If I have correctly understood the DPL documentation, the DxO white paper about the new WGCS and what have been said about WGCS, PSC and PCD on this forum, I notice that the PCD algorithm can only be applied when soft proofing is activatedand when the target profile is a matrix based profile, which means when it’s not a printer profile but only when it’s a display profile.
However, those of us using soft proofing mostly do this when they want to print their images on some physical support. Soft proofing allows them to check whether there will be a problem when sending the file to their printer or to a printing service. In which case, let’s forget about PCD.
So, I will not be able to see the effect of PCD when soft proofing with a profile provided by the printing service or with a profile specific to the paper/printer pair that I’m using. The effect of PCD can only be seen and applied when exporting with a target profile compatible with PCD, that is, most often, sRGB.
Now, if I activate PCD during export with a compatible target profile, what about the corrections I have made when soft proofing with a target profile not compatible with PCD ? There’s some kind of inconsistency in this process, isn’t it ?
I understand that some users may need to soft proof with a display profile when they want to display their images on another display than their working display but I guess that this doesn’t happen that frequently. Soft proofing is mostly used in the case I described above : when printing or using a printing service.
So what ? Did I miss something obvious or is this feature almost useless for the moment, until it can accept printer profiles in a future version ? Or am I really that dense ?
But, since I find PhotoLab’s printing dialog to be inadequate, I always export to correctly sized TIFF files and then use either Canon’s own print manager or macOS preview or ColorSync Utility, which already contain soft proofing tools.
That is correct if you are printing your images or sending it to the printer that knows how to use a printer profile. However if you are not printing your images, and many do not, or if you are printing your images, but you are sending it to some standardized printing lab/shop. Say, wedding photographers. Many would sand hundreds if not thousands of images to a lab for printing, but the lab would not be some tech savvy so they don’t really want you to play around with printer profiles. They want sRGB JPEG or TIFF and they will print just fine. That would be a situation where you would soft-proof for print, but in sRGB. Of course the real benefit is in the algorithms themselves that can be largely automated.
The same is true if you are posting your images on social media or web, even if you print them you still are likely to show a digital version on social media. Since you might be working in DXO Wide Gamut Color space, but that is a working color space, not output color space. For better of worse, the standard is still sRGB. So again, “Preserve Color Details” is a great benefit.
Well, as you have noted earlier. It will work with RGB profiles as far as I’ve seen, but not with CMYK profiles or some really custom ones. Why is that the case I am not sure. Maybe DXO will add it to other profiles eventually.
Well, personally I use “Preserve Color Details” when I export images in sRGB, which is for every use. Weather its for web or personal sharing. So I find it extremely useful myself. Because I don’t have to worry about losing details in saturated areas of the image, I can just pretty much set it and forget it when I export. So for me its a default feature now.
Basically in the pats and in other programs when you want to squeeze out of gamut colors into a smaller gamut, usually sRGB output profile, you usually have to make some compromises. Either desaturate the oversaturated colors or cut off out of gamut colors and lose details.
In the past we had several options. None of the options was very good. We could do it upon conversion from one color profile to the other, using typically either perceptual or relative colorimetric options. One desaturated colors the other cuts off out of gamut colors. They do this with no real control and pretty harshly.
If you wanted more control, you could have manually desaturated colors by using tools such as color saturation sliders or vibrancy. And while saturation slider affects saturation eventfully across the image, vibrancy slider tries to add more saturation to less saturated colors and less saturation to already saturated colors while giving more importance to skin tone colors than other colors. Which is fine. But it’s not very precise when you want to deal with out of gamut colors. It’s more of a creative tool where user relies on appearance.
DXO “Preserve Color Details” is really the best of all these tools with great control and simplicity. You basically are saturating only the colors that are out of the gamut based on target color space, while making sure texture remains visible. This provides the most saturation you can get out of something like sRGB, but with all the rich textures and fine gradations that would be much more of a hassle trying to get in other ways. Here it’s just more or less automatic process where you can just use one check box upon export. I love it and use it by default since it was introduced.
I only apply it from soft proof panel when I’m dealing with particularly tricky images. Otherwise it’s on by default on export.
