Dark corner using a cpl

bychan
Posted 09/03/2010 - 00:41 Link
I've noticed whilst using a cpl on the Sigma 18-50 (72mm), that on certain shots the top left-hand corner is slightly darker.

Whilst not very extreme, I'm guessing that this is caused by the cpl causing vignetting? I'm using a standard Kenko digital cpl.

If this is the case, would a slimmer filter eliminate the problem?
Or get the cpl for use with my Cokin P holders?

Looking back on older photos, the problem seems worse for those taken with the Tamron 18-250 + cpl.

A few examples using the Sigma can be found here

link

Regards
Adrian
K5IIs, Sigma 10-20, Pentax DA 16-85, Pentax DA 55-300, Pentax 70 Ltd, Metz 44 AF-2.
Flickr: https://www.flickr.com/photos/ambott/
Anvh
Posted 09/03/2010 - 01:10 Link
Wrote some things down before seeing the photos, but that's quite strange.

Could it be that it only shows up with photos with a small aperture (high f-number) if so then I've an idea what it might be.
Probably some dirt on the sensor or on the filter but the later is less likely.
Stefan
Comment Image

K10D, K5
DA* 16-50, DA* 50-135, D-FA 100 Macro, DA 40 Ltd, DA 18-55
AF-540FGZ
bychan
Posted 09/03/2010 - 01:43 Link
I've added two more images, taken on the same day as all but one of the others. These were at a focal length greater than 18mm, and the left hand corner is clean (the others were shot at 18mm).

Adrian
K5IIs, Sigma 10-20, Pentax DA 16-85, Pentax DA 55-300, Pentax 70 Ltd, Metz 44 AF-2.
Flickr: https://www.flickr.com/photos/ambott/
gartmore
Posted 09/03/2010 - 08:46 Link
I'm quite sure that it is the filter that is in shot since it disappears as you zoom in. Stefan is right in suggesting that it will be more pronounced at small apertures. TBH it is hardly notiecable and easily sorted in PP.
Ken
“We must avoid however, snapping away, shooting quickly and without thought, overloading ourselves with unnecessary images that clutter our memory and diminish the clarity of the whole.” - Henri Cartier-Bresson -
johnriley
Posted 09/03/2010 - 09:06 Link
If you use a polarizer on a wide angle lens, the saturation of the sky will vary across the frame because the polarization is at maximum 90 degrees to the sun and falls off as we move away. The more breadth of sky in the image, the more we will notice the variation across the frame.

Personally, as ultra wide lenses tend to vignette slightly anyway, I don't use a PL on them at all. On 35mm film, a 28mm lens is the widest I would use a PL on. A 24mm lens shows the variation quite clearly.

The 12-24mm DA lens, for example, will show nicely darkened corners in ultra-wide landscapes. This is exactly what we used to print in when making black and white darkroom prints. Darkened corners concentrate the eyes towards the centre of the image.
Best regards, John
iceblinker
Posted 09/03/2010 - 10:25 Link
It's normal for an object that is causing slight vignetting to disappear as you zoom in. Just try obscuring the corner of the frame with your finger near to the lens, then zoom in.

What confuses me is that only one corner is affected in Adrian's examples. But perhaps the sensor is not quite aligned correctly?

Anyway, a thin polariser is indeed less likely to vignet than a thick one, so give one a go if you do want to use a polariser at 18mm.
~Pete
Edited by iceblinker: 09/03/2010 - 10:27
Anvh
Posted 09/03/2010 - 12:35 Link
gartmore wrote:
I'm quite sure that it is the filter that is in shot since it disappears as you zoom in. Stefan is right in suggesting that it will be more pronounced at small apertures. TBH it is hardly notiecable and easily sorted in PP.
But strange it's only the top left corner, if it was the filter I would expect it to effect all corners, that's why I said some dirt on the sensor.
Just very strange...
Stefan
Comment Image

K10D, K5
DA* 16-50, DA* 50-135, D-FA 100 Macro, DA 40 Ltd, DA 18-55
AF-540FGZ
iceblinker
Posted 09/03/2010 - 13:24 Link
^ Nice pic, but that's got nothing to do with Adrian's problem.
~Pete
Anvh
Posted 09/03/2010 - 13:53 Link
Yes Gareth but it does not completely blank a corner but only darkens it.

if you look at the photos from Adrian you know what he means.
Comment Image

Different right:

The polarization darkens the sky on the right side and he has a completely dark corner on the left so it certainly isn't the polarization effect.
Stefan
Comment Image

K10D, K5
DA* 16-50, DA* 50-135, D-FA 100 Macro, DA 40 Ltd, DA 18-55
AF-540FGZ
Edited by Anvh: 09/03/2010 - 13:55
johnriley
Posted 09/03/2010 - 20:41 Link
I've explained this effect above. Light is more polarised at 90 degrees to the sun, and progressively less so as you look nearer to the sun. This causes different amounts of darkening across the sky. The wider the angle of the lens, the more this effect becomes visible.
Best regards, John
iceblinker
Posted 09/03/2010 - 20:54 Link
That's not Adrian's problem. Look at the extreme top left corner. There's a tiny fairly sharp dark triangle there.
~Pete
bychan
Posted 09/03/2010 - 23:42 Link
Thanks for all your replies and advice, greatly appreciated.

I imagine that if this were a dust problem then all images would be affected (been there as well); touch wood, as far as I can see at present the sensor seems clean.

What I can now also add is that images taken without the cpl are clean.
I've uploaded three new images to the flickr site, all shot at 18mm but without the cpl (two taken with a Cokin P filter).
link

As John Riley specifies, he never uses a cpl on a ultra wide angle lens; this also applies to my Sigma 10-20, and the results are excellent.
I can sense an experiment approaching, results with/without the cpl.

As I specified originally, the triangle is very small, and it has been suggested that it will be easily removed in PP (not sure how, crop perhaps).

PP seems much more viable than the cost of a slim cpl.

Again looking back on old images taken with the Tamron 18-250, using either a cpl or circular GND, the problem is much worse, some examples within here link

So I think that the conclusion is that the cpl causes slight vignetting at the wider focal lengths.

Again thanks all for your input and help/advice, greatly appreciated.

Regards
Adrian
K5IIs, Sigma 10-20, Pentax DA 16-85, Pentax DA 55-300, Pentax 70 Ltd, Metz 44 AF-2.
Flickr: https://www.flickr.com/photos/ambott/
Anvh
Posted 10/03/2010 - 00:16 Link
Glad you found the problem, slim filter would solve it.

bychan wrote:
As I specified originally, the triangle is very small, and it has been suggested that it will be easily removed in PP (not sure how, crop perhaps).
Oh no you clone it out, just like dolly the sheep
Most software for photography has spot/dust removal tool you could use that and a cloning tool works the same.

bychan wrote:
Again looking back on old images taken with the Tamron 18-250, using either a cpl or circular GND, the problem is much worse, some examples within here link
Yeah had the same problem if I stacked two filters, but like your image i got it on all the corners and not just one, that's so strange about it
Stefan
Comment Image

K10D, K5
DA* 16-50, DA* 50-135, D-FA 100 Macro, DA 40 Ltd, DA 18-55
AF-540FGZ
Edited by Anvh: 10/03/2010 - 00:16
melness
Posted 21/03/2010 - 07:40 Link
Just a thought mind ,but you could use a step ring ...ie,a 77mm to a72mm ,that's what i use no problems
Thanks
Sean

Add Comment

To leave a comment - Log in to Pentax User or create a new account.



Proudly supporting Pentax User

Samsung Logo Asahi Pentax Logo