Pages: 1 [2] 3 4 ... 9   Go Down

Author Topic: Colour of light matters  (Read 38237 times)

32BT

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3095
    • Pictures
Re: Colour of light matters
« Reply #20 on: July 09, 2013, 01:51:26 pm »

The whole reason that Thomas Knoll designed Camera Raw to have two separate DNG profiles; one for Standard Illuminate A (2856ºK) and a different profile for D65 was to account for the differences in the sensor's response under those two SPDs. Note that this isn't about correcting for white balance...DNG profiles are designed to correct for different spectral responses AFTER white balance has been corrected.

The problem of course is that Camera Raw manages to obfuscate the two in a horribly incorrect way: by coupling the response profiles to the temperature slider it is effectively making it "about white balance". And then it does so incorrectly because the response is a function of SPD as you mention, but the SPD is not a function of colortemperature. (At least not the SPDs as used in DNG).

Think of it in another way: there are a lot more specific SPDs then the D or A varieties, which get no Camera Raw or DNG love. It's a convenience by oversimplification, which is fine in the RAW converter, but should not be a part of the DNG standard.

/rant over

(sorry, couldn't help myself.)

Logged
Regards,
~ O ~
If you can stomach it: pictures

digitaldog

  • Sr. Member
  • ****
  • Online Online
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: Colour of light matters
« Reply #21 on: July 09, 2013, 01:56:56 pm »

The problem of course is that Camera Raw manages to obfuscate the two in a horribly incorrect way: by coupling the response profiles to the temperature slider it is effectively making it "about white balance". And then it does so incorrectly because the response is a function of SPD as you mention, but the SPD is not a function of colortemperature. (At least not the SPDs as used in DNG).

As opposed to an ICC profile which has a single assumption for an illuminant (usually D50) and usually is based on output referred data.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

digitaldog

  • Sr. Member
  • ****
  • Online Online
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: Colour of light matters
« Reply #22 on: July 09, 2013, 02:14:12 pm »

Think of it in another way: there are a lot more specific SPDs then the D or A varieties, which get no Camera Raw or DNG love.

Yes (with a canned DNG profile) and Eric explained this quite well way back in 2010:
Quote
http://forums.adobe.com/message/3183090#3183090


Hi Andrew,
 
The processing order is roughly as follows. DNG color profiles contain at least a color matrix, and (optionally) one or more lookup tables, and (optionally) a tone curve. So at the minimum, there is a color matrix applied during image processing. This is applied very early in the imaging pipeline, while the image data is known to be in a linear, scene-referred space. The optional lookup tables and tone curve follow, and can be used to correct problem colors and residual issues, and apply a desired scene-to-output mapping.
 
Regarding white balance: It is applied along with the color matrix, i.e., very early. You can think of it as scaling the camera-native RGB coordinates till the desired values are neutral (i.e., R = G = B). This scaling, along with the color matrix in the profile, are effectively responsible for transforming from the camera-native RGB color coordinates -- which vary from model to model -- to a colorimetrically-defined system: CIE XYZ with a D50 white point, in the case of the DNG and ICC models. The user's chosen white balance is applied during this matrix step.
 
In the case of a dual-illuminant DNG profile, the user's chosen WB also has a second role, which is to determine how the matrices are used. As you know, Adobe's supplied "Adobe Standard" profiles are built using standard illuminants A and D65 as the calibration illuminants. So they contain separate matrices for each illuminant. The user's chosen WB determines which of the two matrices is used (more generally, how the two matrices are blended together) at image processing time. For example, if your chosen WB is Temp = 7500 K and you're using an Adobe Standard profile, then the DNG profile model calls for using the D65 color matrix. On the other hand, if your chosen WB for an image is Temp = 5200 K, then you'll be using a blend of the A and D65 matrices, with the latter more heavily weighted (since it's closer to D65 than to A). The DNG spec goes into more details on how this is done.
 
