«

I'm having some trouble with blue and red shifts from my Nikon D50. I'm
not sure if 'shift' is the right word to describe the problem, but it
appears like the colors dont line up. Like one color is shifted away
from the others. In this example image, a blue shift is apparent in the
lower left of the picture.
http://gbookphotos.com/July%2030,%20...a/target0.html

The color shift is much more apparent in other pictures I've taken. The
images are all taken using the default, and full auto mode of the
camera. It's also happened with more than one lens. I've used the same
lenses on a Nikon N75 with no problems. I'm wondering if its the
camera, the settings, or lens? Has anyone seen this type of color
distortion before?

-Greg

<> schreef in bericht
news: oups.com...
> I'm having some trouble with blue and red shifts from my Nikon D50. I'm
> not sure if 'shift' is the right word to describe the problem, but it
> appears like the colors dont line up. Like one color is shifted away
> from the others. In this example image, a blue shift is apparent in the
> lower left of the picture.
> http://gbookphotos.com/July%2030,%20...a/target0.html
>
> The color shift is much more apparent in other pictures I've taken. The
> images are all taken using the default, and full auto mode of the
> camera. It's also happened with more than one lens. I've used the same
> lenses on a Nikon N75 with no problems. I'm wondering if its the
> camera, the settings, or lens? Has anyone seen this type of color
> distortion before?

To me the shifting of colors is not apparant. Although I do see a bleu
hase around the plastic I can not see this as a shift in position.
The picture is unsharp, is taken at 1/200 of a sec on a 450 mm equivalent
setting. So the apparent hase could be caused by an almost steady hand
but just a small movement so a bit of the 'plastic' gives some light which
appears to be a blue hase.

To get a better judgement, you have to provide a better example.
You could fotograph something dark with some highlights in the corner,
use a high enough shutterspeet or use a tripod. Close the aperature
one or two stops, so the lensabrations are less.
So to detect colorshifting of the D50 try to exclude all other
causes for color effects in the picture.

Zooms, especially at the wide angle end are prone to chromatic
abrations. This is where red en blue do sepparate, but this is a problem
of the lens and not of a camera. And often this is only visible in high
contrast situations. (For example tree branches in the corner of the
picture against the sky. Or highlights in the corner of a picture).


ben brugman

>
> -Greg
>

wrote:

>I'm having some trouble with blue and red shifts from my Nikon D50. I'm
>not sure if 'shift' is the right word to describe the problem, but it
>appears like the colors dont line up. Like one color is shifted away
>from the others. In this example image, a blue shift is apparent in the
>lower left of the picture.
>http://gbookphotos.com/July%2030,%20...a/target0.html

That image is out of focus or suffers from camera shake, so it's hard to
judge what the problem is and what caused it. The only thing I see is a
bit of halo effect from the blurred plastic bag.

In fact, three of the four photos on your page are poorly focused. The
only one with some decent sharpness is the second one of the cat with
more scenery.

>The color shift is much more apparent in other pictures I've taken.

Can you post one of those images so we can see what you mean?

Is it possible to also post them with a larger image size? The small
images are lacking detail to study.

> The
>images are all taken using the default, and full auto mode of the
>camera. It's also happened with more than one lens. I've used the same
>lenses on a Nikon N75 with no problems. I'm wondering if its the
>camera, the settings, or lens? Has anyone seen this type of color
>distortion before?

I think the problem might be user error.



Try taking a picture in A mode and set the aperture to f/8 to get some
more depth of field in the image. That will help eliminate focus issues
since more of the scene will be in sharp focus.

Make sure the scene is bright enough to allow a fast shutter speed as
well, like 1/320 or better to make sure there is no blurring from camera
shake.

Ben Brugman wrote:
> <> schreef in bericht
> news: oups.com...
> > I'm having some trouble with blue and red shifts from my Nikon D50. I'm
> > not sure if 'shift' is the right word to describe the problem, but it
> > appears like the colors dont line up. Like one color is shifted away
> > from the others. In this example image, a blue shift is apparent in the
> > lower left of the picture.
> > http://gbookphotos.com/July%2030,%20...a/target0.html
> >
> > The color shift is much more apparent in other pictures I've taken. The
> > images are all taken using the default, and full auto mode of the
> > camera. It's also happened with more than one lens. I've used the same
> > lenses on a Nikon N75 with no problems. I'm wondering if its the
> > camera, the settings, or lens? Has anyone seen this type of color
> > distortion before?
>
> To me the shifting of colors is not apparant. Although I do see a bleu
> hase around the plastic I can not see this as a shift in position.
> The picture is unsharp, is taken at 1/200 of a sec on a 450 mm equivalent
> setting. So the apparent hase could be caused by an almost steady hand
> but just a small movement so a bit of the 'plastic' gives some light which
> appears to be a blue hase.
>
> To get a better judgement, you have to provide a better example.
> You could fotograph something dark with some highlights in the corner,
> use a high enough shutterspeet or use a tripod. Close the aperature
> one or two stops, so the lensabrations are less.
> So to detect colorshifting of the D50 try to exclude all other
> causes for color effects in the picture.
>
> Zooms, especially at the wide angle end are prone to chromatic
> abrations. This is where red en blue do sepparate, but this is a problem
> of the lens and not of a camera. And often this is only visible in high
> contrast situations. (For example tree branches in the corner of the
> picture against the sky. Or highlights in the corner of a picture).
>
>
> ben brugman
>
> >
> > -Greg
> >


