White background prints beige in absolute colorimetric intent

[Andy Fraser]

There's probably a clear (and possibly logical!) explanation for this but I can't find it.  I have a test image with color patches with different densities of C, M, Y, K.  The initial background of the file is white and is left that way.  If I flatten and check with the dropper, it shows 0 for each of C, M, Y, K in that open area so when it prints there should be no ink put down in that area, just the natural paper color.

However, the printable area is filled with beige and perhaps this beige base color has also been added to my color patches also.

The image is CMYK mode, color patches in 25%, 50%, 75%, 100% each of C, M, Y, K (16 little rectangles).

The RGB working space in PS is Adobe RGB.

The printer is Epson P800, Epson premium photo paper glossy, print quality maximum, color mode (management) off.

Photoshop print settings are Photoshop manages colours, printer profile is the Epson profile for the premium photo glossy, normal printing mode, black point compensation is off, rendering intent is Absolute Colorimetric since I read that should put exactly the inks and densities specified by the colour patches in the image onto the paper.

Why would the background be beige, and what rendering intent would result in a white background and exactly those densities of specific CMYK inks being printed?

Relative and Perceptual don't print with this beige background fill. 

In addition to the puzzle of why Absolute adds a background fill, would Relative and Perceptual prints have exactly the density and inks specified by the image, since the densities are given in percentage and therefore can't exceed the gamut of the inks and paper so the colors in-gamut, will be correct?

Thanks

Andy

[Andy Fraser]

Adding to this, I have found D Fosse's post on "Is it possible to use Absolute rending intent without the simulated paper color?" that Absolute is different in that it does not remap the white point so a color cast can appear, so maybe the problem is something is trying make the white paper look white by adding beige to it...

[Stephen Marsh]

Use relative with BPC on.

[Andy Fraser]

Thanks Stephen

For this test though, the input image is assured to be within the gamut of the printer (output space) since it was this printer that had created it, so BPC should be off based on the explanation at https://support.captureone.com/hc/en-us/articles/360002654477-Black-point-compensation

The remapping might make some of the colours inaccurate.  I'll try with and without BPC anyway since it doesn't require much extra printing or work, to see what difference it makes.

Andy

[D Fosse · Accepted solution]

Yes, indeed, this is expected. You should normally use Relative Colorimetric to get correct white point remapping. Absolute will not remap, and is for different purposes.

[Andy Fraser]

Thanks for the replies!  I'll use Relative and that should result in the print being as identical as feasible to the print I scanned into Photoshop.

As for what I'm up to, I suspect one or more of my inks are inaccurate due to being two or three years old so I'm printing the test patch, scan back in and print that scan, repeat the loop two to four times, and see how much the colour patches degrade (an error will compound on each loop).  No scientific benchmark for how much they shouldn't degrade but it will be interesting to do the test and compare after I've install a full set of new inks.