Do i have a problem with my D500 ?

mikew_RIP

Senior Member
Having a bit of a problem with back focus on my 100-400 so have been taking pictures just to check things. This is two screen grabs so the basic details are showing, as you see they where taken one after the other at the same settings, the difference in colour surprises me.

Capture.JPG


Capture2.JPG
 

Horoscope Fish

Senior Member
When I compare the histograms in ACR there's a very slight, but undeniable shift in exposure toward the left; primarily in the shadows and even more specifically with the reds, blues and magentas. I'm not sure why unless it's the internal processing of the camera making a minor adjustment based on your settings.
 

mikew_RIP

Senior Member
When I compare the histograms in ACR there's a very slight, but undeniable shift in exposure toward the left; primarily in the shadows and even more specifically with the reds, blues and magentas. I'm not sure why unless it's the internal processing of the camera making a minor adjustment based on your settings.

Thanks its the first time i have noticed it but with having problems with the lens you can guess what i said to myself, :D i think it was flippin heck thats all i need.
 

mikew_RIP

Senior Member
I find it very unlikely for the lens to be the issue in this case.

I agree thats why i thought oh no not a camera problem as well,its been one of those days,we have wi-fi problems,my wives laptop dipped out and now will find all the wi-fi around us but not ours,my desk top is running on that wi-fi her lap top wont find now,
I have plugged an ethernet in and told her thats it for today :D
 

lokatz

Senior Member
Sorry to hear you're having a bad day. About the D500, however, I'd say don't panic. My money would still be on either an AutoWB shift or something like that. You camera must have perceived a difference between the two shots - that's not unusual even for a body as excellent as the D500.
 

lokatz

Senior Member
Per the data you are showing, it was actually 1/5th of a second. Not enough for the sun to change, but easily enough for the body to decide that a different WB setting is on order. Unfortunately, even the detailed EXIF data won't give a clue here - all Nikon is willing to tell us is 'AutoWB', so there is no way I know of to learn which WB setting the camera picked.

Anyone know how to find out?
 

Bikerbrent

Senior Member
At least on my D7200, if I open a RAW photo in Photoshop CS6, it will tell me auto white balance and then display the actual temperature.
 

lokatz

Senior Member
At least on my D7200, if I open a RAW photo in Photoshop CS6, it will tell me auto white balance and then display the actual temperature.

Interesting - I've been looking through the data with ExifTool and couldn't find the color temp. Where do you see it in PS?
 

lokatz

Senior Member
Thanks. I wonder how PS would know this - or if it just makes some kind of assumption. Need to test this when I find the time...
 

Horoscope Fish

Senior Member
It would have needed to be a quick change at 1/20th of a second between shots.
I think the issue lies with AWB. Auto White Balance, as I understand it, works by way of sophisticated algorithms which look at the entire scene in the frame. It has to make some basic assumptions and the first one is that the scene should be color-neutral. I suspect AWB may tap into the camera's internal database of stored images as well, too help determine what sort of scene it's looking at, much like it does to adjust exposure when using the Matrix metering mode; but that's just my hypothesis.

Anyway... By computing the average color in the scene, and applying the same correction to every pixel, the result should be a frame wherein the average color is neutral; which should mean correct white-balance. And it normally does. AWB is not fool-proof however and certain circumstances can fool it into making bad choices. For instance, when there is a single color dominating the frame, AWB "sees" too much of that particular color, mistakes it for a color-cast, and then tries to correct for it, which of course affects the entire shot. I have a feeling this might be at the heart of the issue here.
 
Last edited:

WayneF

Senior Member
Thanks. I wonder how PS would know this - or if it just makes some kind of assumption. Need to test this when I find the time...

Cameras do not put color Temperature and Tint into the Exif. At best, they might show your WB menu setting, such as Incandescent, which is just a setting, but is probably Not what the actual light was. Direct Bright Sun is probably an exception, and possibly is more of a constant, but otherwise, there are many variations of color for Incandescent, Florescent, Cloudy, Shade, Flash, whatever. Nobody measured the actual light. The settings are pretty much just guesses based on norms.

The camera puts a RGB multiplier into the Exif, for example:

WB RB Levels : 1.8984375 1.44140625 1 1

This is channel multipliers for Red, Blue, Green and Green (for the raw Bayer pattern).

The camera has already done this multiplication for JPG data (the data has been changed), but raw files are not processed yet.

Adobe reads that Exif and tries to figure it out. The default colors come out the same with this same multiplier, but the degrees K Temperature and Tint numbers are approximations. And Not meaning comes out correct, but comes out the same. No one knows what the correct WB ought to be for the exact light at the scene. That's why we use raw, to be able to correct it easily.
 

Woodyg3

Senior Member
Contributor
Cameras do not put color Temperature and Tint into the Exif. At best, they might show your WB menu setting, such as Incandescent, which is just a setting, but is probably Not what the actual light was. Direct Bright Sun is probably an exception, and possibly is more of a constant, but otherwise, there are many variations of color for Incandescent, Florescent, Cloudy, Shade, Flash, whatever. Nobody measured the actual light. The settings are pretty much just guesses based on norms.

The camera puts a RGB multiplier into the Exif, for example:

WB RB Levels : 1.8984375 1.44140625 1 1

This is channel multipliers for Red, Blue, Green and Green (for the raw Bayer pattern).

The camera has already done this multiplication for JPG data (the data has been changed), but raw files are not processed yet.

Adobe reads that Exif and tries to figure it out. The default colors come out the same with this same multiplier, but the degrees K Temperature and Tint numbers are approximations. And Not meaning comes out correct, but comes out the same. No one knows what the correct WB ought to be for the exact light at the scene. That's why we use raw, to be able to correct it easily.

Interesting. Thanks for that info. I always assumed it was a value recorded in the EXIF data.
 

lokatz

Senior Member
I agree thats why i thought oh no not a camera problem as well

Mike, per WayneF's explanation (I just analyzed a few test shots and found the same), as well as the other explanations provided, you can either simply open the two images in Adobe Bridge/LR/PS and compare the color temperatures, or you can use something like ExifTool and compare the WB_RBLevels (look for them in the Nikon section of the data). If these values are different between the two pictures, which I think they will be, you know for sure that the AutoWB in your camera simply made slightly different decisions between them.
 
Top