In article <4s0ejj$6dr@newsstand.cit.cornell.edu>, d_ruether@hotmail.com says...
>In article <4rufvv$a80@walters.East.Sun.COM>, dkc@troi.East.Sun.COM says...

>>When a lens test indicates some type of distortion (pincushion/barrel)
>>it will give the amount in a percentage. I was wondering at what
>>percent will the naked eye be able to detect this distortion.
>>For example: A 50mm/1.7 lens tests indicates Distortion: 1.5% barrel
>>Can I see a barrel distortion of 1.5% on a print or slide
>>or need it be say 5% before the naked eye can detect a barrel effect?
>>The reason I ask, I see most lens vary between 1% and 5%. When should
>>I worry about the number, 1,2,3,4,5%?

>Much of the time you will not be aware of lens distortions, no matter
>how bad they are. Exceptions: horizons or distant large water-body
>edges near the frame edge; buildings shot with straight edges running
>near a large proportion of the frame edge; copy work. The distortion
>is usually (but not always!) greatest near the frame edge, and it is
>always zero with straight lines running through the center of the image.
>There are three kinds of distortion: Barrel, in which straight lines
>not running through the image center bow away from the image center;
>Pincushion, in which straight lines not running through the image center
>bow toward the image center; and "wavy-line", which is a combination of
>barrel over most of the image, switching to pincushion near the edges
>or corners. Wide-angles usually show the first or third types; normals
>often show the first type; telephotos sometimes show the second type;
>macros usually show no distortion; zooms usually show the first and
>second type, and can also show the third type of distortion. (Keep in
>mind that most SLR camera viewfinders have pincushion distortion, so
>it is hard to evaluate lens distortion [or accomodate it, either,
>unfortunately] using the camera viewfinders.)

I thought I should add that some lenses that are commonly thought to
make "distorted" images, don't: fisheye lenses render spherical-
perspective images correctly (without "distortion"), according to the
rules of spherical perspective; super-wide rectangular-perspective
lenses (assuming no barrel, pincushion, or "wavy-line" distortion is
present) render rectangular-perspective images correctly (without "distortion"), according to the rules of rectangular perspective.
A third type of perspective (cylindrical) is rendered (correctly)
by rotating-slit panorama cameras. The appearance of distortion comes
from unfamiliarity with the characteristics of those perspective types,
or when a very wide angle-of-view and/or an unusually close placement
of the subject near the camera are involved. Rounded objects near the
corners of rectangular-perspective super-wides, or straight lines in
fisheyes or rotating-lens panorama cameras may appear to show lens distortions, but, technically, they do not.

(BTW, we see in spherical [fisheye] perspective, contrary to popular
belief, which accounts for why people find some photographic
"distortions", like the corner effects and converging parallel lines
of rectangular-perspective super-wides, disturbing [these are less
prominent in spherical perspective]. Since our vertical angle-of-view
is restricted, the visibility of the curvature of long straight lines
far from the center of our view is reduced [and most people only
attend to a very small part of their angle-of-view, which further
reduces their awareness of the perspective type with which we see].
Also, since most people attend to only a narrow angle-of-view
[a "telephoto"-like image - in which all perspective types have
nearly the same characteristics, making it easy to assume that we
see in rectangular perspective {which has been used in most
image-making for several centuries, and therefore it appears the
most familiar}], as soon as they try a super-wide lens, they tend
to move close to a subject to enlarge it within the confines of the photographic frame, introducing unfamiliar viewing-distance effects.)
Hope This Helps