Here are some more sample pictures:

http://gbookphotos.com/nikon/DSC_0674.JPG - The lower left has some
blue spots, and the grass on the right side has some color distortion
as well.

http://gbookphotos.com/nikon/DSC_0702.JPG - The dark wires and the top
of the flagpole have some distortion.

http://gbookphotos.com/nikon/DSC_0516.JPG - The fencing in the bottom
left appears blue instead of black

http://gbookphotos.com/nikon/DSC_0012.JPG - The purple flower in the
lower left has a blue shadow around it

I hope its not a user issue! But that is something I can fix.

-Greg

chromatic abberation is a problem of lens

http://www.terekhoff.com

wrote:
> Ben Brugman wrote:
>> <> schreef in bericht
>> news: oups.com...
>>> I'm having some trouble with blue and red shifts from my Nikon D50. I'm
>>> not sure if 'shift' is the right word to describe the problem, but it
>>> appears like the colors dont line up. Like one color is shifted away
>>> from the others. In this example image, a blue shift is apparent in the
>>> lower left of the picture.
>>> http://gbookphotos.com/July%2030,%20...a/target0.html
>>>
>>> The color shift is much more apparent in other pictures I've taken. The
>>> images are all taken using the default, and full auto mode of the
>>> camera. It's also happened with more than one lens. I've used the same
>>> lenses on a Nikon N75 with no problems. I'm wondering if its the
>>> camera, the settings, or lens? Has anyone seen this type of color
>>> distortion before?
>> To me the shifting of colors is not apparant. Although I do see a bleu
>> hase around the plastic I can not see this as a shift in position.
>> The picture is unsharp, is taken at 1/200 of a sec on a 450 mm equivalent
>> setting. So the apparent hase could be caused by an almost steady hand
>> but just a small movement so a bit of the 'plastic' gives some light which
>> appears to be a blue hase.
>>
>> To get a better judgement, you have to provide a better example.
>> You could fotograph something dark with some highlights in the corner,
>> use a high enough shutterspeet or use a tripod. Close the aperature
>> one or two stops, so the lensabrations are less.
>> So to detect colorshifting of the D50 try to exclude all other
>> causes for color effects in the picture.
>>
>> Zooms, especially at the wide angle end are prone to chromatic
>> abrations. This is where red en blue do sepparate, but this is a problem
>> of the lens and not of a camera. And often this is only visible in high
>> contrast situations. (For example tree branches in the corner of the
>> picture against the sky. Or highlights in the corner of a picture).
>>
>>
>> ben brugman
>>
>>> -Greg
>>>
>
>
> Here are some more sample pictures:
>
> http://gbookphotos.com/nikon/DSC_0674.JPG - The lower left has some
> blue spots, and the grass on the right side has some color distortion
> as well.
>
> http://gbookphotos.com/nikon/DSC_0702.JPG - The dark wires and the top
> of the flagpole have some distortion.
>
> http://gbookphotos.com/nikon/DSC_0516.JPG - The fencing in the bottom
> left appears blue instead of black
>
> http://gbookphotos.com/nikon/DSC_0012.JPG - The purple flower in the
> lower left has a blue shadow around it
>
> I hope its not a user issue! But that is something I can fix.
>
> -Greg
>
It looks like bad CA to me - so called "purple fringing" as opposed to
lateral CA. AFAIK, unlike lateral CA it is not so easy to fix in post
processing - except by laborious methods.

It is a user issue in that some consumer zooms (70-300 G, ED and Sigma
APO) are not very good at 300mm, especially when used fully wide. EXIF
data indicated that you used f5.7 - so assumed to be fully wide.
If you don't want to see this in your images, then try stopping down the
lens to f8 or smaller, try also to keep used focal length at 200mm or
less. Alternative lenses that perform better at 300mm are expensive.

"Vladislav" <> wrote in message
news: ups.com...
> chromatic abberation is a problem of lens
>
> http://www.terekhoff.com

vladislav got it right - there actually are some pretty effective ways to
counteract and fix that in post processing, if you are familiar with the use
of channels.

