Color Management Canon Software vs. Capture One Pro
Hi,
I have been using the Canon RAW software (Digital Photo Professional) that came with my EOS 30D so far and now I consider moving to Capture One.
When comparing the Canon software with Capture One I noted that colors appear to be shown significantly different, with Canon they appear more vivid than with Capture One. I have tried that with pictures shot both with the sRGB and Adobe RGB camera setting.
I suspected the color management and compared here. In terms of output, I don't have a profile for my monitor yet but I use the same profile (sRGB) in both softwares, so even if the result might be wrong, as far as my understanding goes it should at least be the same for both programs.
In terms of input I use the "Canon EOS 30D generic" profile. Switching to sRGB or Adobe RGB (depending on what the picture was taken with) makes the colors appear even less vivid in Capture One.
So my first question would be what does the 30 D generic profile does differently then the sRGB/Adobe RGB color space? Does the 30 D generic recognizes whether the picture was shot in sRGB or Adobe RGB and behaves differently accordingly?
And second, where could be the difference between the way the Canon software and Capture One displays an image (without performing any modifications)? The Canon software does not have the option to choose a color space for importing a RAW picture so I assume it just takes whatever is embedded in the file. However, then why does the colors look notably different from what Capture One shows?
Would be great if anybody could help me out here because I generally like Capture One more then the Canon software, however the colors trouble me a bit.
Thanks and regards, Robert
I have been using the Canon RAW software (Digital Photo Professional) that came with my EOS 30D so far and now I consider moving to Capture One.
When comparing the Canon software with Capture One I noted that colors appear to be shown significantly different, with Canon they appear more vivid than with Capture One. I have tried that with pictures shot both with the sRGB and Adobe RGB camera setting.
I suspected the color management and compared here. In terms of output, I don't have a profile for my monitor yet but I use the same profile (sRGB) in both softwares, so even if the result might be wrong, as far as my understanding goes it should at least be the same for both programs.
In terms of input I use the "Canon EOS 30D generic" profile. Switching to sRGB or Adobe RGB (depending on what the picture was taken with) makes the colors appear even less vivid in Capture One.
So my first question would be what does the 30 D generic profile does differently then the sRGB/Adobe RGB color space? Does the 30 D generic recognizes whether the picture was shot in sRGB or Adobe RGB and behaves differently accordingly?
And second, where could be the difference between the way the Canon software and Capture One displays an image (without performing any modifications)? The Canon software does not have the option to choose a color space for importing a RAW picture so I assume it just takes whatever is embedded in the file. However, then why does the colors look notably different from what Capture One shows?
Would be great if anybody could help me out here because I generally like Capture One more then the Canon software, however the colors trouble me a bit.
Thanks and regards, Robert
0
-
First, there is no "color space" of a raw file. It just stores the discrete digital values that were converted from the linear, analog voltage levels created by light striking the CMOS (in the case of this Canon) chip.
The color space selected in the camera when shooting raw, is just a tag stored in the EXIF of which software may make use when converting the data to JPEG or TIFF. It just indicates a preference to a raw converter, it does not change the data that goes into the file. It is also completely ignored by Capture One.
If there was anything that could come close to being described a the color space of the raw image, it would be Capture One's camera profile. Indeed if you select "Embed Camera Profile" when processing the images, it does get converted to an ICC profile and truly becomes the color space of the output.
As for why different conversion programs look different, it comes down to how they interpret the discrete digital values in the raw image data. Capture One with its camera profiles, and default exposure settings has chosen to go more for accuracy of color. Canon seems to have gone more to make the colors "pop", and over-saturates a bit.
That said, I too like a little more pop than what I get by default from Capture One, so I have set in the preferences to default to +1.5% on the Color Saturation (CS on the Exposure tab). That makes every image look a little more lively. Most of the time that's just right for my preferred look, but some photos need a little more or less. In those cases I just salt to taste, as they say.
You should be using "Canon EOS 30D generic" profile for input. sRGB is probably best for output unless the rest of the tools you'll be using are properly color managed. If they are, then I'd just embed the camera profile in a 16-bit (really needs to be 16-bit to avoid chances of banding, as the camera's gamut is so huge) TIFF, as this will preserve the largest amount of the original color space. A half-way point would be Adobe RGB in an 8-bit TIFF (if your software color managed, but doesn't handle 16-bits per channel well).
