At 8:24 PM +0000 1/2/03, olympus-digest wrote:
>Date: Thu, 2 Jan 2003 13:04:29 -0600
>From: "gries" <gries@xxxxxxxxxxxxxxxxxxxxxxxxxxx>
>Subject: [OM] bokeh
>
>Is there a good test to determine the quality of a lens' bokeh? Gary's
>tests are quite helpful in determining sharp lenses, but what about the
>bokeh?
>
>I have been immensely satisfied with the bokeh from the 85/2, but less so
>with the 28/2. is there any source that describes the different lenses in
>these terms? What are the terms to describe bokeh (Double-line, etc.)?
I think that bokeh probably has a fairly simple technical explanation: that if
the bokeh is good out-of-focus images are still smooth, and we don't get any
odd artifacts like doubling of lines.
If my theory is true, the test is easy: just run a resolution test with the
lens at varying degrees of defocus, and observe that the three-bar test
patterns smoothly verge to gray as the lines per millimeter increases. The
more defocus, the lower the lines per millimeter at which the patterns will
blur to gray. Look especially for patterns where the number of bars in the
pattern seems to double at some resolution. This is an artifact; the lens
resolution didn't suddenly double. Sometimes you see this even with lenses in
perfect focus, but good lenses don't do this one would expect
If one sees doubling of edges in defocused images, one will see prominant
ripples in the corresponding modulation transfer function, rather than the
traditional monotonic decreasing curve.
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 >
|