Pages: [1]   Go Down

Author Topic: Printing and RGB Values - Anyone clip pure white and black by 3 or 5%  (Read 4452 times)

Brad P

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 258

I've seen conflicting views on whether one should set up images for print with artificial constraints on pure black or pure white.  I think this makes sense because detail in extreme shadows and highlights is very hard to discern in print but am interested in what people think is the best practice.  What's your view?  Work with pure white as RGB 255, 242 (5%) or 247 (3%) or black as RGB 0, 13 or 8?

It certainly will matter what the paper choice is.  In my case it's both Canson Platine (a baryta) and Rag Photographique (a bright matte).
Logged

hugowolf

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1001
Re: Printing and RGB Values - Anyone clip pure white and black by 3 or 5%
« Reply #1 on: December 05, 2015, 10:38:33 pm »

The only point I could see for shifting white would be gloss differential on gloss and lustre papers. Otherwise, that is what a well made profile is meant to deal with.

Brian A

Logged

pfigen

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 534
    • http://www.peterfigen.com
Re: Printing and RGB Values - Anyone clip pure white and black by 3 or 5%
« Reply #2 on: December 05, 2015, 11:16:53 pm »

I don't see any point in doing this. I can see a visible, if slight difference in tone between printing a 255 white vs a 253 white on an Epson 9900 that I've custom profiled. So if you knock your whites down that far, you're also leaving a lot of dynamic range not he table. The shadow end is a slightly different story, and while the profiles are pretty good at maximizing the shadow detail, you will still lose pretty much everything that is below about 10 or 12 in your original file - and I'm referencing high quality photo paper using PK inks. Matte papers lose a bit more of the shadow details. Just one more thing to be aware of when printing.
Logged

howardm

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1984
Re: Printing and RGB Values - Anyone clip pure white and black by 3 or 5%
« Reply #3 on: December 05, 2015, 11:37:04 pm »

Pretty much the same for me.  I like to see a bit of differentiation between paper white and print white unless its really not meant to be.  I can get the blacks down to about 8 before they blend but I dont like to just clip.  I will often just add a 3-6 bump down near the bottom but black is still pretty much black.  And a lot depends on the image and mood.

Brad P

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 258
Re: Printing and RGB Values - Anyone clip pure white and black by 3 or 5%
« Reply #4 on: December 06, 2015, 02:37:25 am »

Thinking about this more, it does seem that if you have a good color proofing mechanism in place, all you would be doing by clipping is throwing away data.

Even for pure black, if when editing the screen's contrast and black point is set to match the paper, it would seem to be throwing away data on the low end too.

The never ending amount of nonsense to read out there! It may make sense though for people who don't have the set up or time to squeeze out all the info in the files. 

Logged

howardm

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1984
Re: Printing and RGB Values - Anyone clip pure white and black by 3 or 5%
« Reply #5 on: December 06, 2015, 09:04:46 am »

That's a big 'if'.  The lowest blacks are the most problematic between printer overinking, visual flare and surround brightness etc etc.  I will print out the Outback image or the shadow checker I have and compare it to the on-screen in softproof mode (be it in Photoshop or the NEC's softproof mode (which is awesome)).  Then tweak the NEC as needed to get closer to what is actually on-paper and save that new setting for that paper.

I know that my printer (3880) and most common paper mostly crushes blacks below 6-8 and not much I can do about that w/o any form of linearization.  So I goose the very low blacks a bit via curves layer so there is some form of differentiation but still a good black.  And then there is the artistic interpretation and goal.

If someone w/ more experience can tell me a better/different way, I'm all ears.  It can be a bit frustrating at times.

Wayne Fox

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4237
    • waynefox.com
Re: Printing and RGB Values - Anyone clip pure white and black by 3 or 5%
« Reply #6 on: December 06, 2015, 11:26:15 pm »

The only point I could see for shifting white would be gloss differential on gloss and lustre papers. Otherwise, that is what a well made profile is meant to deal with.