End the end, I'd just play with the CS slider until you get the look you want. Additionally a custom exposure curve may help a little more. If that still doesn't do it, I hear the color editor in the Pro version is very powerful at making changes to the base camera profiles. I only have the LE version so I'm familiar with it.0 -
[quote="robertpollai" wrote:
However, then why does the colors look notably different from what Capture One shows?
I think "Christopher" explained it very well. Every RAW-converter handles the raw-files in a different way. Not completely different but slightly.
It's a huge advantage of C1 that it manages the input RGB-Data of the raw-files in icc-profiles (or better *.icm, which is the same). The calculation from input data to working or other destination color spaces is accurate from the very beginning.[quote="Christopher Meadors" wrote:
A half-way point would be Adobe RGB in an 8-bit TIFF (if your software color managed, but doesn't handle 16-bits per channel well).
The sole point I disagree with you. AdobeRGB and 8bit does not fit together very well - indifferent as in *.jpg or *.tif.
sRGB is only just able to manage 8bit. The 255 tonal values of each RGB chanel are too much compressed in AdobeRBG-8bit files. The truth is: you are losing differences in tonal ranges. Too, AdobeRGB has less differences in the low tonal ranges according to it's tonal curve.* I know that it is a very common color space. But I don't understand that so much people still using it - especially photographers. If you work with 8bit (maybe because of data size?) choose sRGB. Too and more than ever if you shoot jpg in camera.
A question: does anyone knows which rendering intend C1 uses when transforming the camera profile into working spaces? As far as I notice it should be relative colormetric. But I'm not sure.
Thanks a lot!**
Best Regards.
* edit: this is critical if you work in 16bit too. There are better options as working spaces like eci-rgb (V2), l-star, photogamut.rgb... depending on what you want to output or your workflow is structured at all. and, for sure, you can choose to embed the camera profile itself.
** edit: got it: "All profiles supplied from Phase One are set to Perceptual rendering intent and Best rendering quality." Thanks!0 -
I'm going to disagree about the Adobe RGB in an 8-bit per channel image file. Sure, 16-bit is preferred for any gamut larger than sRGB, but the slightly larger Adobe RGB doesn't really suffer when mapped to 8-bit steps. It is my opinion that the possibility to store some of the more saturated colors that are outside of the sRGB space is worth the small chance of banding artifacts showing up in a very smooth gradient. I doubt any natural photograph could create a situation where bands would be visible in Adobe RGB and not suffer more by being mapped into sRGB.
Ideally the workflow would be completely 16-bit (unsigned) clean, and properly color managed so the workspace would make use of the camera's native range. With the conversion of color space delayed until just before output, and tuned to that device (just like sharpening*).
As for in-camera JPEGs. I doubt many people here use them (being a forum for raw processing software), but I still stand by my opinion that the loss of color space by squeezing the large range of color that today's cameras can capture into sRGB is worse than the possible artifacts of Adobe RGB. But working on JPEGs is futile; too much other information is already lost.
That brings me to the recommendation of 8-bit TIFFs, and not mentioning JPEGs. If there is any chance of an image being re-saved after it is output from Capture One, stay away from JPEG. The lossiness of the Fourier cosine transform coupled with the conversion to the YCrCb color system (does Capture One use 4:4:4 or 4:2:2?) makes JPEG useless as a work file and any other arguments over sRGB vs. Adobe RGB moot. That is the reason I only spoke of TIFFs.
Don't get me wrong, Capture One's High Quality JPEG output is excellent, but I'd not use it for anything more than to e-mail as a proof, or if the images are destined for a web gallery straight out of Capture One; with no further processing applied. Here sRGB is appropriate because the viewer's color management abilities are unknown.
These are just my opinions, and don't have much to do with Capture One. This discussion would be more at home in a general photography forum (and has been held countless times in each of them). So this is the last I'll speak on the subject, because I've seen these discussions deteriorate into a pit of flames all too often.
*Oh dear, that's another can of worms.0 -
I think it has to do with the Picture styles in DPP.
The Standard one seems to boost the reds quite a bit IMHO.
Try using the Neutral or faithful setting in DPP. It makes the colours flatter but closer to what they should be.
Its better to use these if you are going to do more PP in another program as well.
All IMHO of course.0 -
I had a similar problem with a 400D, and i'm sure the issue is indeed with the picture styles. Its just that in some cases the picture styles do just what I want! eg. the landscape style seems to pop the greens and blues, whereas increasing saturation in C1 makes all the colours pop and it looks silly.
Does anyone know of a way of replicating these styles in C1?0
Post ist für Kommentare geschlossen.
Kommentare
5 Kommentare