Export from ProPhoto RGB to sRGB IEC61966-2.1?

[Peter_DL]

Folks that use wide gamut monitors have to be extremely careful when converting to sRGB as there can be very big color shifts. One work around is to just use the proof view settings using sRGB as the target. Then, WYSIWYG when converting to sRGB.

Yes, that’s good advice, to softproof from ppRGB to sRGB before doing the conversion.
Sometimes just minor edits of the out-of-gamut colors e.g. of its hue and lightness can indeed contribute to improve the conversion.

On a case by case basis it can also be of help to insert a Lut-type profile: ppRGB -> … -> sRGB. Rather than v4 sRGB (which I can’t get to work), PhotoGamutRGB is in my toolbox.

Further, with saturated flowers, one key step is often to recover the details which got affected and blurred by the conversion by means of a local contrast enhancement.

Another option, not very famous around here (but I think it should be mentioned to the OP), is when shooting Raw + sRGB Jpeg to have a look at the sRGB Jpeg as well. It can at times be easier to edit it to a suitable rendition in sRGB for sending it to the print lab.

--

[earlybird]

You guys are scaring me... maybe I should have stayed with my sRGB Photoshop workflow. :-)

[earlybird]

Thank you for the explanation about the use of rendering intents.

I am trying to condense the suggestions made in a few of the posts; I think what has been explained is that it is a practical concern when converting from a RGB color space to something like a CMYK color space, and if the print service I use requests sRGB files it is the print service provider's choice of rendering intent, made when converting the sRGB transport file into an actual printing profile color space, that has the most influence on the final results.

Does that seem correct?

I am reluctant to seem pedantic about this, but I hoping someone more comfortable with the subject can confirm or correct; does this seem correct?

Thank you.

[Doug Gray]

Using a printing service that requires sRGB generally means the prints will be made using perceptual intent on whatever device the service uses. These vary. Ultimately all ink printing is done with CYMK.

While you can read all about inkjet printers with gamuts that exceed sRGB and even Adobe RGB what you find little discussion of is that printers can't print significant portions of plain old sRGB. A printer may well be able to print certain colors well beyond sRGB but the gamut volume of colors within sRGB that can't be printed typically exceeds the gamut volume of colors that are printable but outside of sRGB.

For Adobe RGB things are even more lopsided. The number of colors within Adobe RGB that can't be printed far exceeds the number of colors outside Adobe RGB that can be printed.

The reason to use larger gamut working spaces than Adobe RGB is to capture that small percentage of colors that are printable but outside Adobe RGB, should they be in your image.

This is why, even if you work strictly in sRGB it's a good idea to soft proof prior to printing and, of course, it's super critical in Adobe RGB or wider working space gamuts.

[Peter_DL]

... if the print service I use requests sRGB files it is the print service provider's choice of rendering intent, made when converting the sRGB transport file into an actual printing profile color space, that has the most influence on the final results.

Does that seem correct?

You are referring to photo prints, not inkjet, right ?

Even if the lab just mentions sRGB there can be different cases depending what they actually do in terms of color management. One likely scenario is that the lab operates their printer in a so-called sRGB-mode. See this earlier explanation here. In this case there is a kind of hard-wired transform from the RGB numbers for input to the print output, and the lab does not do any ICC-type gamut conversion ahead of it at all. This can be verified by providing them an image in ProPhotoRGB (=> bad print, similar to assign sRGB).

So unless you would have a somewhat accurate profile of the de facto printed gamut, you’ll have to provide them sRGB, and probably the only way to influence the print is by choosing Auto-corrections On or Off. Most labs but not all are offering this choice at least.

I recommend to try both, Auto-corrections On and Off. Auto-corrections may have a bad reputation but in fact there can be quite some "image intelligence" behind it. With Auto-corrections Off, your image might need a second round of editing before providing it to the lab, and it might require a learning curve how to outperform the Auto-On option.

Anyway, another important aspect is the selection of the print lab as such as well as of the print paper. Some labs use Fuji paper, others Kodak. Some labs are simply doing a better job than others. Try different labs, look for reviews in photo magazines, etc.

--

[Doug Gray]

You are referring to photo prints, not inkjet, right ?

Even if the lab just mentions sRGB there can be different cases depending what they actually do in terms of color management. One likely scenario is that the lab operates their printer in a so-called sRGB-mode. See this earlier explanation here. In this case there is a kind of hard-wired transform from the RGB numbers for input to the print output, and the lab does not do any ICC-type gamut conversion ahead of it at all. This can be verified by providing them an image in ProPhotoRGB (=> bad print, similar to assign sRGB).

So unless you would have a somewhat accurate profile of the de facto printed gamut, you’ll have to provide them sRGB, and probably the only way to influence the print is by choosing Auto-corrections On or Off. Most labs but not all are offering this choice at least.

I recommend to try both, Auto-corrections On and Off. Auto-corrections may have a bad reputation but in fact there can be quite some "image intelligence" behind it. With Auto-corrections Off, your image might need a second round of editing before providing it to the lab, and it might require a learning curve how to outperform the Auto-On option.

Anyway, another important aspect is the selection of the print lab as such as well as of the print paper. Some labs use Fuji paper, others Kodak. Some labs are simply doing a better job than others. Try different labs, look for reviews in photo magazines, etc.

--

Good stuff Peter.

Yep, once you get into "smart," content aware processes things get very interesting. You can get very attractive prints but it is hard, if not impossible, to know what they will look like in advance and soft proofing is not going to help much.

It occurs to me that the manufacturers of these machines have the ability in-house to do the content aware magic. I wonder if some sort of plugin might become available for photographers to "soft-proof" using their content aware algorithms? Probably won't happen since the market is probably fairly limited and people have enough difficulty using static profiles.

[tho_mas]

There's a pretty handy tool to "convert" almost any icc profile into a real "printer" profile (containing perceptual, colormetric and saturation tables):

basICColor dropRGB ->

You can even create a profile with the unmatched basICColor gamut mapping from another profile, e.g. a canned profile that came with your printer, just drag and drop the profile onto the basICColor dropRGB icon – done!

Although the software is obviously intended to create printer profiles based on measured printer targets and to optimize existing printer profiles you can also use this tool to easily create table based versions of sRGB, AdobeRGB, ECI-RGB, ProPhoto RGB... and even of input icc profiles.
The software does not provide "settings" ... on the other hand you don't need any when you create a "printer"-profile from a matrix based profile like sRGB. Of course the grey axis of the source profile (Gamma 1.8, Gamma 2.2, L-Star or whatever...) will be preserved in the resulting LUT based profile.
The table based sRGB profile I've created with "dropRGB" works extremely well on real world images. It shows some banding when converting super high saturated blues of synthetically computed images in very large working spaces (like some printer test images in ProPhotp RGB)... but even this is easily fixed with a little adjustment in Photoshop. And, again, with real world images this is a non-issue.

[earlybird]

You are referring to photo prints, not inkjet, right ?

Yes, I order C-prints with a "no corrections" option and send the files in a sRGB colorspace.