Polarizer weirdness

OK, here's something I can't find a reference to, although I just
started looking.

I'm using a Canon Digital Rebel XTi while I wait for whatever Canon's
going to come out with to one-up their own 1Ds Mark II, and I recently
picked up a new 10-22 EFS lens to get some wide angle coverage back (my
other lenses are a 24-105mm L f4, a 100-400mm L f4, a 28-300 L
f3.5-5.6, a 70-200 L f2.8, and a 70-300mm DO).

When I use a polarizer with the new 10-22, I get non-complete
polarizing coverage - i.e., the polarization will only be apparent in
one area of the frame - a circular area above vertical center and dead
center horizontally if the sun is actually 90 degrees to my port or
starboard (g), or a elliptical area to left or right of center if the
angle is not so close to 90 degrees. No vignetting occurs. I can
provide examples if necessary.

I'm using a mid range Sunpak 77mm circular polarizer most of the tiime,
but the effect still happens when I use the high end B+W and/or Hoya
Super Pro polarizers that I normally have on the L series lenses (no
sense putting good glass on a camera and then using crap for filters!),
so it doesn't appear to be polarizer quality related.

Has anyone else encountered this? Is this some inherent problem due to
the more pronounced curvature of the front lens in these extreme wide
angle lenses, or it some weirdness that I can correct for? Would it
help using a Cokin Pro ("Pro", NOT "P") series filter polarizer, as it
would be wider than the lens? Would going to a linear polarizer help
(as digitals can use either, even though they are autofocus)?

Ken Lucke writes:
OK, here's something I can't find a reference to, although I just
started looking.

Look here, the part about using polarizers with wide angle lenses:

http://www.luminous-landscape.com/tu...larizers.shtml

In article , Paul Rubin
wrote:

http://www.luminous-landscape.com/tu...larizers.shtml

OK, so I take it that there's no solution? Damn.

Ken Lucke writes:
http://www.luminous-landscape.com/tu...larizers.shtml
OK, so I take it that there's no solution? Damn.

A "solution" is only needed if you feel that there's a problem.

As the article explains, that wideangle effect can still make good
photos.

And there's always postprocessing...

In article , Paul Rubin
wrote:

Ken Lucke writes:
http://www.luminous-landscape.com/tu...larizers.shtml
OK, so I take it that there's no solution? Damn.

A "solution" is only needed if you feel that there's a problem.

As the article explains, that wideangle effect can still make good
photos.


Yes, but I don't always want that effect.

And there's always postprocessing...

I shoot digital with the intent to do as little post- work as possible.
I want to do it right the first time.

While I have absolutely nothing against _necessary_ post-processing
(photographers have ALWAYS manipulated their images anyway, from as
simple as choice of film, using filters, or cropping, dodging, and
burning to sophisticated emulsion adjustments and processing solutions
to get the look they want, so the "digital vs. film" argument falls on
deaf ears here when it comes to whether or not "post processing" is
bad), and nothing against manipulated images per se (when that is the
intent), I choose to look at an image that needs more than slight
correction as one that I would not normally use, and that the extra
post-processing it would need is a crutch for failing to do it right at
the time. That's not to say that I don't use additional processing at
times [some images are just too good to lose through snobbery), just
that my _goal_ is to be "spot on" when I press the shutter.

you may have to find a bracket type mounted polarizer. but im not sure if
that will even help

Online camera club of friendly photographers
http://www.photochimper.com/forum



"Ken Lucke" wrote in message
...
OK, here's something I can't find a reference to, although I just
started looking.

I'm using a Canon Digital Rebel XTi while I wait for whatever Canon's
going to come out with to one-up their own 1Ds Mark II, and I recently
picked up a new 10-22 EFS lens to get some wide angle coverage back (my
other lenses are a 24-105mm L f4, a 100-400mm L f4, a 28-300 L
f3.5-5.6, a 70-200 L f2.8, and a 70-300mm DO).

When I use a with the new 10-22, I get non-complete
polarizing coverage - i.e., the polarization will only be apparent in
one area of the frame - a circular area above vertical center and dead
center horizontally if the sun is actually 90 degrees to my port or
starboard (g), or a elliptical area to left or right of center if the
angle is not so close to 90 degrees. No vignetting occurs. I can
provide examples if necessary.

