Short tele comparison for portraits (skin colour rendering)

Joined
Mar 23, 2007
Messages
4,958
Location
Collecchio, northern Italy
Guys

so far we have showed pictures on a "lens brand" basis so that all the Nikkor 85 pictures are here, 105 DC and 135 there, 105 micro there.. the Sigma 85 are there, the Zeiss 85 and 100 are another place, so is the Tamron 90 and the Nikkor 105 F/2.5 - just to name a few (also Leica-R 80/90/100 for Nikon are admitted :wink: )

This time I'd like to propose a MIXED thread with a specific purpose in mind, especially for those who have (or have had) more than one of the aforementioned lenses. My aim is comparing not only the general iq (in the end we know them) rather the skin rendering: cooler - cool - neutral - warm - warmer. I noticed that the (old) Nikkor 85 was on the coolish side (which I didn't like at all) while other lenses are warmer and - to me - more pleasant to look at straight out of camera. As soon as I can, I'll post a few samples, but feel free to start yourself in the meantime.
 
Last edited by a moderator:
Joined
Mar 23, 2007
Messages
4,958
Location
Collecchio, northern Italy
Ok, here we go - I'll try to avoid flash pictures, since ambient light reveals more the "nature" of a lens and its "cast", imho. Let's forget sharpness and other parameters for a moment.

105 F/2.5 AIS ( "warm" to my eyes ):

981105807_E3JN3-XL.jpg
Subscribe to see EXIF info for this image (if available)


Tamron 90 ( neutral to warm )

620550520_Ktht5-XL.jpg
Subscribe to see EXIF info for this image (if available)


621599596_pJeno-XL.jpg
Subscribe to see EXIF info for this image (if available)


866853445_jKG2A-XL.jpg
Subscribe to see EXIF info for this image (if available)


105 F/2 DC ( neutral-to-a-tad-cool )

1037690078_RxiH2-XL.jpg
Subscribe to see EXIF info for this image (if available)


142123021_9vtAx-XL.jpg
Subscribe to see EXIF info for this image (if available)


161097021_JtHnP-XL.jpg
Subscribe to see EXIF info for this image (if available)


85 F/1.4 (cool )
171725101_fhbMc-XL.jpg
Subscribe to see EXIF info for this image (if available)


171843731_veC4t-XL.jpg
Subscribe to see EXIF info for this image (if available)


171843746_S2aBf-XL.jpg
Subscribe to see EXIF info for this image (if available)


768251567_PqeaM-XL.jpg
Subscribe to see EXIF info for this image (if available)


Others to follow later
 
Joined
Nov 6, 2008
Messages
2,447
Location
South Orange, NJ
Isn't the warmth based on the white balance?

I think to truly do this as an unbiased comparison, someone that owns all of the lenses should take pictures of the same subject, same background, same light, same exposure settings.
 
Joined
May 20, 2008
Messages
2,505
Location
Fairfax Station, VA
Real Name
Tony Admana
Isn't the warmth based on the white balance?

I think to truly do this as an unbiased comparison, someone that owns all of the lenses should take pictures of the same subject, same background, same light, same exposure settings.

+1 I agree
 

Rob Zijlstra

A Koffie Drinker
Joined
Nov 5, 2008
Messages
999
Location
Netherlands
Dino, I see a tremendous amount of difference regarding the skin color in the set you posted. But, if you shoot raw and save some quick settings for every lens regarding the adjustments you have to make to get a a 'nice'(whatever that is) skin-color, it takes about 20 seconds to get the photo the way you like. And by adjustments I mean of course the WB, but also some other settings: color, sharpness, color-detail etc etc.
Or am i mistaken?
 
Joined
Nov 6, 2008
Messages
2,447
Location
South Orange, NJ
Dino, I see a tremendous amount of difference regarding the skin color in the set you posted. But, if you shoot raw and save some quick settings for every lens regarding the adjustments you have to make to get a a 'nice'(whatever that is) skin-color, it takes about 20 seconds to get the photo the way you like. And by adjustments I mean of course the WB, but also some other settings: color, sharpness, color-detail etc etc.
Or am i mistaken?

Yes, in RAW processing, the WB can be adjusted to be as cool or as warm as you want it. I am not sure if lenses themselves determine the warmth of a picture because it is clear glass. Perhaps a wider lens allows more light in, thus if its a warm light, warmer light enters but then again the camera adjusts for this by modifying WB.
 