So to answer the question of WB and it's relevance to building & using DNG profiles: in principle, the applicability of a profile with a given WB has to do with how close the spectrum of the scene illuminant (used to photograph your real image) is to the spectrum of the illuminant used to build the profile. The closer they are, the better the results. Many flavors of natural daylight are spectrally similar (weighted differently), so this is why a daylight profile tends to work well in many flavors of daylight regardless of the actual CCT (e.g., 4700 K thru 7500 K). So even if you built your own profile under real daylight that ended up being around 6200 K, you should not hesitate to use such a profile under other similar daylight conditions, even if the CCT values vary a lot. But if you end up using a very different type of light (e.g., a fluorescent tube) then you should effect quite different, possibly unpleasant results, even if the CCT measures the same (e.g., a 6000 K office fluorescent tube).
 
Regarding the user's question of RGB & HSV, yes it is the same as the original definition proposed by Alvy Ray Smith. As Sandy notes there is an implementation in the DNG SDK for interested developers to use.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

32BT

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3095
    • Pictures
Re: Colour of light matters
« Reply #23 on: July 09, 2013, 03:00:39 pm »

What they should have done (in fact they still can) they should have build a menu with response options (like "Tungsten, Daylight, Flash, etc...") and include one option called "Auto" which would then use the temperature slider to blend between whatever they think can be blended.

Personally I think response matrices are discrete entities, and shouldn't be blended, but if convenience is your thing and as long as it is optional...
Logged
Regards,
~ O ~
If you can stomach it: pictures

digitaldog

  • Sr. Member
  • ****
  • Online Online
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: Colour of light matters
« Reply #24 on: July 09, 2013, 03:34:21 pm »

What they should have done (in fact they still can) they should have build a menu with response options (like "Tungsten, Daylight, Flash, etc...") and include one option called "Auto" which would then use the temperature slider to blend between whatever they think can be blended.

Like this?

Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

32BT

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3095
    • Pictures
Re: Colour of light matters
« Reply #25 on: July 09, 2013, 03:44:58 pm »

Yes, but for response, not white balance. (You see how it is obfuscated?)

This menu most likely gives you some combination of temperature and tint only, and then the underlying engine selects the blend of "Tungsten" and "Daylight" based on temperature.

So, technically you should only get "Tungsten", "Daylight", and "Auto" in a menu (for the average DNG profiles).







Logged
Regards,
~ O ~
If you can stomach it: pictures

digitaldog

  • Sr. Member
  • ****
  • Online Online
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: Colour of light matters
« Reply #26 on: July 09, 2013, 03:48:58 pm »

Yes, but for response, not white balance. (You see how it is obfuscated?)

Not really. Maybe if I needed to understand everything happening under the hood. But I don't see the present controls having any negative effect on getting the color as I desire. Even when using a DNG profile built for daylight, selecting that or even using the WB tool, I often edit Tint/Temp to produce a rendering I prefer.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

32BT

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3095
    • Pictures
Re: Colour of light matters
« Reply #27 on: July 09, 2013, 03:57:28 pm »

Not really. Maybe if I needed to understand everything happening under the hood. But I don't see the present controls having any negative effect on getting the color as I desire. Even when using a DNG profile built for daylight, selecting that or even using the WB tool, I often edit Tint/Temp to produce a rendering I prefer.

Well, to refer to Eric's own example: if you select the Fluorescent option it doesn't actually give you a Fluorescent response. That would be a potential pitfall. An entirely other problem of course is the whole idea behind Temperature and Tint, which are a mathematical solution to a 2 dimensional problem, but have very little relation to the photographer's concept of tint, which is more akin to colorcast.
Logged
Regards,
~ O ~
If you can stomach it: pictures

digitaldog

  • Sr. Member
  • ****
  • Online Online
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: Colour of light matters
« Reply #28 on: July 09, 2013, 04:05:26 pm »

Well, to refer to Eric's own example: if you select the Fluorescent option it doesn't actually give you a Fluorescent response. That would be a potential pitfall. An entirely other problem of course is the whole idea behind Temperature and Tint, which are a mathematical solution to a 2 dimensional problem, but have very little relation to the photographer's concept of tint, which is more akin to colorcast.

