Pages: [1] 2   Go Down

Author Topic: Color Editor Tool RGB readings anomaly / Windows 10 /  (Read 14815 times)

Dinarius

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1225
Color Editor Tool RGB readings anomaly / Windows 10 /
« on: December 14, 2015, 12:07:47 pm »

I am using C1 9.0 with Windows 10.

This problem existed with C1 8 and Windows 8 and 10, and it's now worse, for me.

But, I'm raising it here first before opening a (yet another!) support case to see if others have the same issue.

The problem is simply as follows:

1. Open the Advanced tab in the Color Editor tool.

2. Pick a color correction on an image, using the tool's eyedropper, from an area of solid color in the image. (an area of solid color makes it easier to compare the readings)

3. Note that in the Color Tool dialog, the before/after RGB readings are the same - which is what they should be. Great.

4. Now, assuming that the color picker tool hasn't been moved from the selection point in the image, note the RGB reading at the top of the screen above the image. Is it the same as the RGB/RGB readings in the Color Tool?

For me it isn't. In fact, I would go so far as to say that the discrepancy is now worse than it was with Windows 8/C1 8.

The RGB readings at the top of the screen are way below those in the Color Tool dialog. Typically, I might get 169/173/188 on the screen, and 248/248, 248/248, 247/247 in the dialog. (This is a reading I just took from an image in order to post this observation.) That's almost an 80 point difference. This is nuts!

Which readings are correct? Which readings do I edit to? Or do I have something wrong in my setup?

Thanks.

D
Logged

Bob Rockefeller

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 548
  • macOS, iOS, OM Systems, Epson P800
    • Bob Rockefeller
Re: Color Editor Tool RGB readings anomaly / Windows 10 /
« Reply #1 on: December 14, 2015, 12:22:08 pm »

In Mac OSX 10.11.2 I see the same thing, but the difference is much smaller. The readings at the top of the window is a few points lower than the readings in the advanced color tool.

So it's on Mac, too.
Logged
Bob Rockefeller
Midway, GA   www.bobrockefeller.com

David Grover / Capture One

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1324
    • Capture One
Re: Color Editor Tool RGB readings anomaly / Windows 10 /
« Reply #2 on: December 15, 2015, 06:20:11 am »

I am using C1 9.0 with Windows 10.

This problem existed with C1 8 and Windows 8 and 10, and it's now worse, for me.

But, I'm raising it here first before opening a (yet another!) support case to see if others have the same issue.

The problem is simply as follows:

1. Open the Advanced tab in the Color Editor tool.

2. Pick a color correction on an image, using the tool's eyedropper, from an area of solid color in the image. (an area of solid color makes it easier to compare the readings)

3. Note that in the Color Tool dialog, the before/after RGB readings are the same - which is what they should be. Great.

4. Now, assuming that the color picker tool hasn't been moved from the selection point in the image, note the RGB reading at the top of the screen above the image. Is it the same as the RGB/RGB readings in the Color Tool?

For me it isn't. In fact, I would go so far as to say that the discrepancy is now worse than it was with Windows 8/C1 8.

The RGB readings at the top of the screen are way below those in the Color Tool dialog. Typically, I might get 169/173/188 on the screen, and 248/248, 248/248, 247/247 in the dialog. (This is a reading I just took from an image in order to post this observation.) That's almost an 80 point difference. This is nuts!

Which readings are correct? Which readings do I edit to? Or do I have something wrong in my setup?

Thanks.

D

Did you not create a support case and already get an explanation for the difference?
Logged
David Grover
Business Support and Development Manager

Bob Rockefeller

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 548
  • macOS, iOS, OM Systems, Epson P800
    • Bob Rockefeller
Re: Color Editor Tool RGB readings anomaly / Windows 10 /
« Reply #3 on: December 15, 2015, 06:47:30 am »

Did you not create a support case and already get an explanation for the difference?

If you did get an explanation, please share it.
Logged
Bob Rockefeller
Midway, GA   www.bobrockefeller.com

David Grover / Capture One

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1324
    • Capture One
Re: Color Editor Tool RGB readings anomaly / Windows 10 /
« Reply #4 on: December 15, 2015, 07:30:29 am »

If you did get an explanation, please share it.

"Hello,

This is becuase the value you see at the top is the color after it has been converted to your output space, whereas the value in Colur editor is based only in the input profile you are editing."
Logged
David Grover
Business Support and Development Manager

Bart_van_der_Wolf

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 8915
Re: Color Editor Tool RGB readings anomaly / Windows 10 /
« Reply #5 on: December 15, 2015, 07:50:43 am »

"Hello,

This is becuase the value you see at the top is the color after it has been converted to your output space, whereas the value in Colur editor is based only in the input profile you are editing."

