@Aleph RawTherapee ships an ICC profile for the Canon EOS 5D, it also ships a DCP profile for it. Look in the iccprofiles\input and dccprofiles folders. If DPP ships 37 profiles for your camera model alone, these profiles likely include some effects baked into them. Unless you meant 37 profiles in total, not just for your camera.
Regardless, this exercise seems like a waste of time. The input profile’s job is to give accurate colors. It does. A different reading of WB temp and tint is not a problem.
Yes I have found the RT icc profile for the 5D, I can see only a subtle diference: the dark blues are less dark. I can see it in the nails, and (in other photographies) in the sky.
I also tried all the 37 DPP Profiles! I only can see two effects:
1- Some of them look very very dark
2- All the CNZ*.icc files look the same. The blues are more dark, and the red, and skin colors are only a little bit more yellow
Yes maybe you are right.
Some times I have some trouble for get the right colors in RT. But probably is my camera and the shooting conditions. I started as nature photographer using DPP because DPP got the best colors for my camera, Adobe Camera Raw and Lightroom are no the best for Nature photographie.
My question was my surprise about the diference 5000K-4650K
Now the question is clear, more o less…
I like RT and I use RT on PCLinuxOS almost exclusively in the last 3 years. Even for nature photographie
@Aleph to be clear, the input profiles in RawTherapee are designed to give accurate colors. This is not the same thing as aesthetically pleasing colors or colors with a certain “look”. “because DPP got the best colors for my camera”, this is likely a result of DPP using a tone curve + a color mapping which gives a certain “look”. You can manually do the same in RawTherapee, then save the settings as a default PP3 for raw files. For example when working with skin one generally wants to smooth-out red splotches and shift reddish colors closer to skin colors (e.g. using the Vibrance - Skin Colors tool https://i.imgur.com/Fk0BgAM.jpg ).
RawTherapee first checks for a DCP input profile when opening a raw. If none exists, it looks for an ICC profile. If that does not exist, it uses hard-coded values from dcraw. As RawTherapee ships both an ICC and DCP input profile for the Canon EOS 5D, it will use the DCP by default. You could select the ICC manually https://i.imgur.com/c3R6Luh.png though it looks almost identical to the DCP.
Do you have a color target like the X-Rite ColorChecker Passport? If so, if you shot it under daylight and tungsten light I could improve the DCP.
DPP’s available files seem to be Canon’s generic profiles; FA.icc is Faithful AdobeRGB, PS.icc is Portrait sRGB, and so on. If I remember correctly, in daktable they can be made usable via the “unbreak input profile” module. They can be used for some comparisons, but I also wouldn’t bother with them too much.
@Aleph A good way to learn about your raw file is to use exiftool or something similar. Other than that, @Morgan_Hardwood pretty much covered everything. A snippet of exiftool output:
WB RGGB Levels As Shot : 2180 1024 1024 1616
Color Temp As Shot : 5019
WB RGGB Levels Auto : 2180 1024 1024 1616
Color Temp Auto : 5019
WB RGGB Levels Measured : 2087 1022 1025 1717
Color Temp Measured : 4648
WB RGGB Levels Daylight : 2236 1024 1024 1577
Color Temp Daylight : 5200
WB RGGB Levels Shade : 2608 1024 1024 1339
Color Temp Shade : 7000
WB RGGB Levels Cloudy : 2427 1024 1024 1450
Color Temp Cloudy : 6000
WB RGGB Levels Tungsten : 1570 1024 1024 2399
Color Temp Tungsten : 3200
WB RGGB Levels Fluorescent : 1949 1024 1024 2245
Color Temp Fluorescent : 3763
WB RGGB Levels Kelvin : 1141 1024 1024 3372
Color Temp Kelvin : 2400
WB RGGB Levels Flash : 2497 1024 1024 1438
Color Temp Flash : 6214
Average Black Level : 2048 2048 2048 2048
Raw Measured RGGB : 415520 180628 178529 249230
Per Channel Black Level : 2047 2047 2048 2048
Normal White Level : 15165
Specular White Level : 15677
Linearity Upper Margin : 10400
This is true, so they aren’t truly neutral or accurate, even though we have a picture profile named neutral.
exiftool is an excellent way to see the metadata of your camera or at least that which the camera decides to reveal. Different raw processors heed different portions of this info. It may also give us a hint on what is baked into the JPEG (or the RAW), if you know what you are looking for.
