I'm wondering if anyone can help me with a problem I'm getting concerning my Canon 350D.
Below are two images I took today, the first image is just a close up of the second. In the first image you can see the white thing that the eagle is standing on has a very distinct blue flare around it.
Was thinking the same thing matt, very bright/white. Is it the kit lense Peter? I had the same problem with mine although it is still a great lense for starting out. As others have said fiddle with your settings/exposure and grab a good quality UV & Polarising filter.
Thanks for the ideas! The lens I used is a Canon Ultrasonic 75-300mm. It is an older lens (5 or 6 years old now?), one I got for my Canon 500N, but still quite serviceable. I've had a flick through other images I've taken with a high contrast or bright light scenario while using that lens and the problem is evident (to a lesser degree) in those images.
I tend to agree with chromatic abberation theory and will toddle off to a camera shop for a UV polarizing lens.
I'll also have a crack at changing the settings (exposure times, aperture etc.). I had been using the camera on automatic when I spotted the bird and wanted to snap off a few before it flew away. I should have realised it was going to be there for some time because it was drying its wings.
It is chromatic aberration and a filter won't fix it. Just so you don't think this is normal, the Canon 75-300 has a pretty bad reputation for image quality.
What may help is reducing the aperture size.
Very good image quality at wide open on a zoom costs quite a bit.
APO, low dispersion, fluorite and all that stuff is about making chromatic aberration disappear.
It seems to be a problem with some digital cameras (not just canon). It is more prevalant on higher MEG camera's. Chromatic adderation is all but invisable on all film SLR's I ever used.
This effect is specifically referred to as "purple fringing". It is quite a common fault. Unfortunately, the current consensus is that neither the lens, nor 'density' OR the quality of the photo sensor are alone responsible for this effect.
There are many websites dedicated to avoiding and minimising the effect.
We do know that is most likely to happen when apertures of greater than f/4 are used, when a wide angle of 28mm (or equivalent) or wider is used and when the subject to background contrast is high. Since it is most likely to happen under those conditions, avoid that combination if possible
Good news is that the effect can be removed quite effectively a lot of the time in post production, decent article from PC World about the technique:
Being an SLR it's quite easy to sort out the culprit. Stick on an expensive telephoto lens and we'll see whether it's the camera that's at fault. Considering that I've never heard of it being the camera's fault and that I don't have any purple fringing, my bet is on the lens being the problem. Especially since that one is known to be bad. http://www.fredmiranda.com/reviews/s...t=27#poststart
Even my $300 Sigma 70-300 APO shows negligible amounts of CA. My Sigma 100-300 F4 EX has none.
"We do know that is most likely to happen when apertures of greater than f/4 are used, when a wide angle of 28mm (or equivalent) or wider is used"
Being a 75-300 it was nowhere near 28 mm or under and the aperture was at F4 or smaller.
Well, it does say that those factors 'contribute', not 'solely responsible'. It looks like several people have issues with the 75-300, even tho it doesn't meet the 'stereotype' mentioned in the other article I linked...
The Canon 75-300 can be had for around £100 second hand, and should cost no more than £150 new. It's light, pretty sharp, suffers badly from flare and chromatic aberration (purple fringing of highlights). If you buy the (optional!) lens hood you will have less problems from flare but the purple fringing is a pain to post-process out of your shots.
You can almost be sure about images turning out with purple fringing if you are shooting something with sky as the background.
Some websites suggest it is a combination of factors, some are convinced it is a "sensor density/voltage leak problem", some are convinced it's the lens, but none of those people are real experts, so they are really just making assumptions based on their own personal experience. I'd be careful about blaming the lens alone, I guess the only way to be sure is as the post mentions above, try to take some test shots with a REALLY good lens and see if the problem persists. The problem is getting your hands on one without buying it, just to find out that wasn't the problem after all...
Any reasonable camera store will let you test lenses. Not many have a selection of the expensive telephotos in stock and on display, but many have the 70-200 L (either F2.8 or F4) which is sufficient for this test IMO.
Have a look at my reasoning for the lens being the problem. I don't think I'm making such a leap of faith by blaming the lens:
There's a whole expensive market dedicated to build quality, sharpness and elimination of chromatic aberration. The special materials like low dispersion glass and fluorite are there just to correct CA, no other reason.
When you buy these lenses you don't get CA.
By sticking a teleconverter on my 100-300 I get some CA. Same sensor, same photo, different lens configuration.
A cheap refractor or binoculars will give you CA visible to the eye. No voltage leaking here. In fact, if you are very picky it takes an extremely expensive refractor to avoid CA. Same deal as camera lenses except camera lenses aren't expected to be diffraction limited.
I hope these theories of the sensor being the problem aren't based on point and shoot cameras. Just because the megapixel war has pushed tiny sensors up to 8 MP doesn't mean the lenses are of a comensurate quality.
Many thanks for your debate as it has opened my eyes to the quality involved with camera lenses.
For my level of photography, this CA problem is nothing more than a minor issue to be dealt with at leisure by post processing (if I really feel like it). I tend to believe that it is the lens at fault as opposed to the camera body simply because I have not noticed the problem in images taken with the 18-55 or 28-80. When time permits I'll go through my film images and pick out any 75-300 shots and try to pick out the CA...although it could be hard in a standard 6x4 print.