So, if I understand you well, when printing, PCD is actually here to replace Soft Proofing and the Perceptual / Relative options. We can either blindly trust PCD when exporting or we can still use the good old Soft Proofing method but without the benefit of the PCD algorithm.
I agree that printing from DPL is not an option anyway. It’s not like in Lightroom where soft proofing and then directly printing from LR is a reliable process.
The DxO documentation gives the feeling that PCD and Soft Proofing are features working together in DPL but that’s exactly the contrary. When printing, it’s one or the other. If you check with soft proofing by using a printer profile (and the Perceptual / Relative options) and then eventually export to sRGB using PCD, the results will be different.
Using PCD might give a better result than using soft proofing, but you have to trust it. There’s nothing to check/see before exporting unless you check the effect of PCD on the display before.
Not a surprise that these new features have generated so much confusion. The White Paper already referenced elsewhere in the forum is not going far enough in the explanations. And a not-so-easy-to-understand feature generally is a source of trouble.
Well there are several ways to use “Preserve Color Details”.
One is obviously in the Soft Proof panel when using RGB profiles such as sRGB.
The other place is yes when you export, you can choose what color profile you want to export as and weather or not you want to use “Preserve Color Details”.
Well, “Preserve Color Details” is not there to replace Soft Proofing, its there to compliment it. As you can see Perceptual / Relative Colormetric options are still there because its usually the part of the conversion process from one color profile to the next. “Preserve Color Details” is there to give you extra control and since its related to problems of dealing with out of gamut colors, you will find it in the soft proofing dialog box.
It is actually located under Advance settings of the Soft Proofing panel. It gives you intensity from 0-100. Meaning you can still rely on Perceptual / Relative Colormetric, methods but you have now more advance, extra complimentary settings “Preserve Color Details”
Worth remembering is that if you are dealing with narrow gamut space or even black and white or very destaturated scene, you may not need “Preserve Color Details” because there is no problem in that area, but you might want to simulate approximately the contrast and brightness differnces of the print (paper and ink) vs what you see on screen. Perceptual / Relative Colormetric, methods are just standards how we do conversions from one color space to the next, you can’t have no method. And while they can deal with some out of gamut color issues, they are very crude and with not much control from the user side. “Preserve Color Details” is complimentary advance setting that adds additional functionality and control. That’s all. Its not a standard the way Perceptual / Relative Colormetric is, its unique to DXO.
It is worth remembering that soft proofing is not the same as hard proofing, meaning first one is approximate simulation of a print or conversion to another color profile and hard proof is actual physical print.
The obvious advantage of soft proofing is that it can saves time and ink and show potential problems before you do hard proof, and it can obviously do something hard proofing cannot do. It can help you simulate other digital devices based on color profiles.
Now, if I activate PCD during export with a compatible target profile, what about the corrections I have made when soft proofing with a target profile not compatible with PCD ?
When softproofing with a non-matrix profile profile (that is any RGB or CMYK paper profile) all other corrections except the then deactivated PCD can be applied with export, which in case of a raw-file includes Color Rendering → Intensity / Protect Saturated Colors – additionally to Exposure, Vibrancy, HSL etc.
Using any of these tools affects the output, which when running a monitor that renders (almost) all colour spaces, can be done visually, but when restricted e.g. to sRGB, one has to rely on the Destination gamut warning – which is more a ‘blind flight’ than real help.
I understand that some users may need to soft proof …
…
So, if I understand you well, when printing, PCD is actually here to replace Soft Proofing and the Perceptual / Relative options. We can either blindly trust PCD when exporting or we can still use the good old Soft Proofing method but without the benefit of the PCD algorithm.
As mentioned, exporting to sRGB is recommended for any social media use and more than often can be necessary for sending out to a printing provider – as I did for photobooks.
Softproofing with a matrix profile then also allows to use PCD and visibly decide between colour saturation and/or texture – no ‘blind flight’ with the right monitor.
REC2020 = is said to be the closest representation to ‘unofficial’ DxO Wide Gamut CG2730 Nat 6500K = my monitor’s native colour space (containing AdobeRGB + DCI-P3) *) SC-P800 CPFR = my ‘best’ paper (semi/glossy type)