Regarding your question, you might be interested, for example, in the WB multipliers, the black / white points and the temperature presets of the camera. One thing that can be seen right away is that your camera’s auto WB wanted to be even warmer than your manual temp but sunlight is a bit less. Looks like the flash may have influenced the colors as well.
Another thing to note is the differences in k between camera presets. DPP would correspond to this but RT clearly has another interpretation. You could probably draw some graphs to compare if you really wanted to, but it might be hard to compare since DPP does more stuff to influence the default output than RT. Photography is just a hobby for me, so I might be wrong on a bunch of things, but hope this helped just a bit .
@Aleph Personally what I do in respect to camera profiles is study raw output via ones that are available with camera jpg’s taken with it’s default settings less noise reduction and sharpening etc. Any editor can usually be used to open jpg’s so even the histograms can be compared.
Might seem to be a strange thing to do but camera manufacturers go to an unbelievable amount of trouble to get the best possible jpg’s out of cameras. It will account for metering and sensor characteristics.
As some one may have mentioned it can also be possible to find the icc files in the manufacturers own software. That doesn’t work out well on Nikon - you might say that they produce a fresh one for every shot the camera takes. Due to that it’s not unusual for people to use Nikon’s software for all raw development. Ufraw can be used to mimic the effect but it’s tricky to do in practice. As far as I am aware it’s the only package that offers the extra style of curve panel that is needed.
On Canon their camera profiles used to seriously limit highlights compared with what is available in raw. There is insufficient info about these days to tell if they still do. Adobe standard bought more in but the net effect has to be less contrast in highlights.
Another good source of camera profiles can be Photivo. It’s an unusual package that specialises in exotic filters to some extent and probably wont be everybody’s cup of tea. Not sure where the profiles come from but one I used on an Olympus camera some years ago was way better than any other I could find in a number of respects. Especially white balance.
I have done a lot of comparisons for select a correct profile. This is a very important thing, aI think, the colors change a lot. However it is very hard to decide wich one is the better.
Photivo maybe has a better red, but the whole image is a little bit grennish in some situations…
The Adobe Camera Standard dcp with the tone curve selected is very pleasant, I can control better the colors posterization, but the blacks are a bit faded.
At the end I think I will stay with the RT auto-select profile (changing a little bit the reds) and compare with others profiles in some situations.
Processing should be capable of boosting the blacks.
I’ve tried to explain what is going on in this area several times but often to no avail. These are the gradation curves built into an E-M1
Each of those curves represents a camera profile. Bottom axis is raw stops in, vertical jpg bit depth out. The slope of the line indicates contrast. It always tends to get lower at each end of the brightness range. When a raw file is opened in an editor what is shown is the jpg bit depth after some curve of some sort is applied. The idea of PP is to do what ever you want to do with it and that can include in a simplistic fashion all of the stops in the raw file but the fact that this is more than can be shown on the screen means that contrast in various places on the tone curve need to be changed. That is what things like highlight and shadow recovery actually do.
Can’t understand how you can get posterisation other than having the wrong working profile or working adobe rgb in srgb or simply pushing things a lot further than they can be. aRGB even with the correct equipment is more likely to have that problem sooner.
Personally I feel that you should be ok with any of the adobe profiles other than maybe high key if there is one and it’s better to try and get to grips with pp rather than trying to achieve what you want to do via profiles. The important thing is a camera auto white balance that works reasonably well in most situations. The colour scaling factors used will have an effect on that so in some respects it’s best to dig out the ones that are provided in the manufacturers software but adobe should achieve the same thing. Faithful,natural or neutral may be a good starting point if there are any named like that or standard.
Maybe the correct word is “oversaturated” I don’t know really.
This happens sometimes, spacially in the red and blue satured colors in low light zones, when I contrast the image with a S curve in “Standard” mode. In “film like” mode the problem dissapear but the colors change…
That is why there are various modes. “Film-like” is also known as “Adobe-like”.
@Aleph if you have any specific issues, post a raw + PP3 + screenshot of the problem, preferably in a new topic if it doesn’t concert the white balance.
.
I’ve tried to explain what is going on in this area several times but often to no avail. These are the gradation curves built into an E-M1
