Yes, that sounds to be a good explanation.
I can’t answer the question in what profile that image is send: wide gamut, the monitors or the soft proof when used.
I lean towards wide gamut. – The color space of the monitor determines what the user sees, while the soft proof is just a screen simulation, a preview of the later printed output.
It’s slowly becoming clearer. I also read through Wolf Hauser’s article on wide gamut again. That also makes things clearer. RAW files are supplied by the camera with 12, 14, 16 bits (perhaps even more in the future). Photolab processes optical corrections and noise reduction in this environment. In the past, further (color) processing was done in AdobeRGB gamut. With the introduction of DXO Wide Gamut, this color space has been expanded or redefined. If I have understood this correctly, it is about the best display compromise for the professional image editing process. This data must then be mapped to the desired printer. There are probably very expensive printing machines that can print the DXO Wide Gamut directly, for affordable printers this information must be mapped to the printer’s capabilities. The mapping is done by the RIP software, whether it is built into the driver and/or firmware of the printer or printing machine. In this case, the RIP software alone is decisive for the quality of the printout. A necessary ICC profile can be read in directly for option 1 via Photolab, in option 2 with external printer management from the configurable RIP driver. In this respect, this is now consistent for me. For now, thank you for your thoughts and further suggestions.
The best working gamut is the output gamut. No conversion has to be done. But for the program doesn’t know what your output will be, a wide gamut is used. One can convert that to a smaller gamut, but not the other way. There’s no quality issue involved.
Every output device has its own gamut. So has a printer. When the program sends the image to an output device it’s using a “Profile Connection Space” to convert it to the output gamut. So also to the printer. I don’t know what happens when it is said let the printer driver do the conversion. If I’ve softproof on and use a ICC profile that only prints B/W it doesn’t matter whether I choose PL or the printer to do the conversion.
See https://www.color.org/profile.xalter for the Profile Color Space.
George
I didn’t know what RIP software meant so I looked up. It does a conversion from a vector based image to a raster based image. PL is raster based.
George
I took a couple of screenshots with my Eizo CG2730 and set it up to sRGB, so they are easily to follow on a sRGB capable screen. The red overlay shows the affected colors, when choosing a target with a small(er) color space.
For better comparison the paper & ink simulation is not activated, while in reality
.
So that happens with the output (color space).
It’s creating a raster image from a vector image. No color gamut involved.
George
I think he asks when using the printer driver.
George
I just found out that the color space in the soft proof is used, irrelevant if SP is on or off. Same result for SP is on or off and with managed by PL or printer. I think the only difference is that when you select managed by PL you can overrule the output color device in SP.
I doubt if the printer driver is doing any color management.
George
This discussion has nicely illuminated the uncertainties and user-unfriendly aspects of allowing printers to manage colors. Why choose this option? I see in the PrintFab manual the recommendation is to allow applications to manage colors. PS and PE are named as photo editing examples and the general message is clear. That would be my recommendation for PL as well.
The current PL print module is anemic, but it does work as advertised. If you choose to print directly from PL, allow PL to manage colors. Assuming you are starting with a RAW file, I would take the extra step of exporting (to the original image folder) a full-size 16 bit TIFF with an embedded actual ICC color profile. Send the exported TIFF to the printer. This extra step will reduce or eliminate any color mismatches or ambiguities that may exist in the file metadata. After printing, you can always delete the TIFF if storage is an issue.
I’ve only ever tried to print from PL once. The print was rubbish. This was in stark contrast to the print of the same image I’d done moments before using an ancient version of Photoshop, it was fine.
Everyone I know who prints follows a simple rule:
If you are printing a colour image, let the application manage the colours BUT if you are printing B&W, let the printer manage the colours.
In my experience there’s a reason this rule exists. It works.
to quickly reiterate … one has 2 possibilities,
about Soft Proofing
SP itself is nothing more than a simulation/preview of what you will see, well, most likely. Of course, SP is not able to simulate the texture of the paper such as glossy, matte or in between, but it does indicate whether and what colors the printer/ink/media combination will produce (if the monitor is capable of it) together with the color of the paper (cool, neutral, warm).
That information is stored in a specific ICC profile,
But SP does NOT change the image. This is what you do yourself, if you want / need to adjust something – while there is an “exception” …
IF you SP
and select to export like
the above settings are applied during export. → Suppose that’s what you noticed. ←
Still, I don’t like exporting while applying a printer profile. – I leave color management to the application (where I choose the paper profile and rendering intent) and print from the printer driver (with custom presets to reduce errors).
I did the same test again and now I get different results. I don’t know why.