I'm using a mid range Sunpak 77mm circular polarizer most of the tiime,
but the effect still happens when I use the high end B+W and/or Hoya
Super Pro polarizers that I normally have on the L series lenses (no
sense putting good glass on a camera and then using crap for filters!),
so it doesn't appear to be polarizer quality related.

Has anyone else encountered this? Is this some inherent problem due to
the more pronounced curvature of the front lens in these extreme wide
angle lenses, or it some weirdness that I can correct for? Would it
help using a Cokin Pro ("Pro", NOT "P") series filter polarizer, as it
would be wider than the lens? Would going to a linear polarizer help
(as digitals can use either, even though they are autofocus)?

On Sat, 10 Nov 2007 06:55:07 -0500, "Rich" wrote:

you may have to find a bracket type mounted polarizer. but im not sure if
that will even help

Online camera club of friendly photographers
http://www.photochimper.com/forum



"Ken Lucke" wrote in message
...
OK, here's something I can't find a reference to, although I just
started looking.

I'm using a Canon Digital Rebel XTi while I wait for whatever Canon's
going to come out with to one-up their own 1Ds Mark II, and I recently
picked up a new 10-22 EFS lens to get some wide angle coverage back (my
other lenses are a 24-105mm L f4, a 100-400mm L f4, a 28-300 L
f3.5-5.6, a 70-200 L f2.8, and a 70-300mm DO).

When I use a with the new 10-22, I get non-complete
polarizing coverage - i.e., the polarization will only be apparent in
one area of the frame - a circular area above vertical center and dead
center horizontally if the sun is actually 90 degrees to my port or
starboard (g), or a elliptical area to left or right of center if the
angle is not so close to 90 degrees. No vignetting occurs. I can
provide examples if necessary.

I'm using a mid range Sunpak 77mm circular polarizer most of the tiime,
but the effect still happens when I use the high end B+W and/or Hoya
Super Pro polarizers that I normally have on the L series lenses (no
sense putting good glass on a camera and then using crap for filters!),
so it doesn't appear to be polarizer quality related.

Has anyone else encountered this? Is this some inherent problem due to
the more pronounced curvature of the front lens in these extreme wide
angle lenses, or it some weirdness that I can correct for? Would it
help using a Cokin Pro ("Pro", NOT "P") series filter polarizer, as it
would be wider than the lens? Would going to a linear polarizer help
(as digitals can use either, even though they are autofocus)?


Odd, but it may actually be the composition of the lens components themselves. I
once had a microscope objective for my phase contrast microscope (Leica optics)
that I ordered with the express purpose for thin-sections of minerals under
polarized light. All my images came out with a maltese-cross polarizing pattern
in them. It was definitely in the objective because when that component was
turned in the optics and lighting configuration then the pattern turned. I had
it exchanged, after much arguing with the sales rep on what the problem was, and
how could it possibly be. For all intents and purposes this should not have
happened due to the type of materials used in that objective. We could only
surmise that that particular blank of glass used to create the lens had a
peculiar stress pattern in it. After exchange for the very same objective model
from the same company, just a different manufacturing date, it didn't have this
defect. If this is the case, then .... so much for "L" meaning anything when
buying glass.

On Sat, 10 Nov 2007 12:23:21 GMT, trevor-johnson
wrote:

On Sat, 10 Nov 2007 06:55:07 -0500, "Rich" wrote:

you may have to find a bracket type mounted polarizer. but im not sure if
that will even help

Online camera club of friendly photographers
http://www.photochimper.com/forum



"Ken Lucke" wrote in message
...
OK, here's something I can't find a reference to, although I just
started looking.

I'm using a Canon Digital Rebel XTi while I wait for whatever Canon's
going to come out with to one-up their own 1Ds Mark II, and I recently
picked up a new 10-22 EFS lens to get some wide angle coverage back (my
other lenses are a 24-105mm L f4, a 100-400mm L f4, a 28-300 L
f3.5-5.6, a 70-200 L f2.8, and a 70-300mm DO).

When I use a with the new 10-22, I get non-complete
polarizing coverage - i.e., the polarization will only be apparent in
one area of the frame - a circular area above vertical center and dead
center horizontally if the sun is actually 90 degrees to my port or
starboard (g), or a elliptical area to left or right of center if the
angle is not so close to 90 degrees. No vignetting occurs. I can
provide examples if necessary.

