[Anya]

Is purple fringing the same as Chromatic Abberation? I was interested because I am thinking about purchasing a new lens for my Asahi Pentax SP.

Also does purple fringing matter on film cameras? I use mostly black and white film and rarely notices it. Does it matter much if I switch to digital in the future?

[MPrince]

Purple fringing is a type of chromatic aberration that can occur in high contrast scenes and is more likely to occur with more open apertures.

For more information, google the phrase "purple fringing" and you will be inundated with search results.

[Robert S Donovan]

Like Matt said, purple fringing is a form of chromatic aberration. It can have an effect on film shots if the lens is particularly bad but its effect is more pronounced on digital due to the sensors' sensitivity to UV light. Digital specific lenses (ie: Pentax DA series) feature specific coatings to filter out the specific wavelengths of light that cause CAs on DSLRs.

CAs in general are less of a concern when shooting B&W since you don't see the color but if it's really bad it will show up as ghosting and fuzzy edges. Even a lens that exhibits particularly bad fringing can work fine on a DSLR if you just avoid high contrast (backlit/black on white) scenes.

If the lens is a good deal and fits your budget I say give it a try!

[Lowell Goudge]

Keep in mind that CA or chromatic aberration, is caused principally in the out of focus areas on lenses that are not aprochromatic, i.e. lenses that only have all the colors converge at the point of focus, but in out of focus areas the colors are not alligned, forming somewhat of a ghost.

While lens coatings may help on purple fringe, they do not really help on CA because that is a function of the lens shapes not UV light. CA usually shows up as a green fringe on one side and red on the other,

[wildman]

1. Taken with a two element achromatic lens (crown+flint glass) Notice the classic purple-green out of focus fringing on the thin branches.
2. Taken with same above lens used carefully to avoid CA
3. Taken with a "perfectly" corrected 3 element APO triplet objective using Schott (Zeiss) CaF2 Fluorite.

Bottom line:
Whither using film or digital, monochrome or color, CA always results in at least some loss of contrast even under ideal conditions.

NOTE:
This are all very long FL glass (over 500mm) at high magnifications where CA can be particularly hard to control. The first shot, for instance, was taken at an equivalent FL of 2000mm (60x).

[stevebrot]

A small note. I have no desire to contradict the previous comments, but PF is not CA per se. Probably the best evidence for this assertion is that it does not happen with film cameras. I have yet to read a good explanation of the actual physics except that the general consensus is that it is due to an interaction between the lens and sensor under specific lighting conditions.

Some lenses are more prone to PF than others, but conditions such as those in wildman's first shot are typical. The classic examples:

• Specular highlights on wavelets in water
• Backlit branches
• Sunlit curved chrome surfaces
• Specular highlights on crumpled cellophane (candy wrappers)

PF is not limited to out-of-focus areas and does not go away at smaller apertures. (I can provide examples of PF for both cases.)

Unlike CA, PF is fairly difficult to correct in post-processing. At least, I have not found an easy method short of subtracting out the color in question.

Steve

[stevebrot]

1. Taken with a two element achromatic lens (crown+flint glass) Notice the classic purple-green out of focus fringing on the thin branches.
2. Taken with same above lens used carefully to avoid CA
3. Taken with a "perfectly" corrected 3 element APO triplet objective using Schott (Zeiss) CaF2 Fluorite.

Bottom line:
Whither using film or digital, monochrome or color, CA always results in at least some loss of contrast even under ideal conditions.

NOTE:
This are all very long FL glass (over 500mm) at high magnifications where CA can be particularly hard to control. The first shot, for instance, was taken at an equivalent FL of 2000mm (60x).

Wildman,
Those last two are gorgeous photos (as always). You really should publish your work!

Steve

[Canada_Rockies]

A small note. I have no desire to contradict the previous comments, but PF is not CA per se. ...<snip>... PF is not limited to out-of-focus areas and does not go away at smaller apertures. (I can provide examples of PF for both cases.)

Unlike CA, PF is fairly difficult to correct in post-processing. At least, I have not found an easy method short of subtracting out the color in question.

Steve

I have read that purple fringe is sensor based chromatic aberration caused by certain lens/sensor combinations. It is very difficult to remove, as pointed out above. I have not yet found any way of working with PF that is satisfactory.

[wildman]

A small note. I have no desire to contradict the previous comments, but PF is not CA per se.

Your point is well taken and understood.

Generally PF is thought to be a subset of CA (lateral chromatic aberration) but other factors may be involved.

But this is a beginner's forum and I didn't want to put too fine a point on it and only confuse the original poster. That's why I posted actual examples rather than a long arid technical explanation.

Sometimes one picture really is worth a thousands words.

BTW my work is published.

Keep clicking away out there...

[Marc Sabatella]

I have no desire to contradict the previous comments, but PF is not CA per se. Probably the best evidence for this assertion is that it does not happen with film cameras.

Really?!?! I've seen discussions about the role the sensor plays in PF, and never really followed them (optics isn't my thing), but I hadn't gleaned from those those that film was totally immune. So, for example, even Tamron 70-300 won't have PF on film?

[wildman]

RE:Software defringing.

I don't have the original file anymore but here is a quick and dirty attempt on the thumbnail I posted. It's still massively poorly exposed but there you are.

[OregonJim]

Really?!?! I've seen discussions about the role the sensor plays in PF, and never really followed them (optics isn't my thing), but I hadn't gleaned from those those that film was totally immune. So, for example, even Tamron 70-300 won't have PF on film?

FWIW, I've never seen PF on film. From what I understand, it is a Bayer-pattern specific phenomenon that is aggravated by certain lenses that are not apochromatic - that is, not all of the color spectrum is focused on the same plane.

[stevebrot]

Really?!?! I've seen discussions about the role the sensor plays in PF, and never really followed them (optics isn't my thing), but I hadn't gleaned from those those that film was totally immune. So, for example, even Tamron 70-300 won't have PF on film?

I have shot film for years and never saw nor heard of PF before I bought my first digicam (Canon G2) several years ago. I still shoot film and while my Zenitar 16/2.8 is prone to PF on the K10D, I have never seen it on film shots with the same lens. I have to admit though to not having done any formal experiments.

Anyone out there had problems with PF on film?

Steve

[stevebrot]

FWIW, I've never seen PF on film. From what I understand, it is a Bayer-pattern specific phenomenon that is aggravated by certain lenses that are not apochromatic - that is, not all of the color spectrum is focused on the same plane.

That was my understanding as well, but was not wanting to cloud the waters with too much technical talk.

Steve

[wildman]

that is, not all of the color spectrum is focused on the same plane.

Neither is it with APO glass only red, green and blue and with achromatic glass red and blue at least for glass intended for visual use only. As far as I understand it.