I still wonder who is responsible for color management during printing when selecting printer driver. From what I understood now it’s the operating system. It’s an extra layer between the program and the printer. The printer driver is taking care that the input is translated in a way that the printer can print. Color management is changing that input first and send it then to the printer driver.
With SP on the image shown in the printing windows of PL is the image converted to the used icc profile when using “managed by PL”, when using “managed by the printerdriver” I don’t know what is used.
I don’t print. I’ve it printed.
George
Oh dear, I didn’t think I would do so much research at first. I had completely ignored the concept of the Profile Connection Spache, but then I started searching. I currently imagine the following:
The Profile Connection Space (PCS) is the superordinate reference that contains all colors that can be perceived by humans. It is standardized as CIE (LAB (Lab*) or XYZ).
The ICC profile describes the mapping from a color space to the PCS (source mapping) or from the PCS to a color space (target mapping). Tables for an interpolation and/or parameter series for a transformation can be used for the description.
If image information is to leave Photolab (saving an image file or printing the image), the Photolab wide gamut and the required/desired target color space must be calculated via the PCS.
Photolab provides an “internal” ICC for this purpose, which describes the mapping from the wide gamut to the PCS.
In addition, the printer interface is provided with the ICC that describes the mapping from the PCS to the printer. This ICC is determined with appropriate measurements for each output device and depends on various properties - paper type, ink, printer series, signs of ageing, etc. and is registered in the user interface by the Photolab user.
The software that processes the print output then does the following: Analyzing the source and target color spaces using the mapping information from both ICC profiles and then (in the vast majority of cases) calculating one defined mapping table(s). This mapping table(s) then allow both color spaces to be mapped to each other without multiple rounding errors.
This means that no conversion of image information from the source to the PCS and from there to the target is carried out, but the PCS is only used as a reference for the calculation algorithms from the source to the target.
This brings me back to my initial question: Who performs the conversion for the printout and what happens when files are saved?
I still assume that the corresponding printer driver of the printer manufacturer or another RIP software takes over this conversion task before printing. Photolab only provides the image information and the “internal” ICC at this interface.
Photolab performs the conversion when a file is saved. This is where two ICC profiles come into play: the “internal” one for mapping from wide gamut to PCS and the device-independent ICC profile from PCS to TIFF and JPG.
I’ll leave DNG out of the equation here, that’s certainly a topic in its own right.
The device-specific ICC profile, which is embedded to the target file, is then needed again when printing out the file to the target device.
I don’t think the printer driver itself is doing any color management. It let the OS do it.
George
According to my findings, this is a controversial topic for Windows. Windows has an integrated color management system that should go beyond ICC color management. This requires WCS profiles instead of ICC profiles. This has failed - try to find a WCS profile.
The consequence is that ICC profiles are not even processed by Windows, but are passed on to the application programs and device drivers for processing. If these programs do not process ICC profiles either, these profiles are useless. Under certain circumstances, this could also mean that your print results are not always reproducible.
With Windows 11, a new approach has been taken, which in some cases means that processing chains that are reliably based on ICC profiles no longer work correctly. We will have to see how the system develops.
Simply drop the idea to leave color management to the printer driver.
Otherwise you have to figure out the right settings as already explained.
I don’t have win11 os but am about to.
What is this color management problem with win11 ?
Does it alter professional color managed softwares and pipelines ?
WOW !
what I read from a message posted 10 mn ago in a reddit discussion :
"The biggest difference you’ll see is that, if you have a wide gamut display, in SDR everything on the desktop that were not color managed before are now clamped to sRGB. So no oversaturated UI, wallpapers, icons, etc. But any previously properly color managing apps like Photoshop, Lightroom, etc, are broken. In MS documents they clearly stated ".
Have to dig into it to know the truth and real consequences !!!
The sky is not falling. Win 11 color management is much improved over Win 10, moving from an sRGB-centric orientation to better encompass wider gamut and hdr environments. Apple people please be kind. DxO PL plays nicely with Windows 11 and there are no real color management issues as far as I am aware. That includes the printing module when DxO PL manages colors. Set up your display properly with current drivers (then leave it alone!) and embed actual ICC profiles in image files whenever possible. A bonus for Win 11 users is that the Microsoft Photos app is now fully color managed and useful for photographers. My understanding is that multiple display management is better too, but I have no recent experience with this.
There were some early problems as app developers made the necessary changes to adapt to the new Win 11 color management system. Those issues have been mostly resolved as far as I can tell. The reddit post is mostly old news. There is an active ongoing discussion among gamers on how best to use advanced color management features in Win 11 to add some extra hot sauce to their displays. Not at issue here.
