RT standard ICC profiles

[Beep6581]

Originally reported on Google Code with ID 938

Hi all,

We found that DCRAW default camera matrixes are sometimes pretty average and Huelight
(Colin Walker) and others can provide way better matrixes plus LUT ICC-profiles.
The matrixes leave not much discussion how to correct them (see issue 930), but question
is how to distribute and UI-support the ICC profiles.

Here is a suggestion to have something to discuss about. Since our discussions tend
to get a bit wild and without clear dicisions, let’s maybe try it with some topic number
to divide the agreement process.

1. We’ll have our own “rtdata\iccprofiles” directory in Mercurial to store and version
the profiles. This directory is distributed with the build and stored in a fixed position
below program root for easier referencing (like e.g. themes and language files)

2. The ICC profile names will be simple to tie to the technical (EXIF) names so they
can easily be auto-mapped. The ICC files are GPLed, however Colin would add his creator
tag in the ICC profiles.

3. We’ll have an additional ICC UI option “Auto selection”. If it finds a corresponding
ICC profile from the EXIF in our directory, it will take it, otherwise fall back to
DCRAWs default matrixes (hopefully improved).

Open question from my point of view: How to communicate the proper white balance settings?
Should we maybe rename the options to clarify to the user which to take? (Simple default
profile/ Advanced ICC profile if available/ Custom profile)

Looking forward to the discussion
Olli

Reported by oduis@hotmail.com on 2011-08-24 19:41:55

[Beep6581]

1,2,3) Agree.
I missed the need to know white balance settings and where?

Reported by ffsup2@yahoo.it on 2011-08-24 20:42:40

[Beep6581]

It's not directly related to the ICC profiles, but the cams have different white balance
settings. For instance a DMC-GH1 (an extreme case) has 5800 in sun with just DCRAWs
standards, while my D700 has around 5000. The information where the standard is would
be important to the user to get the best color. And you can derived it e.g. from a
color checker chart that we use to measure the profile.

Reported by oduis@hotmail.com on 2011-08-24 20:52:24

[Beep6581]

My impression was that ICC profile would be different under different lighting conditions
(E.g. definitely if mixed lighting is used). This is a reason why dcp supports dual
illuminant settings to interpolate between them. In case of dcp this is done for tungsten
2800 and daylight 6500  illuminants as border conditions.

Considering this, I suppose, we could have more than a single icc profile per each
camera. Icc profile set per camera could include profiles for various color temperatures,
for example - 2800K, 5000K, 6500K. ICC profile would be auto-selected from this set
based on the value of the camera white balance temperature, so it would be the closest
to the one for which ICC profile is available.

Is there a way to adapt support for dcp in RawTherapee? May be this is a better direction?

Reported by michaelezra000 on 2011-08-24 22:18:45

[Beep6581]

For ICC profile naming/auto-selection I would suggest
RawTherapee_[Maker]_[Model]_[Temperature]_YYYYMMDD.icc

[Date] - to will allow versioning, although, not entirely sure this is necessary.

Reported by michaelezra000 on 2011-08-24 22:23:14

[Beep6581]

Hi Michael,

OK as you say ACR is using internally two XYZ matrices for each supported camera.

One built under a D65 Kelvin source and another built under a Standard A Tungsten light
source.

ACR interpolates and extrapolates between the two...

ACR's extrapolating approach is mathematically wrong and way to stringent for real
life images, and IMHO it makes for bigger errors all over the Kelvin scale, except
maybe at exactly 2856 and 6503 Kelvin.

But we are not talking about ACR...  

RT4 has only the D65 matrix, so in theory is only correct at 6503 Kelvin.

My profiles are correct at 5000 Kelvin, this does a better job at the outer borders
of the Kelvin scale. This single profile and colour temperature approach is better,
not just for images close to 5000 Kelvin, but also better as a basis for the outer
Kelvin values when a custom WB or click WB is used.

RT4 as it stands would only be 100% correct at 6503k anything below 5500k would be
noticeably wrong to the human eye even at the correct WB.

This is not the case with a good profile at 5000k this would be perceptually good at
any kelvin even on the borders given a custom WB or good click WB.

IMHO offering loads of ICC profiles at exactly x Kelvin will only further confuse and
alienate new users and not really offer any benefits.

Colin

Reported by huelight on 2011-08-24 23:28:38

[Beep6581]

Wow would a single ICC profile be accurate for images shot at various (constant uniform
color) lighting conditions?

Reported by michaelezra000 on 2011-08-25 00:38:03

[Beep6581]

oops - that was "How" :)

Reported by michaelezra000 on 2011-08-25 00:38:35

[Beep6581]

I agree with Colin, I have very few images where the WB temp is over 6000K, most are
in the range 4700-5500 (I shoot mostly outdoors).  Best to have a profile for typical
outdoor lighting, not some extreme that is rarely achieved.  Also, RT internal RGB
color spaces are (Bradford) adapted to D50 illuminant, which is ~5000K color temp.
 So best if the profile matches up with the internal RGB color space, IMO.

We might want to think down the road about a second profile for typical indoor lighting,
and/or better adapting the internals for lower temp illuminants.  So I favor a somewhat
future-proof labelling for profiles that includes the color temp, as Michael originally
suggested (somewhere?).

Reported by ejm.60657 on 2011-08-25 02:50:33

[Beep6581]

I would also just supply the standard outdoor sunshine profile and not one for tungsten,
but for practical considerations.
Here in Europe the normal tungsten lamps are stepwise forbidden by law (CO2 waster)
and replaced by LED and espc. small fluorescence lamps of various warmth levels (and
a difficult spectrum by the way). So you find very mixed light situations with some
tungsten, some fluorescent, some LED and you cannot really build a one-size-fits-all
profile for that I guess.
So I think it’s better to stay with one standard sunlight profile, as Colin suggest.

Reported by oduis@hotmail.com on 2011-08-25 06:00:40