Cause of Fringing

kfong wrote in post #5391248
Still don't explain how the purple part of a white light hitting a microlens (even at an acute angle) managed to fall into ONLY blue and RED neigbouring photo-sites to generate purple fringes.

Well, red and blue light will both be filtered out by the green filters over the green photo-sites.

Well now i'm more confused than ever. Why would Canon be addressing the fringing issue with Flourite in the 60's and 70's if it's caused by sensors? They're weren't any sensors back then.

Chromatic aberration isn't quite the same as purple fringing; CA shows a whole rainbow at its extreme; purple fringing is mainly purple and green. Fluorite was used for classical CA, not the purple fringe.

Ahh, I get it now. Thanks.

Jon wrote in post #5394326
Well, red and blue light will both be filtered out by the green filters over the green photo-sites.

Yeah, but by the same token red and green will be filtered out by the blue filters, and green and blue by the red filters, so I don't think that's the answer.

However, it does make me wonder if it could somehow be related to there being 2 "green" photosites for each red and blue. I can't form a reasonable hypothesis on exactly how though.

Terrywoodenpic wrote in post #5393432
In digital photography the sensor micro-lenses are the main cause see...
http://en.wikipedia.or​g/wiki/Talk:Purple_fri​nging

I think I'm getting it. According to the above Wiki, purple fringing in micro-lenses are due to red and blue lights scattering off the larger numbers of green filters in a Bayer pattern. If this is the case, does it mean that the filters are interference filters instead of absorption filters ? I thought interference filters are typically narrow band and results in poorer overall quantum efficiency?

Ken

Here is an example of what appears to be purple fringing. The image was made with my XTi and EF 75-300mm f/4-5.6 III lens. It was shot into the late afternoon sun so there are a lot of saturated or nearly saturated reflections on the water ripples. This situation resulted in the GBH in the foreground being very underexposed so there are some very strong contrast areas where the "purple fringing" shows up. The second image is a 300% crop from the area where the effect is most noticeable.

HOSTED PHOTO
please log in to view hosted photos in full size.

1. A lens problem such as flare from internal light scattering, although I can't explain why it would be purple.
   
2. Sensor sites are at or near saturation and the excess charge is affecting the surrounding sites that are relatively dark. Again, I don't see an obvious connection to this causing a purple color.

HOSTED PHOTO
please log in to view hosted photos in full size.

Just my opinion, but the better the glass, the closer you are to the sweet spot of the lens (2-3X stopped down from max aperture), & the more in focus the area you're looking at is, the less artifacts you will see.
There's a reason that "L" glass costs a lot.

Just saw this thread and for whatever reason, my 40D does show purple fringing on some shots when I crop them fairly heavily or look at them at 100% in CS3. I never saw any of this with the 30D or my 1D classic using the very same lenses.

You probably just notice or CAN see it now.

I see purple fringing on TV, in movies, in my old film shots. I try to avoid it by having good lenses. Try.

Optics are always a compromise. The usual compromise is $$$$.

No, I said I just saw this thread, not the purple fringing. And while I do not see it on my TV, movies, old film shots or Polaroids of my old Uncle Cliff, I do see it on my newish 40D on some shots taken against heavy backlighting.

And I'm not new to shooting towards heavy backlighting, but I never saw it on my 30D shots or the 1D shots.

I notice it everywhere. Older Kubrick movies especially. Telephoto TV news shots...

Go back and look at your 30D and 1D shots. Did you changes lenses? Did you always shoot stopped down back then?

I'm just saying, factor in the largest variable here, human perception.

I know what your saying. But purple fringing is like a slap in the face. It's not something you would overlook when it happens. Looking back, I guess it should be surprising that it didn't happen more often to me, cause I do shoot "into the sun" quite often, and in fact, I have seen fringing on others shots, which turned out to be older lenses or non-canon lenses. I even have a PS tutorial bookmarked on how to get rid of it. It just never happened to me, until the 40D, and I have not bought any new glass in a long time, and my shooting style hasn't changed, so it's not that.

Unfortunately, I delete all that trash, so I don't have any to show. But I will no doubt get some more, as soon as the sun comes out again, which could be a long time, give we have snow flurries today.

canonloader wrote in post #5407995
I know what your saying. But purple fringing is like a slap in the face. It's not something you would overlook when it happens. Looking back, I guess it should be surprising that it didn't happen more often to me, cause I do shoot "into the sun" quite often, and in fact, I have seen fringing on others shots, which turned out to be older lenses or non-canon lenses. I even have a PS tutorial bookmarked on how to get rid of it. It just never happened to me, until the 40D, and I have not bought any new glass in a long time, and my shooting style hasn't changed, so it's not that.

This is about what the disgruntled poster said on the thread I quoted. I'd never seen or heard it mentioned before, but keep in mind although it was on a Canon forum, it was in response to a post that referenced a blog about Canon's quality control. The OP's blog cited poor QC, but as many pointed out, he didn't substantiate his claims.

The thread has been completely removed (at least I can't find it), no doubt because it elicited some pretty childish and nasty reponses, but perhaps also because the site has Canon advertising.

Well, whatever the cause, it's fairly common or there wouldn't be tutorials on how to fix it.