Luminous Landscape Forum
Raw & Post Processing, Printing => Digital Image Processing => Topic started by: bill t. on April 29, 2013, 10:47:22 pm
-
The dirty little secret of soft proofing is that monitors have limitations, yet we like to think they don't.
Here are some slapdash iccview.de graphs.
The wireframe is my 2690WUXi monitor profile.
The colored hull is a moderately high gamut glossy canvas, which is in the same ballpark as most barytas .
Look at all the important colors that my poor monitor can't adequately represent! And I have seen the results of this effect with my own eyes.
In soft proofing, even a pretty good monitor can't go a few places where the media itself can! Explains a lot of things I've been noticing recently, such as how low gamut matte media soft proofs more accurately than high gamut glossy media. You can make your own conclusions, but this is definitely food for thought. Maybe my theory is wrong, but it seems supported by my empirical observations.
All I'm asking here is a monitor that can cover 100% ProPhotoRGB, what's the holdup? Or at least represent every color the human eye can see. Andrew knows the name of that concept, which I can't remember.
-
The problem is less it being possible....it is how much it would cost. ;D
-
All I'm asking here is a monitor that can cover 100% ProPhotoRGB, what's the holdup?
ProPhoto RGB has two primaries that are not physically realizable. The green one is only a little way outside the xy or uv horseshoe, but the blue one is wildly outside. So, as long as monitors have three primaries, and light mixes additively, we'll never see a monitor that can cover the whole Pro Photo RGB gamut, or even that part of it that's visible to humans.
Sad, huh?
Jim
-
Last year, I looked at what you could do to build a really wide-gamut monitor with three primaries (http://blog.kasson.com/?p=1880); the challenges are daunting, even with spectral primaries.
Adding a fourth primary helps. (http://blog.kasson.com/?p=1887)
Jim
-
In soft proofing, even a pretty good monitor can't go a few places where the media itself can! Explains a lot of things I've been noticing recently, such as how low gamut matte media soft proofs more accurately than high gamut glossy media. You can make your own conclusions, but this is definitely food for thought. Maybe my theory is wrong, but it seems supported by my empirical observations.
Your theory is bang-on. Take a look here (http://blog.kasson.com/?p=1872) for similar conclusions.
Jim
-
Andrew knows the name of that concept, which I can't remember.
Not sure what Andrew calls it, but I call it "unobtainiaum (http://en.wikipedia.org/wiki/Unobtainium)" which is anything that is, well unattainable...
"In engineering, fiction, and thought experiments, unobtainium is any fictional, extremely rare, costly, or impossible material, or (less commonly) device needed to fulfill a given design for a given application."
Will displays get better? Yes...however, unfortunately, computer display technology is actually driven by consumer TV technology, so we are at the whim of a different, related but not directly connected tech.
The trick is to learn how to adapt and adopt what we have now, to do what we want to do now and learn how to deal with it's limitations...
-
Jim, I like your enthusiasm...now, we just need to connect you to some of the guys who move the display tech forward like Karl Lang and Will Hollingsworth. Karl designed the original Colormatch display system for radius and Will works for NEC Displays. Both are LuLa members (although they don't post a lot).
-
Hey, but wasn't my old LaCie Electron Blue better in this regard? Until if faded to under 80 cdm2's?
A couple years ago there were LCD TV's down at BestBuy with a yellow patch in their RGBY matrix. Sure had bright yellows and greens. Doesn't seem to have driven much into our little world, however.
OK, I'll shut up and adapt. But I've lost all respect for this monitor.
-
But I've lost all respect for this monitor.
Well, that's one way of dealing with disappointment, the other is to learn how to use what you have to the best of your ability!
-
OK, will try not try to wallow in a lifetime of Gamut Regret.
PS needs one of those LR style monitor gamut buttons.
-
PS needs one of those LR style monitor gamut buttons.
