[slackercruster]

With film, chromatic aberrations were not a big deal. With digital they very noticeable. Why the difference?

[joe.penn]

With film, chromatic aberrations were not a big deal. With digital they very noticeable. Why the difference?

The way the digital sensors interpret the light - also, the aberrations will change from sensor to sensor, about crapped my pants when I first saw the aberrations with my VS1 2.8 210 on the K5, same shot with that lens produced little to none on the Kx sensor...

[Adam]

Film was a lot more tolerant, plus nowadays people pixel-peep, whereas you couldn't really do that in the film era.

[RobG]

Film was a lot more tolerant, plus nowadays people pixel-peep, whereas you couldn't really do that in the film era.

Funny thing - I never noticed the chromatic aberration on my film-era 28mm Makinon lens until I scanned some photos and looked at the detail. It was disappointing, because the lens is very sharp. However, my main point here is that I agree with Adam - the aberrations were there, but we didn't look at the details so easily or so often.

[PPPPPP42]

Just to clarify the question, are you asking why digital cameras have more chromatic aberrations with new lenses built for them or with old lenses, high end lenses or cheap ones? Also are you clear on the difference between C.A. and purple fringing? I sometimes get purple fringing with wildly different contrasts (recently with back lit stain glass in a church) but closing a stop or two clears it right up.
Not sure about C.A. since I never really see any of that even with my old A lenses I use. Anything but new glass optimized for the requirements of digital would in theory cause more C.A. They've both improved with computer design and tweaked in general the characteristics of new stuff to keep the sensors happier.

Plus what Adam said, film is just better.

EDIT: Ok I guess these old A's do make some C.A. (depending on contrasting light) I just had to zoom WAAAAY in to see it (like Mario Brothers level of pixelation zoomed in), and since I am in the habit of cropping by zooming or just moving when using a prime to get better quality full size photos, I just never saw it before.

[RobG]

Also are you clear on the difference between C.A. and purple fringing? I sometimes get purple fringing with wildly different contrasts (recently with back lit stain glass in a church) but closing a stop or two clears it right up.

I thought what you are describing above is an aspect of chromatic aberration, so apparently I'm not clear on the difference. FWIW I just read the review of the DA12-24 which apparently has CA that is not improved by stopping down.

[JohnBee]

With film, chromatic aberrations were not a big deal. With digital they very noticeable. Why the difference?

The focal plane tolerance is much lower with digital than it ever was with film.

[stevebrot]

I agree with Adam - the aberrations were there, but we didn't look at the details so easily or so often.

...What he said...

What we can see with pixel-peeping on digital would have required a microscope with film or a very high-power loupe. It was only as I started scanning some of my old slides that I realized how poor many of the images actually are and how often CA is part of the "picture" (so to speak).

It seems that our tolerance for imperfection is much less now that we have such great tools for evaluation!


Steve

[RobG]

It was only as I started scanning some of my old slides that I realized how poor many of the images actually are and how often CA is part of the "picture" (so to speak).

Out of interest, what are you scanning with? I don't think that the image on the film is necessarily as poor as scanning makes it seem. The interaction of film grain with scanner technology has a tendency to make the faults in film look much worse than when the image is printed onto photographic paper. Particularly my scanner - a Nikon LS30 - which uses LED technology that makes the grain look worse and highlights cuts and scratches (while providing an IR channel that allows a lot of the imprefections to be removed at a cost). I don't know if you've ever tried scanning a printed photo from a magazine on a flatbed scanner? The scanner will generate moire patterns from the offset colour dots on the page. The same sort of thing happens when a scanner sees film grain. But sorry, I digress - we were talking about chromatic aberration not the pecularities of film scanning. Your main point was that once scanning, you discovered that there was often CA that wasn't obvious in the past, and I agree.

It seems that our tolerance for imperfection is much less now that we have such great tools for evaluation!

Very true! Good for the equipment manufacturers too, since our passion for perfection keeps their R&D and sales going.

[RioRico]

With film, chromatic aberrations were not a big deal. With digital they very noticeable. Why the difference?

Maybe because digital sucks and we all need go back to Kodachrome. Oops, too late.

[PPPPPP42]

I suppose purple fringing is a type of chromatic aberration, and usually you get it on one side with another color on the other side often yellow (like when you look at the lines on the bottom of a swimming pool) but this is straight purple fringing https://www.pentaxforums.com/forums/photographic-technique/179889-dont-get-bl...-fringing.html which is caused by a dark object with a super bright background (or sometimes a white dress with dark background, think weddings) confusing the hell out of the digital sensor, better lenses seem to have less of this, I only get it wide open with white and dark contrasts. I'm no engineer or scientist so I couldn't say why anything does that.

[ElJamoquio]

Film was a lot more tolerant, plus nowadays people pixel-peep, whereas you couldn't really do that in the film era.

It's very easy to crop nowadays, too - incredibly easy - and there's already a crop with the Pentax sensors. So instead of 200mm we're using the lens as if it were 500 mm, etc.

[stevebrot]

Out of interest, what are you scanning with? I don't think that the image on the film is necessarily as poor as scanning makes it seem.

Nikon 5000 ED...and I am pretty proficient in its use.

The good images look great. The bad ones...well, it is enough to say that things like CA are plainly visible at the high resolution that my monitor offers and that focus error or camera motion are also incredibly troubling on screen for images that were perfectly acceptable when projected.


Steve

[RobG]

Nikon 5000 ED...and I am pretty proficient in its use.

Nice! No criticism of you intended. I was just talking about how a scanner can make film images look worse than they really are. The only film which has ever scanned really well on my LS30 is Fujichrome 100F because the grain is so fine that the scanner doesn't see it (which makes me curious how much more image information might be there if scanned at 4000ppi instead of 2700ppi). Unfortunately shadow details in slide film tend to be lost because the LED light source simply isn't bright enough to penetrate the slide. It's sad, because the scanner can't reproduce what I can see if the slide is projected.

The good images look great. The bad ones...well, it is enough to say that things like CA are plainly visible at the high resolution that my monitor offers and that focus error or camera motion are also incredibly troubling on screen for images that were perfectly acceptable when projected.

I know what you mean. Some of the images from my DSLRs which I have deleted probably would have been keepers with a film camera because I wouldn't have been so aware of fine deficiencies. On the other hand, because I have the option of deleting them at zero cost, I've taken a lot of photos which I would not have taken with film. The DSLR also has image stabilisation which made some images possible that would have been much more difficult to achieve on film.

My understanding is that most colour digital sensors use an array of elements with a colour filter - each element in the array doesn't pick up every colour, but is actually delivering intensities of red, green or blue. As a result, any image from the sensor is actually a matrix of RGB points which the camera firmware has to interpolate into a pattern of RGB pixels. I suspect that the camera firmware could also be guilty of generating artifacts in certain circumstances, or highlighting things like chromatic aberration from the lens.