contrast and resolution are intertwined, you cant specify
a resolution without also specifying the loss of contrast
at the resolving limit.
JCO
----------------------------------------------------------------------------
----
J.C. O'Connell mailto:[EMAIL PROTECTED] http://jcoconnell.com
----------------------------------------------------------------------------
----
-----Original Message-----
From: Dr E D F Williams [mailto:[EMAIL PROTECTED]
Sent: Monday, September 15, 2003 2:20 PM
To: [EMAIL PROTECTED]
Subject: Re: Lens resolution
Correct. 'Lines per Millimetre' is the way Kodak described the resolution of
their plates. Those we used for high resolution work resolved about 1200
lines per millimetre. The Holographic plates were much better and more
expensive. Now that I've thought about this a little our Kern Switars may
have done about 450 lpm.
But in ordinary 35mm photographic applications 50-60 lines per millimetre on
the film is great ... and good enough for nice big prints; which will not
show more than about a tenth of this anyway. But contrast is more important
than resolution and without wanting to start a whole new ball rolling, very
high resolution pictures with low contrast are terrible. In transmission
light microscopy one needs some of each and for that a very carefully
adjusted light source, according to 'K�hler's' principle, is required. But
for high resolution one needs high resolution film. I think T-Max 100 should
resolve about a hundred lpm with care and proper processing. And it has a
range of about 8 stops which makes it, and Technical Pan, Kodak's best
offerings for B&W photomicrography on 35mm.
Don
_______________
Dr E D F Williams
http://personal.inet.fi/cool/don.williams
Author's Web Site and Photo Gallery
Updated: July 31, 2003
----- Original Message -----
From: "John Francis" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>
Sent: Monday, September 15, 2003 8:31 PM
Subject: Re: Lens resolution
> >
> > One black line and one white line is a "PAIR"
> > of lines to me. Why would you need 4 lines
> > to make up a pair?
>
> You don't. It's line *pairs* per mm if you count both the black
> and the white line. If you only count the black lines (treating
> the white as just the background) you get lines per mm.
>
> In other words:
>
> 100 line *pairs* per mmm is 100 black lines and 100 white lines.
>
> 100 lines per mm is 100 black lines (on a white background), or
> alternatively 100 white lines (on a black background).
>
> They all describe the same resolution pattern. The terminology
> using line pairs is technically more correct, but the other usage
> is widespread and is a closer analogue to the measurements used in
> diffraction gratings, which generally talk about rulings per mm.
>
