Blown reds & Adobe RGB

[echelon2004]

color spaces should only affect out of gamut colors.

Looking at the faces in the image, it is clear that even colors not out of gamut is taking a toll=)

it WOULD help to examine the problem had the images posted carried a color space tag...
--
Anders

Some of my pictures can be seen at;
http://teamexcalibur.se/US/usindex.html

event photography and photo journalism

 

[kevm14]

it WOULD help to examine the problem had the images posted carried
a color space tag...

I labeled them with text above each image. What do you want me to do exactly?

 

[hotwag]

Your White Balance appears to be set to "flash." Is that correct? You need "Dir Sunlight" in those conditions.
--
GHW

 

[echelon2004]

use convert to color space and make sure that the colorspace shows in the save dialog.

That way color space aware software (such as all proper image editors) will map the colors correctly relative to which workspace is selected.

Never just assign a color space.
--
Anders

Some of my pictures can be seen at;
http://teamexcalibur.se/US/usindex.html

event photography and photo journalism

 

[echelon2004]

is any of those images actually Adobe RGB when posted here?

If so, no internet browser used on a windows computer will show the images anywhere close to what they look like...

And since the images are uncalibrated for color space, no image editing software will know how to map colours either
--
Anders

Some of my pictures can be seen at;
http://teamexcalibur.se/US/usindex.html

event photography and photo journalism

 

[kevm14]

Your White Balance appears to be set to "flash." Is that correct?
You need "Dir Sunlight" in those conditions.

Yes my dad was using my camera that day and to simplify matters, I just set the WB to flash and left it. Flash WB is actually pretty accurate for direct sunlight pictures, if a tad warm. Nothing wrong with a little warmth. These were all RAW so it doesn't matter anyway.

 

[kevm14]

use convert to color space and make sure that the colorspace shows
in the save dialog.

That way color space aware software (such as all proper image
editors) will map the colors correctly relative to which workspace
is selected.

Never just assign a color space.

If I open the NEF in NC, and switch the color mode to II, then hit the send to photoshop button, am I OK so far? Then I can convert to sRGB color space using some built-in feature? Is this correct? Because all I did was send to CS2 then save as a JPEG. I thought it converted to sRGB at that point.

 

[kevm14]

is any of those images actually Adobe RGB when posted here?

I thought they were sRGB. I will try to follow your instructions so we know for sure.

If so, no internet browser used on a windows computer will show the
images anywhere close to what they look like...

This brings me to an interesting point. I understand the concept of gamut limitations. But then how can we look at a gamut illustration of Abobe RGB or ProPhoto RGB right here on the web in sRGB?? That makes no sense. Or is it just a visual aid to show the gamut is wider, even though the actual colors are obviously not correct?

 

[echelon2004]

If I open the NEF in NC, and switch the color mode to II, then hit
the send to photoshop button, am I OK so far? Then I can convert
to sRGB color space using some built-in feature? Is this correct?
Because all I did was send to CS2 then save as a JPEG. I thought
it converted to sRGB at that point.

I can't help you with Nikon Capture, but let us assume that the color space information follows the file from NC.

Do one conversion in NC using sRGB and another using Adobe RGB.
Send both to photoshop.

Convert both images to sRGB (depending on your version of photoshop it might be found in different places, but look at edit/adjustment/convert or just under the edit menu).

Then post the images again after saving as jpg, not save for web.

If NC doesn't render sRGB or Adobe RGB properly or the colors are out of gamut in sRGB, we should see sthe same difference as earlier.

--
Anders

Some of my pictures can be seen at;
http://teamexcalibur.se/US/usindex.html

event photography and photo journalism

 

[echelon2004]

the nice 3D maps of color spaces is just a visual guide, the colors shown are not accurate. Often the size and placement are quite accurate though.
--
Anders

Some of my pictures can be seen at;
http://teamexcalibur.se/US/usindex.html

event photography and photo journalism

 

[THKTHK]

If I open the NEF in NC, and switch the color mode to II, then hit
the send to photoshop button, am I OK so far? Then I can convert
to sRGB color space using some built-in feature? Is this correct?
Because all I did was send to CS2 then save as a JPEG. I thought
it converted to sRGB at that point.

