RGB to CMYK colour loss

[ZoeLione]

I reacently leanred the differance off RGB and CMYK. I was wondering whether it would be possible, once an image is converted to a CMYK can you 'artificially' regain the colour and the light lost when converting via adjustment layers or are there other things that will be lost after converting from RGB to CMYK?

[c.pfaffenbichler]

If data has been destroyed it is gone. 

Out of gamut colors in the RGB image can be »compressed« or »clipped« in the CMYK image – in neither case the CMYK image itself provides an unequivocal way of determining the original colors. 

What one can do, though, is maintain RGB-images as Smart Objects in CMYK Images (psd, psb, tif). 

But, in my experience, it is prudent to just work on and keep the layered RGB Images anyway and separate on final output from a layout application like Indesign. 

Perfectionists may want to further edit the separated images, but in that case one should keep the layered RGB and CMYK images. 

[D Fosse]

CMYK is a representation of actual physical inks on actual physical paper. Each CMYK profile relates to an offset press calibrated to a certain standard, using specific inks on specific paper stock.

The color gamut is determined by the purity of the inks. There's nothing Photoshop can do about that.

If this isn't for commercial offset print you should work in RGB. Inkjet printers work with RGB data.

[c.pfaffenbichler]

And there is also the issue of the Rendering Intent. 

If one were to convert to CMYK with the Intent »Absolute Colorimetric« the paper color would be included in the CMYK. 

[NB, colourmanagement net]

IN many cases the CMYK colour space is smaller than the RGFB colour space.

Ink on paper isn't capable of the colour range [gamut] of a decent well calibrated monitor display screen.

The RGB ICC to CMYK ICC conversion (using an ICC profile that truly represents the printing condition) will potentially reduce the image colour by remapping it to the CMYK "gamut volume". The choice of "rendering intent" decides how that is done.

Sometimes a Relative Colorimetric conversion will be better.

A CMYK file has specific max ink values, the paper can only take so much ink before sheets offset one to the next or maybe even ink runs..

If you adjust the CMYK file after conversion, maybe increasing saturation, its really easy to violate the ink limits, it might look good on your screen but on the press you could "overink" and have ink running down the paper. 

CMYK needs care

I hope this helps
neil barstow, colourmanagement net :: adobe forum volunteer:: Co-Author:Getting Colour Right
google me "neil barstow colourmanagement" for lots of free articles on colour management

[TheDigitalDog]

The short answer is no. And this can occur going RGB to RGB as well, again depending on the color gamut of the scene, the capture after rendering and the final output color space. We have to live with the fact that there are wider ranges of colors than we can display or print.

What you want to attempt while soft proofing in the wider gamut space (that you can see), is a 'harmonious' conversion using output specific edits when possible. Profiles and rendering intents don't do this, they are only a starting point in viewing an entire, simple conversion. Profiles and modern color management know nothing of color in context. They 'see' and treat an image one pixel at a time. As an image creator, you can see the image in context, again within the limitation of your display. You can pick a rendering intent you visually prefer first, then you can make output specific edits to better, if necessary, express your vision for that image.

Soft proof and select a RI you prefer.

Consider in Photoshop making layers that are output specific to better control the image through the soft proof.

Lable them with the profile name and RI, you can toggle them on or off so you in essence have the 'master' TIFF/PSD, and a layer group of the output specific edits.

Flatten convert to the output color space and Save a Copy that way for output AFTER you convert to the output color space.

[Conrad C]

---

@ZoeLione wrote:

…are there other things that will be lost after converting from RGB to CMYK?

---

Yes, here’s one example. RGB has no black, you get black by specifying 0 for all three channels. But that doesn’t work for print, because if you print 100% of C, M, and Y you don’t get a solid black. So the K (black) channel/plate is added to CMY, and that way you can print a nice dense solid black. But what parts of the RGB colors do you move into that black K channel? There are multiple “black generation” formulas depending on what combination of ink and substrate (paper) you’re printing on, and you’ll pick the black generation that is appropriate for your print job.

But now suppose you want to go the other way, from CMYK back to RGB. There’s no black channel in RGB, so how do you divide up the K channel and redistribute it into RGB so that you get the same RGB values you started with? You can’t, especially if you don’t know exactly which CMYK conversion was used. So that is one of the other things you lose if you try a round trip conversion; you can’t accurately reverse the black generation. The software can complete the CMYK to RGB conversion by guessing, but you won’t end up with exactly the same RGB values you started with.

[Mylenium]

I was wondering whether it would be possible, once an image is converted to a CMYK can you 'artificially' regain the colour and the light lost

Certainly not artificially or magically, but you could strategically add more color separations and use spot colors. Everything else the others already have explained.

Mylenium