At 3:02 AM +0000 12/24/02, olympus-digest wrote:
>Date: Tue, 24 Dec 2002 11:22:45 +1300
>From: "Brian Swale" <bj@xxxxxxxxxxxxxxx>
>Subject: [OM] Film and lens resolution (was Re: Dipping our Toe Into Digital
>(LF lenses))
>
>Hi folks,
>
>Joe Gwinn wrote
>
> > We are somewhat going in circles here. The reason that all photgraphic
> > lenses
> > seem to have resolution within a factor of two of 50 line pairs per
> > millimeter
> > (100 pixels per millimeter) is that this is more or less the resolution of
> > film
> > of reasonable sensitivity (ASA 100 or greater), so the lenses are designed
> > to
> > that resolution.
> >
> > Whatever the size of the film, 35mm, 6x6, 4x5, 8x10, the film resolution is
> > the
> > same, so all lenses will have more or less the same on-film resolution,
> > with the
> > more expensive lenses being better, and the less expensive lenses being
> > worse.
> > The limits, from the websites mentioned below, seem to range from 100 line
> > pairs
> > to 25 line pairs per millimeter.
>
>In my opinion, it's not reasonable to generalise so much.
Why not? I'm talking about the typical camera user, the one who still consumes
a few billion frames a year taking lousy snapshots of friends and family. They
use ASA 200 and 400, mostly.
>With lenses there are very measurable and significant differences.
Sure. Nobody said that there wasn't. The point is that the ratio of best to
worst (excluding the hopeless dogs) is about four to one, and that this is no
accident.
>For film, there are also significant differences between emulsions. Go to
>http://www.google.com and do a search for " film resolution limits " without
>the quotes.
Also true, and there are spectrographic films that can do thousands of lines
per millimeter. But the typical ASA 200 to 400 color negative film is nowhere
that good.
>Near the top of the search you will find a couple of Robert Monaghan's pages.
>
>You will need to download the cached versions as the original sites have
>exceeded their paid-up download limits and have all access denied except
>for the author.
>
>The file names are "lenslpm.html" and "limits.html"
>
>They are very instructive as to what film, lenses and systems can deliver.
>
>If anyone can't get these files give me a call and I can e-mail copies.
I guess I can wait until the website becomes available again. It would be
interesting, but I assume it's large.
The bottom line is that lenses and film have coevolved. Most people take their
shots handheld, at something like 1/60-second, so motion blur will limit the
achievable resolution. Given this resolution, the film needs to be only a bit
better, and the rest of the film's tradeoff space can be spent on making the
sensitivity to light good. Given the resolution of typical film and the fact
the camera in mostly handheld, the necessary lens resolution is determined, and
the rest of the lens trade space can be spent on color correction, big
apertures, etc. Each camers manufacturer sells into a market that determines
how much the camera system can cost. The higher price points allow a better
camera to be produced within those constraints.
I recall a story from a photo mag about the genesis of the original XA camera,
designed by Maitani (sp?) as a young man. The crux of the story was that
Maitani dropped some typical feature and was able to buy a very good
multielement glass lens for the savings, thus preserving the intended price
point. It turned out to be a good tradeoff, as the XA line sold like hotcakes.
Joe Gwinn
< This message was delivered via the Olympus Mailing List >
< For questions, mailto:owner-olympus@xxxxxxxxxxxxxxx >
< Web Page: http://Zuiko.sls.bc.ca/swright/olympuslist.html >
|