Archived posting to the Leica Users Group, 2000/11/04

[Author Prev] [Author Next] [Thread Prev] [Thread Next] [Author Index] [Topic Index] [Home] [Search]

Subject: RE: [Leica] Re: Greg Bicket's focus thread
From: Austin Franklin <austin@darkroom.com>
Date: Sat, 4 Nov 2000 22:45:33 -0500

> > Since you, the human, who are looking at the ground glass screen, know 
what it
> > is that you want in focus, how can a computer program written by a 
programmer,
> > sitting in a cubicle somewhere, know what you want to focus on

> Autofocus is a tool and, like any
> other tool, it can be used effectively or not effectively.  I'm surprised
> that you, Jim, with your experience of digital photography would voice 
such
> a nonsense argument.

> Autofocus knows when *something* is in focus: it is up to the 
photographer
> to choose *what* to focus upon.  The autofocus doesn't try to guess 
*what*
> you want to focus on any more than it tries to guess what you had for 
lunch
> last Thursday.

Jim is absolutely right, and it is not nonsense.  The autofocus systems are 
one, not guideable, two, have an angle of view, and three typically require 
some contrast.  If one could guide the autofocus to a specific point in 
space, that would be a different story...but the current systems don't work 
that way.  There is no way you can tell it to focus halfway down the face 
so only the left eye of your subject is in focus.  If the subject's face is 
in the autofocus domain, it will focus 'somewhere' on the face.

The technology that would work is to have retina tracking that knows what 
spot you are looking at, with a very small angle of view, and focus there. 
 No one has that availability yet, and the sensors, and computing power 
necessary for that, in such a small package, are possibly tens of years 
away.

Replies: Reply from Greg Locke <locke@straylight.ca> ([Leica] OTTAWA?)
Reply from Jim Brick <jim@brick.org> ([Leica] Re: Focus thread)