First off, IF I had a daylight captured image and selected Fluorescent, I'd get a less than acceptable image (IMHO) so the take home is, don't select it! That said, the setting could provide a rendering someone might like. On a few daylight captured images, it doesn't appear as I'd expect from a daylight to Fluorescent "conversion" if you will, mostly a rather cool rendering.

2nd, what you call a problem isn't one IMHO but you are welcome to call it a problem. You may have a beef with the names, fine. Once one explains what the sliders are doing, or even if one ignores the explanation of what's happening under the hood, one just moves them to get a desired appearance, all is fine (no problem). Now if you have specific images whereby the sliders fail to provide an acceptable rendering, we have an actual problem to discuss.

I can't speak for all photographers as to what they conceive for just about anything! <G>
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

Tim Lookingbill

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2436
Re: Colour of light matters
« Reply #29 on: July 09, 2013, 04:14:29 pm »

Not really. Maybe if I needed to understand everything happening under the hood. But I don't see the present controls having any negative effect on getting the color as I desire. Even when using a DNG profile built for daylight, selecting that or even using the WB tool, I often edit Tint/Temp to produce a rendering I prefer.

And the purity of these Tint/Temp effect perception of the overall image that has nothing to do with the effects provided by the matrices.

I was surprised to find how other apps that can edit Raw can inject their own hue/sat of Tint/Temp that differs greatly from the previews in ACR/LR combined with the actual scene which adds its own patina.

Some Raw converters (iPhoto & SilkyPix my Pentax's Raw converter) think golden hour sun has the color of dingy yellow straw while ACR/LR maintains the required level of non-linear/non-uniform "selective" saturation increase to highlight/lower mids with the right=(more realistic) blue/amber hue to mimic the effect of how the sun low in the sky actually looks shining on real objects.

Other Raw converters add somewhat of a distortion to the hues and squelch this "selective" luminance/saturation increase to highlights to keep the overall image uniform and linear behaving so when actual saturation is applied all objects whether lit by the sun or in shadow increase equally in saturation giving an overall dingy patina and less realistic appearance.
« Last Edit: July 09, 2013, 04:16:16 pm by Tim Lookingbill »
Logged

stamper

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 5882
Re: Colour of light matters
« Reply #30 on: July 10, 2013, 04:48:59 am »

Hi,

Well, that depends if e.g. you are into accurate reproduction of artwork ... The mix of Red and Green pigments may produce the same yellow as a more pure yellow pigment, depending on the spectrum of the illuminant, and the sensitivity of the Bayer CFA filtered sensels.

In other circumstances, e.g. photo journalism, it may be less of a show stopper. It only becomes a hinderance if it's important and cannot be rectified by accurate profiling. After all, this is a forum about Color Management, so feel free to ignore if you do not care about that.

Cheers,
Bart

Colour management isn't an end in itself, or at least it shouldn't be. Unfortunately a lot of the posts on this forum seem to lead you to think so? At the end of the day it has to/should relate to the final output. That was my point.

Tim Lookingbill

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2436
Re: Colour of light matters
« Reply #31 on: July 10, 2013, 01:04:07 pm »

Colour management isn't an end in itself, or at least it shouldn't be. Unfortunately a lot of the posts on this forum seem to lead you to think so? At the end of the day it has to/should relate to the final output. That was my point.

If you care about the subtleties and overall look of the image (patina) derived from edits and you want that level of image quality preserved for all to see and appreciate, then color management is the best thing you've got to assure this.

If that's not of major concern to you then you might as well just shoot jpegs and present them as they are straight out of the camera. No one will really know the difference.

In fact I was at my local art galleries photography section where there were quite a few prints created that way. Of course I could see from the perceptually non-uniform levels of saturation between background and foreground objects (in this case a field of flowers) subtlety, nuance and overall balance of color was not of concern and I passed it by because I couldn't stand looking at it for more than a couple of seconds.
« Last Edit: July 10, 2013, 01:07:26 pm by Tim Lookingbill »
Logged

Iliah

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 770
Re: Colour of light matters
« Reply #32 on: July 10, 2013, 04:59:17 pm »

> Not really. ... But I don't see the present controls having any negative effect on getting the color as I desire.