That is what I also expected, but was not sure enough about to state it as a fact. It also suggests that the Color Editor works at a low level, early in the Raw conversion pipeline. The R/G/B and brightness display values at the top of the preview, and the optional Color readouts in the preview area are proofed output RGB colors. It is the latter colors that are meaningful to me, even when I use the Advanced or Skin Color Editor tabs.

It's a bit similar to Lightroom which shows RGB percentages while in Raw data, and 8-bit RGB values when in proofing mode.

Cheers,
Bart
Logged
== If you do what you did, you'll get what you got. ==

Dinarius

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1225
Re: Color Editor Tool RGB readings anomaly / Windows 10 /
« Reply #6 on: December 15, 2015, 12:56:32 pm »

"Hello,

This is because the value you see at the top is the color after it has been converted to your output space, whereas the value in Colur editor is based only in the input profile you are editing."

David,

Just to be clear, I want to edit and save in ProPhoto RGB.

So, with ProPhoto set as my Proof Profile (View/Proof Profile) and then Exporting in ProPhoto (obviously), I must ignore the RGB values in the Color Tool dialog and only read those at the top of the screen. Correct?

Thanks.

D.
Logged

AlterEgo

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1995
Re: Color Editor Tool RGB readings anomaly / Windows 10 /
« Reply #7 on: December 15, 2015, 01:09:33 pm »

"Hello,

This is becuase the value you see at the top is the color after it has been converted to your output space, whereas the value in Colur editor is based only in the input profile you are editing."

why the change of the proof space changes the values in ColorEditor then ? if you say "only" that implies that no other color profiles shall affect the numbers in color editor, but it is not so... I change the proof from Prophoto to sRGB and I see that the same color selected in ColorEditor will get new RGB numbers... can actual developers (a-la Esben for example) explain the flow of data for color editor ? is it [raw rgb] -> [color transform using input profile] only to display RGB numbers in color editor ? then there shall be no difference when I change the proof profile... if it is [raw rgb] -> [color transform using input profile] -> [color transfor using proof profile] to display RGB numbers in color editor then it is not "only".
« Last Edit: December 15, 2015, 01:25:46 pm by AlterEgo »
Logged

AlterEgo

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1995
Re: Color Editor Tool RGB readings anomaly / Windows 10 /
« Reply #8 on: December 15, 2015, 01:24:51 pm »

It also suggests that the Color Editor works

because C1 Color Editor can save adjusted input profiles as new input profiles it is certainly so - but what color editor displays as numbers is a different matter - see the point about changing proof profile affecting numbers in color editor... where and when it works modifying something (with AtoB0 tage from input profile with LUT, that converts raw rgb to a proper colorimetric color space, PCS = cieLAB or cieXYZ) is not linked as to what it displays to you in C1 user interface (it seems)
Logged

Bart_van_der_Wolf

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 8915
Re: Color Editor Tool RGB readings anomaly / Windows 10 /
« Reply #9 on: December 15, 2015, 02:47:27 pm »

because C1 Color Editor can save adjusted input profiles as new input profiles it is certainly so - but what color editor displays as numbers is a different matter - see the point about changing proof profile affecting numbers in color editor... where and when it works modifying something (with AtoB0 tage from input profile with LUT, that converts raw rgb to a proper colorimetric color space, PCS = cieLAB or cieXYZ) is not linked as to what it displays to you in C1 user interface (it seems)

Hi,

I agree it is not that intuitive, but maybe it is due to Rendering intent that feeds back to how the base color conversion is approached? I don't know the exact reason, because I do not have insight in the exact conversion engine's inner workings.

What's most important for me, is that the (delta) changes in the Color Editor are indicative, and they do not translate 1-on-1 to the output RGB changes (which is ultimately what counts, for me anyway). If one is interested then cloning a before-variant and placing a/some Color Readout(s) will do the trick.

Cheers,
Bart
Logged
== If you do what you did, you'll get what you got. ==

AlterEgo

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1995
Re: Color Editor Tool RGB readings anomaly / Windows 10 /
« Reply #10 on: December 15, 2015, 02:52:21 pm »

I agree it is not that intuitive, but maybe it is due to Rendering intent that feeds back to how the base color conversion is approached? I don't know the exact reason, because I do not have insight in the exact conversion engine's inner workings.

so why don't we ask David Grover to be so kind again (as he did recently) and ask people in P1 who knows to provide a more precise answer ! I mean as to why the change in proof profile means change in rgb numbers in color editor... input (camera) and output profiles stay the same
 
Logged

Dinarius

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1225
Re: Color Editor Tool RGB readings anomaly / Windows 10 /
« Reply #11 on: December 16, 2015, 04:58:35 am »

The reason I wrote to Support again was because the discrepancy between the RGB readings in C1 9.0/Windows 10 was even greater than C1 8/Windows 8. (I haven't had a reply)

Anyway, below is the text of my last correspondence with them re C1 8/Windows 8. I presume it's applicable to the current situation.