I'm using a mid range Sunpak 77mm circular polarizer most of the tiime,
but the effect still happens when I use the high end B+W and/or Hoya
Super Pro polarizers that I normally have on the L series lenses (no
sense putting good glass on a camera and then using crap for filters!),
so it doesn't appear to be polarizer quality related.

Has anyone else encountered this? Is this some inherent problem due to
the more pronounced curvature of the front lens in these extreme wide
angle lenses, or it some weirdness that I can correct for? Would it
help using a Cokin Pro ("Pro", NOT "P") series filter polarizer, as it
would be wider than the lens? Would going to a linear polarizer help
(as digitals can use either, even though they are autofocus)?


Odd, but it may actually be the composition of the lens components themselves. I
once had a microscope objective for my phase contrast microscope (Leica optics)
that I ordered with the express purpose for thin-sections of minerals under
polarized light. All my images came out with a maltese-cross polarizing pattern
in them. It was definitely in the objective because when that component was
turned in the optics and lighting configuration then the pattern turned. I had
it exchanged, after much arguing with the sales rep on what the problem was, and
how could it possibly be. For all intents and purposes this should not have
happened due to the type of materials used in that objective. We could only
surmise that that particular blank of glass used to create the lens had a
peculiar stress pattern in it. After exchange for the very same objective model
from the same company, just a different manufacturing date, it didn't have this
defect. If this is the case, then .... so much for "L" meaning anything when
buying glass.


It sounds to me like the OP is observing the way the light in the sky
is polarized, nothing at all wrong with the camera or polarizer. It's
a common observance with polarizers and wide angle lenses. Examine
the sky through the polarizer without mounting it on the camera, some
parts of the sky will show more effect than others.

On Sat, 10 Nov 2007 12:31:10 GMT, Charles wrote:

On Sat, 10 Nov 2007 12:23:21 GMT, trevor-johnson
wrote:

On Sat, 10 Nov 2007 06:55:07 -0500, "Rich" wrote:

you may have to find a bracket type mounted polarizer. but im not sure if
that will even help

Online camera club of friendly photographers
http://www.photochimper.com/forum



"Ken Lucke" wrote in message
...
OK, here's something I can't find a reference to, although I just
started looking.

I'm using a Canon Digital Rebel XTi while I wait for whatever Canon's
going to come out with to one-up their own 1Ds Mark II, and I recently
picked up a new 10-22 EFS lens to get some wide angle coverage back (my
other lenses are a 24-105mm L f4, a 100-400mm L f4, a 28-300 L
f3.5-5.6, a 70-200 L f2.8, and a 70-300mm DO).

When I use a with the new 10-22, I get non-complete
polarizing coverage - i.e., the polarization will only be apparent in
one area of the frame - a circular area above vertical center and dead
center horizontally if the sun is actually 90 degrees to my port or
starboard (g), or a elliptical area to left or right of center if the
angle is not so close to 90 degrees. No vignetting occurs. I can
provide examples if necessary.

I'm using a mid range Sunpak 77mm circular polarizer most of the tiime,
but the effect still happens when I use the high end B+W and/or Hoya
Super Pro polarizers that I normally have on the L series lenses (no
sense putting good glass on a camera and then using crap for filters!),
so it doesn't appear to be polarizer quality related.

Has anyone else encountered this? Is this some inherent problem due to
the more pronounced curvature of the front lens in these extreme wide
angle lenses, or it some weirdness that I can correct for? Would it
help using a Cokin Pro ("Pro", NOT "P") series filter polarizer, as it
would be wider than the lens? Would going to a linear polarizer help
(as digitals can use either, even though they are autofocus)?


Odd, but it may actually be the composition of the lens components themselves. I
once had a microscope objective for my phase contrast microscope (Leica optics)
that I ordered with the express purpose for thin-sections of minerals under
polarized light. All my images came out with a maltese-cross polarizing pattern
in them. It was definitely in the objective because when that component was
turned in the optics and lighting configuration then the pattern turned. I had
it exchanged, after much arguing with the sales rep on what the problem was, and
how could it possibly be. For all intents and purposes this should not have
happened due to the type of materials used in that objective. We could only
surmise that that particular blank of glass used to create the lens had a
peculiar stress pattern in it. After exchange for the very same objective model
from the same company, just a different manufacturing date, it didn't have this
defect. If this is the case, then .... so much for "L" meaning anything when
buying glass.


