Pages: [1]   Go Down

Author Topic: Soft Proofing Failure with M1 (and sometimes M0) with high OBA Paper  (Read 1648 times)

Doug Gray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2197

There's an interesting problem that comes up when softproofing for these papers. I've noticed it on some matte and glossy papers that have high OBAs. It can't be fixed but fortunately only applies when "show paper white" is selected. It affects both Perceptual Intent and Relative Intent and is most likely to come up when you are soft proofing against a print in a viewing booth using the currently recommended D50 including D50 uV. That is, of course, exactly when you are most likely to select "show paper white."

It is particularly bad when you are trying to soft proof a high key image.

As an example I used Epson Ultra Premium Glossy which has a fairly high level of OBA. The M1 white point of the paper sheets I have is Lab (97.3 1.8 -9.2). which is pretty bluish.

The problem exists because the uV kicks up the B in RGB values beyond 255 and these are not showable on a monitor. For this paper, it's (244,246,265) on both standard and wide gamut monitors. But what gets shown is (244,246,255) which is about half as bluish. This drops the b* from -9 to under -5. But only on pure white.. When L* drops to 94, blue channel stays unchanged because it is already clipped and the b* climbs to -9. The result is a strange tint shift in the whitest areas that becomes correct only when the L* drops below about 93.

If you are using soft proofing and have these high key areas, expect the soft proof to fail. The only solution is to unselect "show paper white."  Of course this will not show correct colors matching a light booth that has D50 w D50 uV levels but will at least not produce shifting color in these ranges.

A few notes:
The lower the white point L* value, the less the problem. It's the papers with a high L* together with OBAs that clip the blue channel.

For instance a white point L* of 94 won't cause any issue if the b* of the paper white is below (in magnitude)  -10 because there is enough headroom on the blue channel.

I've included a few gradients that show the effect with Epson Ultra Premium Glossy. The image "Clip", when viewed in Photoshop shows what the "Neutral" looks like when soft proofed. In particular the gradient quickly becomes more bluish than the surrounding white starting to the left. This can also bee seen by j

The image "NoClip" is what the soft proof would look like if the blue channel wasn't clipped at 255. It is very slightly reduced in luminance to prevent the clipping. The overall image has the strong blue tint of

If you have M1 Profiles with similar high L* and OBA papers you can see the same effect soft proofing the "Neutral" image
Logged

arobinson7547

  • Full Member
  • ***
  • Offline Offline
  • Posts: 187
Re: Soft Proofing Failure with M1 (and sometimes M0) with high OBA Paper
« Reply #1 on: October 18, 2017, 04:41:28 pm »

I stopped keeping up but when the new Gracol 2013 standard came out, proofing was switched for Papers 'without' OBAs to Papers 'With' OBAs. Measuring Devices are needing to be replaced with devices that were capable of M! "AND" and new light standard was indroduced (new tubes). I don't know much about the new lights for viewing.

Since you have the M1 and OBA paper, maybe you need [to try] the new lighing
Logged

Doug Gray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2197
Re: Soft Proofing Failure with M1 (and sometimes M0) with high OBA Paper
« Reply #2 on: October 18, 2017, 05:39:24 pm »

I stopped keeping up but when the new Gracol 2013 standard came out, proofing was switched for Papers 'without' OBAs to Papers 'With' OBAs. Measuring Devices are needing to be replaced with devices that were capable of M! "AND" and new light standard was indroduced (new tubes). I don't know much about the new lights for viewing.

Since you have the M1 and OBA paper, maybe you need [to try] the new lighing

Not a lighting issue. The soft proofing failure is really a limitation of the colorspace RGB B values being clipped at 255. It's intrinsic and not something that can be fixed with profiles. A workaround is to set the monitor's whitepoint to match the luminance and chromaticity of the D50 (wiith uV) viewing booth. Then don't select show paper white.

If one can easily switch between monitor white points and associated profiles (such as with ColorNavigator inter alia) then the workaround fixes the problem. It's also particularly effective when the viewing station uses lighting that doesn't have the required levels of uV. That's most of them made in the last 20 years. Full D50 w uV booths are pricey. Andrew Rodney's videos on soft proofing follows that approach.

I see proofing paper is now specified at Lab(95,1,-4) from (95,0,-2) in 2006. It's a pretty small change and not really large amounts of fluorescent additives. In any case these small levels have a tiny effect on hard proofing in that industry but there should be no issue with soft proofing as one can simulate  95,1,-4 white points without clipping the monitor's blue channel. Really, the only place I see the effect is soft proofing high key images with a white border for papers with extremely large levels of OBA combined with a high L* white point.
« Last Edit: October 18, 2017, 05:44:15 pm by Doug Gray »
Logged
Pages: [1]   Go Up