Me: "Thanks for the reply. So, let me rephrase the question........ If I want to edit *AND* save in ProPhoto RGB, what must I do? What preference settings must I apply to the software and where? And will I then see the same readings both at the top and in the Colour Tool dialog?"

Support: "Hi,

 

Editing in and saveing in Profot RGB does not make sense... becuase your camera is not aligned to Profoto RGB which is a working space, not a camera input space.  If you want to see the same numbers you can set your output profile to "Embed Camera profile."

 

I assume what you are really trying to do is hit certain numbers in the final image.  In that case place a color readout dropper (can be found in the list of eyedroppper tools in the curser toolbar) and place a color readout on the patch you want to edit.   then adjust the color editor for that color but pay attention to the final numbers displayed on readout, not the color editor tool.

 

Best regards,
Team Phase One Support"

Make sense?

D.
Logged

Dinarius

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1225
Re: Color Editor Tool RGB readings anomaly / Windows 10 /
« Reply #12 on: December 16, 2015, 05:48:59 am »

One further point on all of this...........

If I select View/Proof Profile and select Adobe RGB (1998), and then select Color/Base Characteristics/Adobe RGB (1998),  when I now take a sample from an image using the colour picker in the Color Tool/Advanced Tab, the RGB readings in the Color Tool dialog, and the RGB readings at the top of the screen are identical.

This is great and exactly what I wanted all along.

My problem is this: unless I'm going blind, ProPhoto RGB is NOT an option in the Base Characteristics drop down menu. Is it there? If not, can I add it to the list?

Thanks.

D.

Ps. Or let me rephrase my question above: MUST I leave Base Characteristics set to the camera model whose files I'm editing? (In which case the RGB reading discrepancies would continue)
« Last Edit: December 16, 2015, 05:57:06 am by Dinarius »
Logged

David Grover / Capture One

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1324
    • Capture One
Re: Color Editor Tool RGB readings anomaly / Windows 10 /
« Reply #13 on: December 16, 2015, 07:34:28 am »

One further point on all of this...........

If I select View/Proof Profile and select Adobe RGB (1998), and then select Color/Base Characteristics/Adobe RGB (1998),  when I now take a sample from an image using the colour picker in the Color Tool/Advanced Tab, the RGB readings in the Color Tool dialog, and the RGB readings at the top of the screen are identical.

This is great and exactly what I wanted all along.

My problem is this: unless I'm going blind, ProPhoto RGB is NOT an option in the Base Characteristics drop down menu. Is it there? If not, can I add it to the list?

Thanks.

D.

Ps. Or let me rephrase my question above: MUST I leave Base Characteristics set to the camera model whose files I'm editing? (In which case the RGB reading discrepancies would continue)

Base Characteristics describes the ICC profile of your camera.  If you want totally wacky colours, then sure, change it to something other than your camera profile!  But this makes no sense whatsoever.

But if its what you want then go ahead.

Are you adjusted colours in the color editor to achieve an exact output based on readings of a colour patch?  If so, then simply ignore the readings in the Color Editor and use the readings at the top of the screen.

Logged
David Grover
Business Support and Development Manager

Bart_van_der_Wolf

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 8915
Re: Color Editor Tool RGB readings anomaly / Windows 10 /
« Reply #14 on: December 16, 2015, 08:08:53 am »

Base Characteristics describes the ICC profile of your camera.  If you want totally wacky colours, then sure, change it to something other than your camera profile!  But this makes no sense whatsoever.

But if its what you want then go ahead.

Are you adjusted colours in the color editor to achieve an exact output based on readings of a colour patch?  If so, then simply ignore the readings in the Color Editor and use the readings at the top of the screen.

I agree, and suggest people to forget their old paradigms, like editing in ProPhoto RGB inside Capture One. It makes no sense, and one can always output (by making a recipe for that) to "Embed Camera Profile" (which is a large enough space to avoid losing data to clipping), and then in Photoshop convert from that colorspace to ProPhotoRGB if one must. Or one could do it the easy way, and just output in Prophoto RGB from an output recipe which is what the output was proofed for.

Cheers,
Bart
Logged
== If you do what you did, you'll get what you got. ==

Dinarius

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1225
Re: Color Editor Tool RGB readings anomaly / Windows 10 /
« Reply #15 on: December 16, 2015, 08:13:47 am »

Base Characteristics describes the ICC profile of your camera.  If you want totally wacky colours, then sure, change it to something other than your camera profile!  But this makes no sense whatsoever.

