Soft proofing papers with OBAs

[coellis]

Hi All,

When I softproof papers with OBAs and there is a lot of white in the photo then the bluish colour cast is really strong. Usually I softproof with Perceptual rendering intent with simulate paper color checked. The only way to get rid of the colour cast is to uncheck simulate paper colour. So what is the best way to softproof papers with OBAs like the Epson premium luster?

Many thanks
Colin

[digitaldog]

Hi All,

When I softproof papers with OBAs and there is a lot of white in the photo then the bluish colour cast is really strong. Usually I softproof with Perceptual rendering intent with simulate paper color checked. The only way to get rid of the colour cast is to uncheck simulate paper colour. So what is the best way to softproof papers with OBAs like the Epson premium luster?

[a href=\"index.php?act=findpost&pid=174250\"][{POST_SNAPBACK}][/a]

I'd expect to see the cast with an Absolute colorimetric intent but not necessarily with the others. Where did the profile come from? Often, the "graying down" of the paper white tends to appear cooler, I'd agree, but blue?

[coellis]

Thanks Andrew.

I've attached two screenshots of a photoshop file softproofed with renderimg intent perceptual and simulate paper color on and off. In the off the white next to the photo is white and in the on is bluish-white. The profile is the canned Epson 2400 for premium luster.

[attachment=5112:attachment]

[attachment=5113:attachment]

[digitaldog]

Doesn't look all that off based on what I'd expect from having the Paper White on. Notice it doesn't make a huge effect on the image itself. It could also be how you have the display white point calibrated.

Also, give your eyes time to adapt to the paper white simulation. One issue is, watching it redraw before your eyes (look away while using a key command to turn it on and off) and note that you really need to be in full screen mode where NO other UI elements that don't undergo the simulation are in your field of view.

[coellis]

It could also be how you have the display white point calibrated.

Andrew, what do you mean by that?

[digitaldog]

Andrew, what do you mean by that?
[a href=\"index.php?act=findpost&pid=174310\"][{POST_SNAPBACK}][/a]

The white point of your display profile.

[coellis]

It's set to D65.

[Schewe]

I've attached two screenshots of a photoshop file softproofed with renderimg intent perceptual and simulate paper color on and off. In the off the white next to the photo is white and in the on is bluish-white.
[a href=\"index.php?act=findpost&pid=174296\"][{POST_SNAPBACK}][/a]

Yep, as expected and as designed. The proble is, you really shouldn't be looking at ANY monitor white when you are looking at the softproofed image. Why? Because the design of using the Display Options is to simulate paper white (which WILL be darker and often bluer if the paper has whiteners) and the blacks will be less black. That's REAL based on what the actual print can provide in terms of dynamic range.

Ideally, until you know how to force your eyes to adapt mentally, hit the F key to get to full screen black and the Tab key to hide any UI. Then set up your print viewing and your softproof. With accurate display and output profiles, it's scary how accurate Photoshop's softproof can be.

Yes, it WILL make your image, on screen look like crap–that's what it's designed to do...drop the white down to that of the paper and lighten the black to match that of the ink. By comparison to the RGB image, unsoftproofed on the display, it is SUPPOSED to look like crap. But, it will be an accurate and effective prediction of what the image will look like in ink, on paper.

[coellis]

Thanks very much Jeff. I know you're right but then how come when I print the softproofed image, the white WITHIN the image doesn't print bluish-white but white?

What I'm trying to say is when there is a white area WITHIN the image that white area in the softproof looks bluish-white but when printed without corrections is white (paper white, no colour cast).

Sometimes I print for others and have to send them my profiles. How can I explain to them not to correct the bluish-white when they softproof the image?

Again many thanks.

[digitaldog]

Thanks very much Jeff. I know you're right but then how come when I print the softproofed image, the white WITHIN the image doesn't print bluish-white but white?

That's somewhat a display technology issue. We don't really have good control (or any control) over the dynamic range of the display. You have manufacturers trying to sell us on units that have 3000:1 contrast ratio's and you want to view a print simulation at 200:1 so now Photoshop and the current ICC architecture has to attempt to do this for you. You'll notice it's not difficult to take black and make it milky appearing to simulate the ink density. But white? Not easy.

[Geoff Wittig]

Thanks very much Jeff. I know you're right but then how come when I print the softproofed image, the white WITHIN the image doesn't print bluish-white but white?

What I'm trying to say is when there is a white area WITHIN the image that white area in the softproof looks bluish-white but when printed without corrections is white (paper white, no colour cast).

Sometimes I print for others and have to send them my profiles. How can I explain to them not to correct the bluish-white when they softproof the image?

Again many thanks.
[a href=\"index.php?act=findpost&pid=174476\"][{POST_SNAPBACK}][/a]

All "white" is relative. Seen in isolation, the whites in your image printed on (e.g.) Epson premium luster will look, well, white. However, if you put that print with its white border against the same image printed on, say, Epson ultrasmooth fine art, the border on the (OBA-free) fine art paper will look quite warm or cream-colored compared to that on luster paper. Or seen the other way, the border on the luster paper will look noticeably blue when placed against the ultrasmooth. And of course any paper white (0,0,0) within the image will reflect that paper base color. Photoshop's soft proof really does quite accurately show you what you'll get; it's up to us to interpret that correctly and see if any image adjustment is necessary for the print to look like we want it to.

[coellis]

Photoshop's soft proof really does quite accurately show you what you'll get; it's up to us to interpret that correctly and see if any image adjustment is necessary for the print to look like we want it to.

I'm really sorry but I'll have to disagree with you Geoff. I don't think Photoshop does a very good job. It does not show me what I'll get and I see this as a limitation of Photoshop or as Andrew said a limitation of technology. Photoshop through softproof shows a colour cast on screen (bluish-white for papers with OBAs, creamy-white for fine art papers  without OBAs), that does not show on the actual print. On the other hand while Profilemaker takes care of OBAs when profiling it should somehow hide the colour cast in the soft proof table. As I said before sometimes I print for others and I have to send them my profiles to softproof themselves. What am I going to say to them when they ask me about the bluish-white cast?

Best
Colin

[digitaldog]

On the other hand while Profilemaker takes care of OBAs when profiling it should somehow hide the colour cast in the soft proof table. As I said before sometimes I print for others and I have to send them my profiles to softproof themselves. What am I going to say to them when they ask me about the bluish-white cast?

[a href=\"index.php?act=findpost&pid=174653\"][{POST_SNAPBACK}][/a]

Here's a case where some careful profile editing of JUST the soft proof table may be in order. You probably could counter the blue appearing "tint" but I suspect that, depending on the white point of the display, what other UI elements are showing and so forth, its still going to look fudgly. What we really need are displays that have actual control over the dynamic range (something ICC profiles can't do). The old Sony Artisan had such an option thanks to its ability to calibrate and control both black and white extremes. I'd also think that if you lowered the overall luminance like we used to use on CRTs, that might help (but present issues in viewing the prints and getting that luminance in sync with the display).