[wim]

[quote name='Sylvain' timestamp='1291158324' post='4616']

Wim, even though you wrote "'nuff said", I'm afraid you did not say anything :/. I did find the article was not well put together (no explanation whatsoever on graph data collection, etc...) but that what it's trying to say could be rather interesting. Or did I misunderstand everything?

DxO still inks to this article and what I could find in the LL forum wasn't really saying anything else than "no explanation". Care to enlighten me? Doesn't have to be a Wim Encyclopedic answer



TIA,

S.

[/quote]

Hi Sylvain,



I didn't want to butcher DxO, Mark Dubovoy or LL more than already was done, and I was rather disappointed with the matter, hence my cryptic message.



It really did it for me with the claim that there was no use for minDoF type application of lenses, because you wouldn't get less DoF anyway, or something to that extent. As I own a fair number of large aperture lenses, two of which are the fastest AF production lenses available for dslrs, I'd have to call it complete male bovine manure as a result.



Of course, it may well be that camera manufacturers have a few tricks up their sleeve to manipulate the results from a sensor, but the article didn't make sense at all. One thing that is interesting however is that PF or purple fringing, which is an aberration oftentimes considered to be CA (which it isn't by the definition of CA, because it isn't sagittal or transversal in nature), is much stronger than it is with film, and always disappears when stopping down a few stops. Now, this could either be caused by someting to do with the sensor assembly itself and extremely high contrast transitions, the A/D processing, both of these, or generally indeed by anything that causes an increase in contrast due to signal amplication, whether optical or electrical/digital and the signal clipping in a weird way. Since we are talking signal amplification here, that could be caused by this, it it were true.



As BC and GTW indicate, all of this could actually be tested quite easily, by breaking the contacts between lens and camera and just shooting at maximum aperture. I guess I may give this a try myself, too <img src='http://forum.photozone.de/public/style_emoticons/<#EMO_DIR#>/biggrin.gif' class='bbc_emoticon' alt='' />.



Kind regards, Wim



P.S.: sorry for my late reply - I am in a new assignation since about 3 1/2 weeks, and I generally have bad to very bad and intermittent or even no internet connections during the week, and currently only very little time during the weekend.

[genotypewriter]

[quote name='wim' timestamp='1291334181' post='4720']

One thing that is interesting however is that PF or purple fringing, which is an aberration oftentimes considered to be CA (which it isn't by the definition of CA, because it isn't sagittal or transversal in nature), is much stronger than it is with film, and always disappears when stopping down a few stops. Now, this could either be caused by someting to do with the sensor assembly itself and extremely high contrast transitions, the A/D processing, both of these, or generally indeed by anything that causes an increase in contrast due to signal amplication, whether optical or electrical/digital and the signal clipping in a weird way.[/quote]

PF is a really weird thing... it's both lens and sensor related. I say lens related because well-corrected lenses (e.g. 125 f/2.5 APO-Lanthar, 200 f2L IS, 200/4 Micro-Nikkor and I'm sure the 60 f/4 UV-VIS-IR too) don't show it. But it is also a sensor related thing because lenses like the 85 1.2L wide open [url="http://www.flickr.com/photos/genotypewriter/4635819634"]show it on digital[/url] but [url="http://www.flickr.com/photos/genotypewriter/4495611132/"]not on film[/url].



I think calling it CA is fairly wrong because CA is caused by refraction of light of different wavelengths. But when light is passing through the center of a lens, there's no refraction so there can't be CA but you can often see [url="http://www.flickr.com/photos/genotypewriter/4635819634/sizes/o"]PF in the center[/url]. Similarly I haven't seen PF getting worse towards the corners like CA typically does.





[quote name='wim' timestamp='1291334181' post='4720']

As BC and GTW indicate, all of this could actually be tested quite easily, by breaking the contacts between lens and camera and just shooting at maximum aperture. I guess I may give this a try myself, too <img src='http://forum.photozone.de/public/style_emoticons/<#EMO_DIR#>/biggrin.gif' class='bbc_emoticon' alt='' />.

[/quote]

I just did earlier this morning before turning up to work... hmm the early findings are a bit shocking and not in favour of us DxO-skeptics <img src='http://forum.photozone.de/public/style_emoticons/<#EMO_DIR#>/sad.gif' class='bbc_emoticon' alt='' /> I'm going to do a thorough test and upload an article within the next few days, hopefully.



GTW

[popo]

[quote name='genotypewriter' timestamp='1291349452' post='4722']

PF is a really weird thing... it's both lens and sensor related. I say lens related because well-corrected lenses (e.g. 125 f/2.5 APO-Lanthar, 200 f2L IS, 200/4 Micro-Nikkor and I'm sure the 60 f/4 UV-VIS-IR too) don't show it. But it is also a sensor related thing because lenses like the 85 1.2L wide open [url="http://www.flickr.com/photos/genotypewriter/4635819634"]show it on digital[/url] but [url="http://www.flickr.com/photos/genotypewriter/4495611132/"]not on film[/url].



