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: [Leica] Re: Greg Bicket's focus thread
From: Larry Kopitnik <>
Date: Sat, 4 Nov 2000 20:03:17 -0600

>Now as long as we're considering the remote possibility of photographer
>error, let me jump to confess that after lots of trying with the 80mm R, I
>could never consistently put the focus plane where I wanted it.  Some rolls
>would come back with hilariously misplaced planes of focus.  One by one
>revealing that while there was indeed a razor sharp focus plane, and
>beaucoup bokeh, often nowhere near I intended it.  I got it right between a
>third and half the time!  There were even shots that turned out interesting,
>but fully unintentional in terms of what I had attempted to do.
>Compliments about the unintended focus plane in a particular photograph made
>it worse!

Greg's post is a great post. Everyone should read it in it's entirety. Not
just this snip.

What Greg has outlined is exactly why AUTOFOCUS camera/lenses are useful
for only a very limited and select set of conditions. Like bright sun, ISO
400, f/8-f/22. Weddings, flat walls with graffiti, etc. Autofocus
algorithms look for vertical or horizontal (whichever your camera maker has
provided) contrast lines. 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? They cannot, the camera can't, and even you
the human, will sometimes have trouble.

Hmmmm. I agree that Greg's post is an excellent post. And I read it 
thinking, "This is where autofocus excels."

My camera knows what I want in focus by my putting an autofocus 
sensor over it. Just like putting a rangefinder patch over it. As a 
cross-shaped sensor, doesn't matter if what I've put it over is 
vertical or horizontal. It focuses regardless. And most often -- not 
every time, I'll grant you, but more often than not -- it focuses 
faster and more accurately than I would by turning the focus ring and 
guaging focus on the ground glass.

My 85 mm autofocus Nikon lens is a prime example. I'll shoot it 
reguarly with 100 speed film at f/1.4 and 1/60 second (the slowest 
speed I can reliably handhold it; I'll sometimes try 1/30 and 
autofocus works equally well in that light, but my holding the lens 
steady is less reliable). Put one of the autofocus sensors over the 
subject's eye. Press a button. Recompose (and with multiple autofocus 
spots in modern autofocus cameras, recomposing usually means moving 
the frame just a smidge) and shoot the picture. And if the subject 
didn't move, by golly, that eye is in focus. Just like I, not a 
programmer, decided. And in focus more quickly and more often than 
when trying to focus the same lens myself on the screen of a 
non-autofocus body (and therefore a focus screen optimized for manual 

Autofocus is no panacea, but in the situation Greg described, it 
works. It really does. Which may be one reason why Greg gave up on 
his 80 mm Summilux R while my 85 mm f/1.4 Nikkor is my most-used SLR