Yes, this is exactly the same what I usually do with this kind of pictures.

Because my PS is set up to use sRGB, it alerts that the received picture from NC is not in that colour space, and asks what should it do. I hit "convert to working colour space", and that's it, it will be in sRGB - without the clipped, orange-like reds

--
Henrik

 

[segarci1]

My observations on blown reds vs. color spaces vs. conversion options are identical to yours... but let me add one more wrinkle:

If I save a NEF in aRGB and then use Nikon View to convert to sRGB JPG, the results are ok... they pretty much match my color-managed aRGB preview.

So... all because of blown reds... I stay in aRGB all throughout my flow except for the last step of conversion to JPG, for which I use Nikon View.

--
sg

 

[kitgard]

a good article by oldskool how to convert from aRGB-> sRGB w/ Nikon View

http://www.digitalkb.com/nikon/d70/nikon_d70_articles/work_with_argb/

rgds
---
kitgard

 

[echelon2004]

The blooming, which might be a contributing factor to the problem according to Nikon Professional in Sweden, is caused by the electronic shutter technology.

From other posts in this thread it has become rather clear that other factors, partly involving Nikon Capture! might contribute or enhance the problem.

--
Anders

Some of my pictures can be seen at;
http://teamexcalibur.se/US/usindex.html

event photography and photo journalism

 

[RomanJohnston]

assigning the color space and CONVERTING the color space.

Wonder if Adobe being the gurus they are......programed PS to convert.....so it takes the full color space in consideration and renders it properly in the resulting compression.

Roman
--

Photoshop is like bondo. Use it to smooth out and polish your work, not to compensate for your lack of body work skills.

http://www.pbase.com/romansphotos/

 

[THKTHK]

Thanks - it works!

Now this is even more weird: Nikon View can convert to sRGB correctly, Nikon Capture clips reds and shifts them towards orange - just like the camera in sRGB

If NV can do it correctly, NC why not?

--
Henrik

 

[RomanJohnston]

The modle shows exactly what my testing has shown....

the cutoff is in the reds and blues/purples... between the D70 and sRGB...less so in aRGB....and none in ProPhotoRGB.

How much testing and examination of the actual models have you done?

Your answer leads me to believe none...te data is what the data is..

Roman
--

Photoshop is like bondo. Use it to smooth out and polish your work, not to compensate for your lack of body work skills.

http://www.pbase.com/romansphotos/

 

[echelon2004]

The modle shows exactly what my testing has shown....

As I told you, I can't see their models. If they are mapped the most common way you'll suffer most in the blue-green area.

the cutoff is in the reds and blues/purples... between the D70 and
sRGB...less so in aRGB....and none in ProPhotoRGB.

Agreed that ProPhotoRGB is a much larger gamut obviously, while retaining the fine steps of sRGB all the way. As you probably know Adobe RGB have the same number of colors spread over a larger area, resulting in larger spaces between possible colors.

As I always shoot raw I have not had the pleasure to examine the D70 native color spaces, to me they are rather pointless. The raw data isn't though, and the early clipping of red happens on sensor level.

How much testing and examination of the actual models have you done?

Your answer leads me to believe none...te data is what the data is..

No sense in testing any of those color spaces, far greater minds have done that before=)

As you so wisely puts it, the data is what the data is.
And the "problem" is within the data.

The examples shown here seems to be strongly exagerated by poor color management though.

Are you using NC much?

Apart from disagreeing with you know, I have taken great pleasure in your work and have not seen any evident and big mistakes in the processing.

--
Anders

Some of my pictures can be seen at;
http://teamexcalibur.se/US/usindex.html

event photography and photo journalism

 

[Louie M]

kevm14 wrote:

IIIa crop:

Fascinating...

I don't t know if this helps, but eyedropper shows no green at all in the shirt -- just red and blue. The Mode II shows green.

 

[ccarrier]

If you change the color mode from II to Ia or IIIa in NC it will have the same effect as if you shot in that mode.

BUT :

In your color managment option, set the default RGB color space to NKsRGB.icm

and then check the "use this instead of an embedded profile when opening files".

You can still chose color mode II but when you save the file in JPEG, il will be saved in sRGB but with the color signature of mode II. Meaning, no blown red.

--
Claude Carrier