But if its what you want then go ahead.

Are you adjusted colours in the color editor to achieve an exact output based on readings of a colour patch?  If so, then simply ignore the readings in the Color Editor and use the readings at the top of the screen.

David,

Perfect.

What I had been doing is leaving the ICC Profile as per my camera, and editing to what I'm seeing at the top of the screen. It's exactly as it should be.

But, purely from a convenience point of view, if I could use the CT dialog RGB readings, there would be less too-ing and fro-ing, so to speak.

But, thanks for the clarification. Sorted now.

D
Logged

AlterEgo

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1995
Re: Color Editor Tool RGB readings anomaly / Windows 10 /
« Reply #16 on: December 16, 2015, 09:10:42 am »

"...then adjust the color editor for that color but pay attention to the final numbers displayed on readout, not the color editor tool..."

Make sense?


this part does not make any sense - why in the world you have some RGB numbers (and HSV representation of the same numbers) in Color Editor and nobody can clearly tell you what they actually are  ;D - and a direct recommendation from P1 not to pay attention to the numbers that C1 itself displays is odd, isn't it ?
Logged

Dinarius

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1225
Re: Color Editor Tool RGB readings anomaly / Windows 10 /
« Reply #17 on: December 16, 2015, 12:14:47 pm »

In editing, for colour accuracy purposes, I'm trying to hit specific RGB numbers - sometimes in ProPhoto, sometimes in Adobe 1998.

Unlike Bart, I want to edit in these spaces within C1.

Do anyone have the remotest clue of what their camera profile (Base Characteristics/ICC Profile) RGB numbers correspond to on a Gretag Colour Checker? I don't. But, I know exactly what the RGB numbers are for PP, Lab, 1998, sRGB and Apple. They're readily available.

So, I guess what I'm really saying is, I don't understand why the RGB readings in the Colour Tool retain the camera's ICC Profile values, when you've decided to edit in a standard space.

The colour I'm getting via the RGB numbers at the top of the screen is excellent. But, the workflow would be greatly simplified if I could fully utilize the Colour Tool.

Hope I'm making sense.

D.
Logged

AlterEgo

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1995
Re: Color Editor Tool RGB readings anomaly / Windows 10 /
« Reply #18 on: December 16, 2015, 12:34:25 pm »

I want to edit in these spaces within C1.

we were through this recently in a different topic - you can't edit in those spaces in C1 (as in when you move exposure slider the C1 code does not use any of those color spaces to adjust the data) - you can only use them as a proof profile (for example to see RGB numbers in color readout) and/or output profile (proof profile can be set to be = output profile automatically, if that's what you want), that's it... so while you are not editing (as in what RGB coordinates are that C1 code deals with inside) in them you (visually and/or by numbers and/or clipping indications) can control to some extent the result of your manipulations.

but then you need to clearly define what do you mean by "editing" and what is the purpose of what you try to achieve.


Logged

Bart_van_der_Wolf

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 8915
Re: Color Editor Tool RGB readings anomaly / Windows 10 /
« Reply #19 on: December 16, 2015, 12:54:45 pm »

In editing, for colour accuracy purposes, I'm trying to hit specific RGB numbers - sometimes in ProPhoto, sometimes in Adobe 1998.

Unlike Bart, I want to edit in these spaces within C1.

Hi D.,

You can hit these numbers, just set the output recipe to the required colorspace (assuming you want to also output in that colorspace). Alternatively you can set the proofing profile to the particular output colorspace via the menu, but that doesn't automatically becomes the output profile (it's only for proofing on display).

So, to me, switching between output recipes has more benefits. Firstly you can leave the proofing settings to use the active output recipe, and secondly you cannot forget to set the correct profile for output, WYSIWYG.

The only thing is that one has to use the colors readout above the preview (that's why it's a preview), or set one or more Color readout(s) in the image itself. Simple.

Quote
Do anyone have the remotest clue of what their camera profile (Base Characteristics/ICC Profile) RGB numbers correspond to on a Gretag Colour Checker? I don't. But, I know exactly what the RGB numbers are for PP, Lab, 1998, sRGB and Apple. They're readily available.

And all colors that are not in the colorchecker may be more or less off ... Really, in my experience, it's more important to have a good profile than to match 18 color and 6 greyscale patches out of the hundreds / thousands of actual/different colors in the subject. Only matching CC24 patch values (have you really measured the ones of your target? There is some variance depending on batch and age ...) creates a false sense of security if that's the only check. Real subject colors may suffer from metamerism, and the CC24 will never tell you that, because its colors have been designed to avoid metamerism issues ...

Cheers,
Bart
Logged
== If you do what you did, you'll get what you got. ==
Pages: [1] 2   Go Up