Can anyone here post some examples of varying color renditions between lenses? To this point, I've always found the idea that the color rendition varies between lenses to be vague at best. It strikes me that the sensor you're using and the lighting conditions are far more important, unless the glass is either extremely good or cheap.
I dont see a color difference with Nikon lenses, whether its a G or AIS lens, they are the same to me. But my Zeiss and (former) VL lenses clearly produce different colors then Nikon. Its more vibrant, darker and stronger.
The times when i notice it the most, is especially at winter and night. I dont remember seeing a difference in daytime.
The materials in a glass can certainly affect colour transmission - even different ions of the same atom/compound can affect the colour transmission greatly. That said, I would think that the manufacturers would be shooting for 100% accurate colour reproduction, and trying to get pleasing levels of contrast and saturation (among other aspects). I only have fundamental knowledge of the physics of optics - it would be nice to hear from an actual expert.
I've never noticed certain lenses producing a colour cast of any sort, but I've never wasted my time looking. As you alluded to, any noticeable colour cast in a pro-level lens would be picked up on quickly and create some serious backlash. I'm approximately 36,000 times more concerned with the vast differences in how an image's colours display on different devices than I am with my lenses affecting the colours.
EDIT: I should note that I've only ever shot with Nikon lenses, so I have no experience with other manufacturers.
Lenses of a brand tend to be similar in their color rendering. Zeiss ZF lenses tend toward warmer than Nikon, which tend towards cool. Sigma lenses came with built in warming filters for years. In general use it's not an issue unless you're using multiple lenses during a shoot and need to have matched results at the end.
This is picky stuff and you're going to need a calibrated monitor and a pro print house to regularly see the differences.
It seems somewhat moot in these days of digital imaging, where the WB setting would have the most impact, either in camera .jpg or in post with RAW. I always adjust color temp. in PP if it helps the image to do it.
Jammy Straub wrote:
Zeiss ZF lenses tend toward warmer than Nikon, which tend towards cool.
I can't agree with that. The Zeiss lenses seems "truer" and letting more nuances through, which might allow you to PP in a way that still looks good when using a warm WB. But the largest difference I can see (without a direct A/B comparsion) is the Zeiss lenses gives nice, blue shadows without affecting the allower color balance. The Nikkors are more "muddy" and the modern lenses tend to look a bit oversaturated when applying hard PP.
pdxflint wrote:
It seems somewhat moot in these days of digital imaging, where the WB setting would have the most impact, either in camera .jpg or in post with RAW. I always adjust color temp. in PP if it helps the image to do it.
Still you can't get the same results with a bad lens as with a good lens, which is just as true with colors as with anything else.
My sigmas have always seemed a bit warmer than my G lenses. My AIS 105 f/2.5 has a "character" about it also, but I'm not sure if I can pin it on color.
Also, I think it may be rather hard to seperate "color" differences from other variables, such as contrast and micro-contrast, metering differences between lenses, etc. I think even resolution would have a role in fooling the human eye into seeing "better" color, etc.
It's an easy-enough test to pull, though.
I can compare my Sigma 50-500 to my AF-D 50mm f/1.8 and AIS 105 f/2.5, all at equivalent focal lengths and apertures and post them up, when I have time.
I've got a Sigma 28-200 D that always "seems" a bit cooler than my Nikon lenses - but it's in MY eye & MY mind. Nothing drastic
I also feel that some of the older Nikon MF glass have greater contrast and more saturation of the colours than some of the newer AF ones. But again. nothing major
And I agree with previous post about the 105mm f2.5MF - it really has it's own character. Just a joy to use!
I don't think it's going to be possible to display a color cast you might notice all the time but can't show on the web. There are too many other variables contributed by the processing of the image.
I think that only if you've used film and maybe if you shoot RAW will you notice a propensity of one lens to have a cast over another. Once you have it on the screen in Lightroom or Photoshop, the image is having presets applied to it. In fact, just the variables applied by various monitor settings would override any noticeable differences, I would imagine.
