index] [Topic Index] [Home] [Search]Austin writes: > Digital image capture is NOT based on "information > theory" ... Information theory is the quintessence of digital image capture, and indeed of all digital processing of all kinds. > I have shown, why that IS a rule. No, it is an empirical guideline, imposed by engineering limitations up to now. It's rather like saying that no lens can be faster than f/0.7. Actually, a lens _can_ be faster than f/0.7, but from an engineering standpoint, it is extraordinarily difficult to build lenses that are faster than that (indeed, even f/1 is a major feat). But it is important to distinguish between what is practical and what is actually possible, otherwise you never actually achieve all that is truly possible. > It's such a basic rule of signal processing, it is > obviously you don't have any real background in this > field. The most basic rules of signal processing are embodied in information theory, and you've just told me that digital image capture is not based on information theory. > If you really believe yourself, then show me > that you can design a scanner that can reliably > scan a line of .009mm with a sensor that is > .009mm without doing some technique to (effectively) > increase the sample rate. I've just explained what is possible, not what is actually done. I suppose that if some engineers believe that certain things are not possible, out of a lack of familiarity with theory, they'll probably never design scanners that achieve those things. It wasn't so long ago that people believed that aspherics and diffractive optics were blue-sky or "impossible," too.