Brian A
+1 (and gloss differential to me is a pretty rare problem these days)


Thinking about this more, it does seem that if you have a good color proofing mechanism in place, all you would be doing by clipping is throwing away data.

If you push the black point up with a curve or even a levels adjustment, you aren't clipping things, as everything moves.


I know that my printer (3880) and most common paper mostly crushes blacks below 6-8 and not much I can do about that w/o any form of linearization.  So I goose the very low blacks a bit via curves layer so there is some form of differentiation but still a good black.  And then there is the artistic interpretation and goal.

I use Lightroom or a raw adjustment layer in photoshop and try to open the shadows up just a little, either globally or in specific areas with an adjustment brush.  pulling those dark tones apart seems to work pretty good to see a little bit of detail, but I still get the blackest black possible in areas where I want.
Logged

Brian Gilkes

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 443
    • http://www.briangilkes.com.au
Re: Printing and RGB Values - Anyone clip pure white and black by 3 or 5%
« Reply #7 on: December 08, 2015, 07:25:53 am »

A good profile will give close to linear response form 0-255. The trick is to have perceptual rendering matching average human luminosity response.
Logged

William Chitham

  • Full Member
  • ***
  • Offline Offline
  • Posts: 170
Re: Printing and RGB Values - Anyone clip pure white and black by 3 or 5%
« Reply #9 on: December 08, 2015, 12:36:48 pm »

Have we go the terminology right here? Surely the problem is that the printer is "clipping" whites and blacks, the solution is to ensure that all the detail you want to retain lies within the range of tones the printer can produce.

William.
Logged

Ernst Dinkla

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4005
Re: Printing and RGB Values - Anyone clip pure white and black by 3 or 5%
« Reply #10 on: December 08, 2015, 03:25:45 pm »

The Economy Mode set gloss enhancer of the Z3200 is only applied on the image area, with some papers the 255 whites then can even run up to the edge of the image and the gloss enhancer separates the image whites enough from the margin white. In Qimage I have to apply a print filter that has the red shifted to 254 but the GB can be 255 to get the gloss enhancer on the whites, no color ink is printed though in the whites. On matte papers at least for images where the image white runs up to the white margins you have to keep some slack in the image whites, it can not be an RGB 255 white, the image falls apart, usually in the sky. Of course a light grey border can be made to solve it in another way.


Met vriendelijke groet, Ernst

http://www.pigment-print.com/spectralplots/spectrumviz_1.htm
December 2014 update, 700+ inkjet media white spectral plots
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: Printing and RGB Values - Anyone clip pure white and black by 3 or 5%
« Reply #11 on: December 08, 2015, 04:02:25 pm »

A good profile will give close to linear response form 0-255. The trick is to have perceptual rendering matching average human luminosity response.
I think you mean Brightness (which is a perceptual, visual response)? Luminosity is a measure of the total radiant energy from a body. It has nothing to do with what a human observer perceives.


I agree with others, soft proof with good profiles, let the profiles deal with conversions and all should be well. Humans need to view images in context and make subjective decisions about how they want  the image to appear/print.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

Wayne Fox

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4237
    • waynefox.com
Re: Printing and RGB Values - Anyone clip pure white and black by 3 or 5%
« Reply #12 on: December 09, 2015, 02:54:41 pm »

Have we go the terminology right here? Surely the problem is that the printer is "clipping" whites and blacks, the solution is to ensure that all the detail you want to retain lies within the range of tones the printer can produce.

William.
The printer doesn’t clip colors, and you really don’t have to make a special effort to keep the range of tones within the colors the printer can reproduce ...this is all handled by a good color management workflow.