The first sentence and the second do not belong together.

Really, for different light spectrum one needs a different colour transform, not different chromatic adaptation. Tint and selecting colour temperature do not solve it.

As for the second sentence, you have no option of the correct workflow, so you can't compare.
Logged

digitaldog

  • Sr. Member
  • ****
  • Online Online
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: Colour of light matters
« Reply #33 on: July 10, 2013, 05:22:42 pm »

The first sentence and the second do not belong together.
Really, for different light spectrum one needs a different colour transform, not different chromatic adaptation. Tint and selecting colour temperature do not solve it.
As for the second sentence, you have no option of the correct workflow, so you can't compare.

Oh, there really IS a problem, I and the others using this product just don't see it? And you're quite certain this is the only product I've ever used? Thanks for setting me (us) straight and uncovering this a, problem we don't recognize.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

Iliah

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 770
Re: Colour of light matters
« Reply #34 on: July 10, 2013, 05:28:01 pm »

> there really IS a problem, I and the others using this product just don't see it?

Yes, there is a problem, and you and some other users may not see it.

Other products except Nikon Capture do not automatically take into account spectral properties of the light, you need to create necessary profiles yourself.
Logged

digitaldog

  • Sr. Member
  • ****
  • Online Online
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: Colour of light matters
« Reply #35 on: July 10, 2013, 05:39:16 pm »

Yes, there is a problem, and you and some other users may not see it.

We're all color blind? We don't see it, that's true. How that makes it a problem only you can say for certain if we are to accept your premise. You'll forgive me if I don't jump off that bridge (blindly)...
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

Iliah

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 770
Re: Colour of light matters
« Reply #36 on: July 11, 2013, 02:17:18 am »

We're all color blind? We don't see it, that's true. How that makes it a problem only you can say for certain if we are to accept your premise. You'll forgive me if I don't jump off that bridge (blindly)...

Your desperate attempts of sarcasm, Andrew, are funny. But you know well from your own experience - they do not work.
Logged

stamper

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 5882
Re: Colour of light matters
« Reply #37 on: July 11, 2013, 03:49:08 am »

If you care about the subtleties and overall look of the image (patina) derived from edits and you want that level of image quality preserved for all to see and appreciate, then color management is the best thing you've got to assure this.

If that's not of major concern to you then you might as well just shoot jpegs and present them as they are straight out of the camera. No one will really know the difference.

Unquote

If nobody has seen the initial file then they won't know what subtleties were there? They won't - or nor will they care - what hoops you jumped through to get there. I only shoot raw - not jpegs - and a good workflow is needed. However it seems to me - and others - that some get caught up in the workflow process and think it is an end to itself. I bet you have seen - and liked - some prints that have started life as a jpeg. As you stated...No one will really know the difference. :)

digitaldog

  • Sr. Member
  • ****
  • Online Online
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: Colour of light matters
« Reply #38 on: July 11, 2013, 09:17:41 am »

Your desperate attempts of sarcasm, Andrew, are funny. But you know well from your own experience - they do not work.

Now you're just being ridiculous! You make a point that something is broken, that thousands if not tens of thousands of users of ACR/LR can't see and don't recognize? IF you have a raw file the rest of us can examine whereby you are unable to produce a rendering and can explain this, we're all ears. Otherwise you are taking us down a faith based rabbit hole. As I said and will say again with a library of 30K+ raws all handled within the ACR engine: There's no problem on this end, I have no issues, no problems producing a rendering I desire.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

Iliah

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 770
Re: Colour of light matters
« Reply #39 on: July 11, 2013, 09:28:29 am »

> Now you're just being ridiculous!

You just added yet another reason to trust your perception, dear Andrew ;)
Logged
Pages: 1 [2] 3 4 ... 9   Go Up