If you buy Diglloyd's ZF guide there are examples showing the color difference between ZF, Nikon, and Canon lenses in a very clear and understandable manor. He has them set up as mouse overs that flop between difference lenses.
From his review:
"Color rendition is notably richer and warmer with the Zeiss ZF lens line as compared with the Canon and Nikon lenses, a finding confirmed in a number of other field tests. For the image below, the late-day light was warm and nicely uniform, being reflected off a large grassy (yellowish) hillside, pleasing light which caught the eye of this photographer."
Back in the "'chrome" days of reversal film color balance was accomplished with film emulsion and filters. Lens coatings and aging affected colors to a lesser extent. But with color prints so difficult and expensive to print in home darkrooms many/most sent their work out to the tender mercies of some rather random color "management."
I remember I sent a treasured slide out for a "Dye Transfer" 8x10" print (the gold standard then) in the middle '70s. I paid, in today's dollars, $300-450 for ONE PRINT! It stood the test of time though and didn't fade. Today it's all different in the electronic darkroom. Amatures have tools and techniques that surpass the most seasoned, chemical, pros. With Photoshop etc you can adjust WB and trim colors to most anything you want.
the only color differences I notice with nikon is in their DC lenses (135,105) where the red channel is held back a tad to make skin look smooth and milky
domdog31 wrote:
the only color differences I notice with nikon is in their DC lenses (135,105) where the red channel is held back a tad to make skin look smooth and milky
That's very interesting! I've never actually heard of that before, it'd help a lot with the D300's sensor, which tends to 'bloom' in the red easily on skin.
"Lens colour" actually has very little to do with glass types or "tints". Spectral transmission differences caused by these reasons are always very "benign" in the way they work, if a glass has better UV than IR transmission the colours will not change/shift after a correct whitebalance has been applied. Well, actually - it will, ever so very slightly - since a skewing of the overall slope of the light power curve will very subtly "push" the hues placed at the x-over between the camera filter slopes (yellow and cyan) slightly towards short wavelengths, but this effect is so small that it is perfectly invisible unless you have two very large patches of the colours produced by one tilt next to one of the other.
What you're looking after is quite a lot more complicated than that.
Since a lens is quite complex and has quite a lot of glass-air-glass transitions, and every anti-reflex coating has different properties, you induce different "contamination" colours of veiling flare into the darker areas of the picture. Certain AR-coatings (they all do this) may disperse or chaotically refract some wavelengths more than others, and this veiling flare is the main part of what we perceive as "lens colour".
Say that a certain lens construction uses a combination of coatings that sums up to give dominantly "cold" (blue) veiling flare:
If you WB after the brighter portions of the picture, all shadows will take on a slightly "cold" hue.
If you WB after the "average" of the picture, brights will be slightly warm and the darks slightly cold.
Most of the time we do something in between those two. This of course alters the way we "perceive" a lens colour rendering, but the effect is very hard to quantify - since the lens reacts differently depending on if you're judging on the bright or the dark parts of the picture, and yet again differently if the bright areas are small and the darker areas large (or the other way around).
Most lenses we perceive as "warm" flare in the short end of the spectrum and make the shadows slightly colder after WB is set after "average" or "bright" pars of the picture.
Lenses we perceive as "cold" flare in the long end of the spectrum. This is actually quite contrary to what you'd initially believe, but the end effect of the capture and the subsequent WB results in this. We usually take more notice of the brighter portions of a picture.
Personally I prefer the Zeiss choice. Colour hue differentiation, or "visual colour resolution" is a lot easier with blue contamination than with red contamination.
As is quite easy to see, lens "colour" is not that easy to quantify - as it depends on how you judge the result, and attempting to do it by comparing large uniform areas of colour is also quite meaningless. As as with any multivariate subject, discussing it is bound to cause some differing opinions about the results... :-)