I think calling it CA is fairly wrong because CA is caused by refraction of light of different wavelengths. But when light is passing through the center of a lens, there's no refraction so there can't be CA but you can often see [url="http://www.flickr.com/photos/genotypewriter/4635819634/sizes/o"]PF in the center[/url]. Similarly I haven't seen PF getting worse towards the corners like CA typically does.[/quote]

I thought PF was LoCA. At least in the specific case of the Canon 85mm f/1.8, which shocked me how strong it was when I first got it. It looks fine until you hit a highlight. Any bright spot and bang, purple or green fringes depending on if it is behind or in front of the focal plane.



It also fits that true APO lenses wouldn't suffer from it. I don't have experience on film to make any claim there though.

I just did earlier this morning before turning up to work... hmm the early findings are a bit shocking and not in favour of us DxO-skeptics <img src='http://forum.photozone.de/public/style_emoticons/<#EMO_DIR#>/sad.gif' class='bbc_emoticon' alt='' /> I'm going to do a thorough test and upload an article within the next few days, hopefully.

I don't want to put words in your mouth, but that reads as you found indication confirming DxO's claims. Of course I'll wait and see the full report.

[mst]

[quote name='popo' timestamp='1291365255' post='4726']

I thought PF was LoCA.

[/quote]



Nope. Unlike LoCA, PF happens at the focal plane, too. And regardless of the position, it's always purple.



Two different animals. Similar in morphology, but definitely not the same species <img src='http://forum.photozone.de/public/style_emoticons/<#EMO_DIR#>/wink.gif' class='bbc_emoticon' alt='' />



-- Markus

[genotypewriter]

[quote name='popo' timestamp='1291365255' post='4726']

It also fits that true APO lenses wouldn't suffer from it.[/quote]

It has a lot to do with lens correction. Maybe it's related to the angles at which light hits the sensor... but that still doesn't explain why there can be PF in the center and not get worse toward the corners. Wish someone could chime in on this... I'm a bit lost apart from the observations I've made.





[quote name='popo' timestamp='1291365255' post='4726']

I don't want to put words in your mouth, but that reads as you found indication confirming DxO's claims. Of course I'll wait and see the full report.

[/quote]

From my early measurements, this actually is the indication. But I want to be sure... what they said was that manufacturers alter ISOs in fast aperture lenses due to the light loss. This implies it's a f-number specific thing. I need to check for that. And I want to check if they do this only for wide lenses more than long ones. That's about 4 tests right there.



GTW

[popo]

[quote name='genotypewriter' timestamp='1291373550' post='4732']

It has a lot to do with lens correction. Maybe it's related to the angles at which light hits the sensor... but that still doesn't explain why there can be PF in the center and not get worse toward the corners. Wish someone could chime in on this... I'm a bit lost apart from the observations I've made.[/quote]

Longitudinal CA is due to the different focus planes for different wavelengths of light right? That applies regardless of where in the frame you are hence can apply equally in the middle as well as the edges. Actually, that could explain why you can also get PF on the (apparent) focal plane as commented by Markus, since there is actually not one but a region dependant on wavelength. Or am I missing something here?



Looking up some samples, I got my colour direction mixed up. It is purple on or in front of the focal plane for the Canon 85mm f/1.8, and green behind. Some early test samples:











Note these may be cropped so look worse than if I had kept full frame. Is this LoCA or PF or are they the same thing?

[Brightcolours]

http://toothwalker.org/optics/chromatic.html

[Guest]

[quote name='Brightcolours' timestamp='1291380855' post='4734']

http://toothwalker.org/optics/chromatic.html

[/quote]



Interesting. However I'm a bit confused by the following (longitudinal chromatic aberrations):



"As the human eye and autofocus systems are particularly sensitive to green light, both manual focus and autofocus tend to bring the green image in sharp focus. The other colors of the spectrum are left defocused and add up to a magenta fringe. For this reason purple fringing is more common than green fringing."



Is that so?

[popo]

[quote name='Lomskij' timestamp='1291400276' post='4741']

Interesting. However I'm a bit confused by the following (longitudinal chromatic aberrations):



"As the human eye and autofocus systems are particularly sensitive to green light, both manual focus and autofocus tend to bring the green image in sharp focus. The other colors of the spectrum are left defocused and add up to a magenta fringe. For this reason purple fringing is more common than green fringing."



Is that so?

[/quote]

Remember in the bayer pattern you get as twice as many green detectors as you do of red or blue individually.

[Brightcolours]

[quote name='Lomskij' timestamp='1291400276' post='4741']

Interesting. However I'm a bit confused by the following (longitudinal chromatic aberrations):



"As the human eye and autofocus systems are particularly sensitive to green light, both manual focus and autofocus tend to bring the green image in sharp focus. The other colors of the spectrum are left defocused and add up to a magenta fringe. For this reason purple fringing is more common than green fringing."



Is that so?

[/quote]

Is what so?



Nature is "green". evolutionary our vision is more sensitive to greens, we are able to detect more green shades than other colour's shades.



This makes scientists, developers and manufacturers to put the biggest importance in greens. For RGB sensors (for instance with bayer CFA) it is not by accident that they have chosen to make 2 out of every 4 sensels detect green. And it is probably also true that most AF systems try to get green in focus.