It sounds to me like the OP is observing the way the light in the sky
is polarized, nothing at all wrong with the camera or polarizer. It's
a common observance with polarizers and wide angle lenses. Examine
the sky through the polarizer without mounting it on the camera, some
parts of the sky will show more effect than others.

That's what I thought at first, but I read the description more closely. If it
was due to normal atmospheric polarization of sunlight then it would be always
occurring in a wide band 90 degrees to the sun. Not in offset areas as defined
circles and ovals. I frequently use ultra-wide and fish-eye lens configurations.
Sometimes with a polarizer just for interesting effects. Highlighting a
particular cloud formation in the sky by deepening the sky behind it. Or just to
intensify fall colors--banding in the sky be damned, I'll play with
post-processing later and see if I can fix that. The band in the sky is there as
I described. Sure it will take on a curvature as any straight-line will when
using such wide angles, but it's never as a disparate circle or oval area
standing apart from horizon or FOV edge.

This is what makes me suspect something else is at play. Two different
polarizers causing the same effect leads to another optics component as the
cause. Primary suspect = main lens. (for now)

"Charles" wrote in message
...
On Sat, 10 Nov 2007 12:23:21 GMT, trevor-johnson
wrote:

On Sat, 10 Nov 2007 06:55:07 -0500, "Rich" wrote:

you may have to find a bracket type mounted polarizer. but im not sure
if
that will even help

Online camera club of friendly photographers
http://www.photochimper.com/forum



"Ken Lucke" wrote in message
...
OK, here's something I can't find a reference to, although I just
started looking.

I'm using a Canon Digital Rebel XTi while I wait for whatever Canon's
going to come out with to one-up their own 1Ds Mark II, and I recently
picked up a new 10-22 EFS lens to get some wide angle coverage back (my
other lenses are a 24-105mm L f4, a 100-400mm L f4, a 28-300 L
f3.5-5.6, a 70-200 L f2.8, and a 70-300mm DO).

When I use a with the new 10-22, I get non-complete
polarizing coverage - i.e., the polarization will only be apparent in
one area of the frame - a circular area above vertical center and dead
center horizontally if the sun is actually 90 degrees to my port or
starboard (g), or a elliptical area to left or right of center if the
angle is not so close to 90 degrees. No vignetting occurs. I can
provide examples if necessary.

I'm using a mid range Sunpak 77mm circular polarizer most of the tiime,
but the effect still happens when I use the high end B+W and/or Hoya
Super Pro polarizers that I normally have on the L series lenses (no
sense putting good glass on a camera and then using crap for filters!),
so it doesn't appear to be polarizer quality related.

Has anyone else encountered this? Is this some inherent problem due to
the more pronounced curvature of the front lens in these extreme wide
angle lenses, or it some weirdness that I can correct for? Would it
help using a Cokin Pro ("Pro", NOT "P") series filter polarizer, as it
would be wider than the lens? Would going to a linear polarizer help
(as digitals can use either, even though they are autofocus)?


Odd, but it may actually be the composition of the lens components
themselves. I
once had a microscope objective for my phase contrast microscope (Leica
optics)
that I ordered with the express purpose for thin-sections of minerals
under
polarized light. All my images came out with a maltese-cross polarizing
pattern
in them. It was definitely in the objective because when that component
was
turned in the optics and lighting configuration then the pattern turned. I
had
it exchanged, after much arguing with the sales rep on what the problem
was, and
how could it possibly be. For all intents and purposes this should not
have
happened due to the type of materials used in that objective. We could
only
surmise that that particular blank of glass used to create the lens had a
peculiar stress pattern in it. After exchange for the very same objective
model
from the same company, just a different manufacturing date, it didn't have
this
defect. If this is the case, then .... so much for "L" meaning anything
when
buying glass.


It sounds to me like the OP is observing the way the light in the sky
is polarized, nothing at all wrong with the camera or polarizer. It's
a common observance with polarizers and wide angle lenses. Examine
the sky through the polarizer without mounting it on the camera, some
parts of the sky will show more effect than others.

I noticed the same when using a polarizer on my zoom at 17mm (APS size
sensor camera). Only a small area of sky would be affected, giving odd
looking dark areas in the images. It works much better at longer focal
lengths. I might try graduated neutral density filters next time.
John