sid

frederick wrote:
> wrote:
> > Ben Brugman wrote:
> >> <> schreef in bericht
> >> news: oups.com...
> >>> I'm having some trouble with blue and red shifts from my Nikon D50. I'm
> >>> not sure if 'shift' is the right word to describe the problem, but it
> >>> appears like the colors dont line up. Like one color is shifted away
> >>> from the others. In this example image, a blue shift is apparent in the
> >>> lower left of the picture.
> >>> http://gbookphotos.com/July%2030,%20...a/target0.html
> >>>
> >>> The color shift is much more apparent in other pictures I've taken. The
> >>> images are all taken using the default, and full auto mode of the
> >>> camera. It's also happened with more than one lens. I've used the same
> >>> lenses on a Nikon N75 with no problems. I'm wondering if its the
> >>> camera, the settings, or lens? Has anyone seen this type of color
> >>> distortion before?
> >> To me the shifting of colors is not apparant. Although I do see a bleu
> >> hase around the plastic I can not see this as a shift in position.
> >> The picture is unsharp, is taken at 1/200 of a sec on a 450 mm equivalent
> >> setting. So the apparent hase could be caused by an almost steady hand
> >> but just a small movement so a bit of the 'plastic' gives some light which
> >> appears to be a blue hase.
> >>
> >> To get a better judgement, you have to provide a better example.
> >> You could fotograph something dark with some highlights in the corner,
> >> use a high enough shutterspeet or use a tripod. Close the aperature
> >> one or two stops, so the lensabrations are less.
> >> So to detect colorshifting of the D50 try to exclude all other
> >> causes for color effects in the picture.
> >>
> >> Zooms, especially at the wide angle end are prone to chromatic
> >> abrations. This is where red en blue do sepparate, but this is a problem
> >> of the lens and not of a camera. And often this is only visible in high
> >> contrast situations. (For example tree branches in the corner of the
> >> picture against the sky. Or highlights in the corner of a picture).
> >>
> >>
> >> ben brugman
> >>
> >>> -Greg
> >>>
> >
> >
> > Here are some more sample pictures:
> >
> > http://gbookphotos.com/nikon/DSC_0674.JPG - The lower left has some
> > blue spots, and the grass on the right side has some color distortion
> > as well.
> >
> > http://gbookphotos.com/nikon/DSC_0702.JPG - The dark wires and the top
> > of the flagpole have some distortion.
> >
> > http://gbookphotos.com/nikon/DSC_0516.JPG - The fencing in the bottom
> > left appears blue instead of black
> >
> > http://gbookphotos.com/nikon/DSC_0012.JPG - The purple flower in the
> > lower left has a blue shadow around it
> >
> > I hope its not a user issue! But that is something I can fix.
> >
> > -Greg
> >
> It looks like bad CA to me - so called "purple fringing" as opposed to
> lateral CA. AFAIK, unlike lateral CA it is not so easy to fix in post
> processing - except by laborious methods.
>
> It is a user issue in that some consumer zooms (70-300 G, ED and Sigma
> APO) are not very good at 300mm, especially when used fully wide. EXIF
> data indicated that you used f5.7 - so assumed to be fully wide.
> If you don't want to see this in your images, then try stopping down the
> lens to f8 or smaller, try also to keep used focal length at 200mm or
> less. Alternative lenses that perform better at 300mm are expensive.


Thanks for the tips. Upon searching for purple fringing, I came across
this article:
http://www.pcworld.idg.com.au/index....pid;1585691688
, which was extremely helpful in fixing my pictures. I also found a
Photoshop plugin called PTLens which is good for fixing some chromatic
abberation.

Searching has also shown that CA is pretty common in digital cameras.
Why is the case? I dont ever remember seeing it in the pictures taken
with my Nikon N75, even with the same lenses. I was thinking maybe it
was there in the film pictures, but I didn't have the resolution to see
it clearly...

thanks for the help!

-Greg

"frederick" <> schreef in bericht
news:1156815064.876900@ftpsrv1...
> wrote:
>> Ben Brugman wrote:
snip
>>>
>>> To get a better judgement, you have to provide a better example.
>>> You could fotograph something dark with some highlights in the corner,
>>> use a high enough shutterspeet or use a tripod. Close the aperature
>>> one or two stops, so the lensabrations are less.
>>> So to detect colorshifting of the D50 try to exclude all other
>>> causes for color effects in the picture.
>>>
I did ask for some more information. But Frederick has allready responded
to the 'new' information. And I do agree with the given answer.

Ben

Unlike film in which the image circle is projected onto the photo
sensitive media, there additional filters and sensors above the photo
sensor upon which the image circle is projected and then needs to be
refocused through the IR, low pass filter, Bayer filter and boundaries
of the photo sites to get to the photo sensitive sites. These
additional filters add additional thickness which can add the
opportunity for additional CA. For film the light just has to pass
through the layers of the photo sensitive emulsion..

http://www.shortcourses.com/choosing...e%20to%20color

http://groups.google.com/group/rec.p...al+lens+design

Ben Brugman wrote:
> "frederick" <> schreef in bericht
> news:1156815064.876900@ftpsrv1...
> > wrote:
> >> Ben Brugman wrote:
> snip
> >>>
> >>> To get a better judgement, you have to provide a better example.
> >>> You could fotograph something dark with some highlights in the corner,
> >>> use a high enough shutterspeet or use a tripod. Close the aperature
> >>> one or two stops, so the lensabrations are less.
> >>> So to detect colorshifting of the D50 try to exclude all other
> >>> causes for color effects in the picture.
> >>>
> I did ask for some more information. But Frederick has allready responded
> to the 'new' information. And I do agree with the given answer.
>
> Ben