Rob Zijlstra

A Koffie Drinker
Joined
Nov 5, 2008
Messages
999
Location
Netherlands
Milan,
Probably that phrase 'Or am I mistaken' didn't make clear what I meant.
I meant if Dino has another way of taking pictures and pp'íng them. I know a lot of folks here won't use raw, for whatever reason. If Dine takes his photographs in JPEG I can imagine that his original question is much more relevant.
If you're using a studio with everything controlled, I imagine that after some time you have an excellent saved setting in raw to get perfect pictures, without even using the photoshop's controls themselves.
 
Joined
Mar 18, 2008
Messages
5,113
Location
Miami, Florida, USA.
Isn't the warmth based on the white balance?

I think to truly do this as an unbiased comparison, someone that owns all of the lenses should take pictures of the same subject, same background, same light, same exposure settings.

+ 1. What seems as a cool rendition of the lens is indeed a wrong WB.
Nikon optics have been pretty neutral over the years with lots of similarities in the rendition of images among lenses.
Skin colors can be manipulated at will with software and my favorite is Nikon Capture NX. I find that, in general, Nikon cameras tend to clip the red channel
and my D70S is very prone to do that. Correction is very simple.
Sorry if my flash picture of the grandson did not meet your approval for the thread. It has been removed.

William.
 
Last edited by a moderator:
Joined
Jun 29, 2009
Messages
1,485
Location
Florida
I think you're trying to evaluate color shifts and if a lens favors a particular channel (R,G,B) . lenses will have color shifts under different lighting but do tend to favor (saturate) one particular channel.

Usually modern lenses favor the red channel, therefore your "warm" colors. The Two lenses that come to mind with high color shift to the red are the 17-55mm DX and 24-70mm F/2.8. What I've noticed is the lenses with ED will favor the Red channel (and to lesser degree the green channel) and have added contrast, which causes that "pop" look, somethimes it can be overdone, the cheaper ED glass used in consumer zooms and the 35mm F/1.8 dx makes the contrast too harsh for portraits, but it works great for landscapes.

The ones I've found with minimal or neutral shift under various lighting conditions are the DC's, the 35mm F/2 AF-D, the old 35-70mm F/2.8

The ones I've used which are cooler are the 85 F/1.4 AF , the old 35-135mm F/3.5 - 4.5 AF
 
Joined
Mar 23, 2007
Messages
4,958
Location
Collecchio, northern Italy
Let me reply
of course I know the WB thing, however they were both shots in Auto WB mode so either the D200 was always "wrong" (at that time I couldnt process RAW files, though, only later) or the 85 was effectively cooler
 
Joined
Nov 10, 2008
Messages
3,969
Location
Chicago
The only way this experiment has validity is if all the pictures were taken with one camera, identical lighting conditions, and the lenses were simply interchanged. Then they have to be processed on a single computer or at least a calibrated computer.

There are way too many variables here to make accurate assessments.

Post processing will get any color shifts to match.
 
Joined
Apr 25, 2009
Messages
1,304
Location
Kitchener, ON
Let me reply
of course I know the WB thing, however they were both shots in Auto WB mode so either the D200 was always "wrong" (at that time I couldnt process RAW files, though, only later) or the 85 was effectively cooler

The camera sensor has to be trying to adjust for any differences between the lenses as the light it uses to evaluate white balance goes through that lens first. The camera has no way of knowing and doesn't care if a colour shift is caused by the lighting or by the lens.

If you are really concerned with this issue, you have to lock the camera down to a fixed value and choose consistent lighting. Then do a/b testing between two lenses to see what difference there is.
 
Joined
Jun 29, 2009
Messages
1,485
Location
Florida
Post processing will get any color shifts to match.

Not really, Bayer filters only record one channel (R,G,B) per pixel and sample their neighboring pixels (interpolate) to produce a combined color output per pixel. For example a model with brown eyes, if you were to pixel peep to the individual brown pixels, each single pixel recorded either a single R,G,or B channel and borrowed it's neighbor channels to create a single brown pixel of information.

If you want to adjust for a Red channel shift, it will inevitably have an effect on the interpolating formula of the neighboring channels. This is a limitation of the Bayer filter and the reason why color accuracy from the lens is crucial.

Post processing to match colors would be more effective if D-SLR's utilized a 3CCD or 3CMOS set up like the ones used in profesional video recording. Each CCD and each pixel per CCD is dedicated to each channel, the article below is a good read on the subject.

http://en.wikipedia.org/wiki/3CCD
 

Latest posts

Links on this page may be to our affiliates. Sales through affiliate links may benefit this site.
Nikon Cafe is a fan site and not associated with Nikon Corporation.
Forum post reactions by Twemoji: https://github.com/twitter/twemoji
Forum GIFs powered by GIPHY: https://giphy.com/
Copyright © Amin Forums, LLC
Top Bottom