Since most images have colors outside the gamut of the printer, the conversion to the printer profile handles keeping those colors within the printer capability.  How you choose to handle the colors that are out of the printer gamut determines if they get “clipped” into the nearest  color the printer can produce (relative), or if the conversion can adjust all colors so it can try to retain the relationship of the colors to each other (perceptual).
Logged

JRSmit

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 922
    • Jan R. Smit Fine Art Printing Specialist
Re: Printing and RGB Values - Anyone clip pure white and black by 3 or 5%
« Reply #13 on: December 10, 2015, 11:18:05 am »

A good profile will give close to linear response form 0-255. The trick is to have perceptual rendering matching average human luminosity response.
A good profile indeed does that.
But the contrast range of paper is much lower than that of an average display, which is your visual reference.
To cater for that one choice is blackpoint compensation, other choice is to open up somehow the vey dark areas so that on the print you still see the details of interest, while still have the aximum black on the pritn where needed.
In Jeff Schewe's book on printing he clearly describes workflows to match those contrasts.
Logged
Fine art photography: janrsmit.com
Fine Art Printing Specialist: www.fineartprintingspecialist.nl


Jan R. Smit

AFairley

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1486
Re: Printing and RGB Values - Anyone clip pure white and black by 3 or 5%
« Reply #14 on: December 10, 2015, 12:31:16 pm »

I tested this a while ago and found I could knock whites down to 250 or so and see the difference in brightness of the area on the print but still get the gloss differential effect I saw if that area was left at 255.  That's with Exhibition Fiber on a 3880.  So my take is that it's not worth it to kill your top whites for little benefit, and that to lay down enough ink to completely eliminate any gloss differential you have to wreck your whites in the print.
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: Printing and RGB Values - Anyone clip pure white and black by 3 or 5%
« Reply #15 on: December 10, 2015, 12:32:07 pm »

But the contrast range of paper is much lower than that of an average display, which is your visual reference.
Which is why my SpectraView can control black and thus contrast ratio. Easy to set it for 300:1 for a glossy paper, say 200:1 for matt etc.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

datro

  • Full Member
  • ***
  • Offline Offline
  • Posts: 231
Re: Printing and RGB Values - Anyone clip pure white and black by 3 or 5%
« Reply #16 on: December 10, 2015, 02:07:22 pm »

Which is why my SpectraView can control black and thus contrast ratio. Easy to set it for 300:1 for a glossy paper, say 200:1 for matt etc.

+1
I've found that lowering the contrast ratio in SpectraView on my NEC display is the single most important thing to do to get a better environment for working up images to be printed.  I'm running at 350:1 but have considered going down to 300:1 or even 250:1.  I print on both matte and glossy papers.

However there is one caveat which I've lately been thinking about:  Lowering the contrast ratio on the monitor is good for evaluating photographic images which are going to be printed, but not so good for my (aging) eyes when it comes to reading text on web sites and in documents.  With text you want a fair amount of contrast, and these days with many web sites and applications using grey or light-colored text (instead of black) it is even more of an issue.  So lowering the contrast ratio ends up being a bit of a compromise if you do a lot of reading on the display as well.

Dave
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: Printing and RGB Values - Anyone clip pure white and black by 3 or 5%
« Reply #17 on: December 10, 2015, 02:11:48 pm »

I've found that lowering the contrast ratio in SpectraView on my NEC display is the single most important thing to do to get a better environment for working up images to be printed.  I'm running at 350:1 but have considered going down to 300:1 or even 250:1.  I print on both matte and glossy papers.
However there is one caveat which I've lately been thinking about:  Lowering the contrast ratio on the monitor is good for evaluating photographic images which are going to be printed, but not so good for my (aging) eyes when it comes to reading text on web sites and in documents.  With text you want a fair amount of contrast, and these days with many web sites and applications using grey or light-colored text (instead of black) it is even more of an issue.  So lowering the contrast ratio ends up being a bit of a compromise if you do a lot of reading on the display as well.
What makes SpectraView so awesome is you can build as many calibration targets as you wish, switch on the fly. So make one for 300:1, another for 250:1, yet a 3rd for viewing the web or just emulate sRGB. Alter white point, cd/m^2 etc, along with contrast ratio. You can have all three and more.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

howardm

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1984
Re: Printing and RGB Values - Anyone clip pure white and black by 3 or 5%
« Reply #18 on: December 10, 2015, 02:24:46 pm »

Dave, use the Multiprofiler software to change the luminance and black point on the fly w/o having to load a different profile.
Pages: [1]   Go Up