Naw, it'll only be yet another fixation that will distract photographers. Just shoot raw, use Pro Photo RGB, soft proof with good prifiles in LR/PS and make great prints...pretty easy, actually.
-
The problem is less it being possible....it is how much it would cost. ;D
Doesn't matter! ProPhoto is a theoretically constructed color space (like all the RGB working spaces). All you need are some chromaticity values to plot a gamut and even if one or two don't fall within human vision, no problem. Well it's a problem if you want to design a real world device to produce something we can't see (this ProPhoto display). IOW, ProPhoto defines "colors" we can't see (so I'd submit, they are not colors).
You'll never see a ProPhoto RGB display unless the time comes those super fetus aliens from 2001 a space odyssey up our evolutionary powers such we can see outside the current limit of our visual systems.
-
The dirty little secret of soft proofing is that monitors have limitations, yet we like to think they don't.
Only those who are trying to sell something, or don't fully understand the differences in gamuts in devices and their inherent limitations.
You know, there are colors we can see a digital camera can't capture while there are 'colors' it can capture we can't see (although much is done to reduce these). There are colors in working space we can't print or display, there are colors we can display we can't print. No dirty little secret, but often a revelation for those starting to understand the limits of digital color imaging!
Breaking news: We don't live in a prefect world.
-
ProPhoto is a theoretically constructed color space (like all the RGB working spaces). All you need are some chromaticity values to plot a gamut and even if one or two don't fall within human vision, no problem. Well it's a problem if you want to design a real world device to produce something we can't see (this ProPhoto display). IOW, ProPhoto defines "colors" we can't see (so I'd submit, they are not colors).
Ahh, those fun-loving guys at Kodak. They're the same people that brought us PhotoYcc, which was based on an RGB color space with physically-realizable primaries, but allowed negative amounts of them. Try that on your display!
Jim
-
And a more current monitor will not help either....
(NEC PA301w vs Canson Baryta)
cheers
afx
-
Ahh, those fun-loving guys at Kodak. They're the same people that brought us PhotoYcc, which was based on an RGB color space with physically-realizable primaries, but allowed negative amounts of them. Try that on your display!
Couple points. First, last time I looked, Lab which has a strange following also had negative numbers.
Have you ever seen what PhotoYCC in say Photoshop looks like (without the proper transforms)? Acid trip.
PhotoYCC was in a large way, the predecessor of the raw data so many of us today cherish. It was way ahead of it's time.
I'll let other's argue if "PhotoYcc, which was based on an RGB color space" is valid.
We're talking numbers here which have no useful definition without a scale. That scale tells us the gamut among other things. The numbers can be CMYK, Lab or RGB and all could be out of display gamut. Life's a bitch. If your only interest is the output to a display, that's the only gamut you need to be concerned with. I'd shutter to think that in 10 years, the gamut of output devices the masses are using isn't wider than today. This is a moving target. We've seen this in relatively few years with both inkjet and traditional 4 color CMYK output.
WYS isn't always WYG but we strive to get as close as we can with the limitations of technology and a CMS and it's somewhat archaic basis and architecture.
If you want to dumb down the gamut of your output to match your display, that's pretty darn easy to do.
-
I'll let other's argue if "PhotoYcc, which was based on an RGB color space" is valid.
A few quotes from this Kodak white paper (http://www.kodak.com/global/en/professional/products/storage/pcd/techInfo/pcd-045.jhtml):
"Note that the RGB values delivered by the reference image capturing device are not constrained to positive values. This allows the PhotoYCC scheme to include colors outside the CCIR 709 display primary gamut. Furthermore, the reference image capture device values may exceed the values for a 100% white in order to preserve important specular highlight information."
"If you are familiar with video broadcasting standards, you may recognize that for positive values of R, G, and B, the nonlinear transformation corresponds to the optoelectronic transfer characteristic defined in CCIR 709. As a result, colors that are outside of the gamut defined by the CCIR 809 [sic] primaries are encoded by the negative values."
Jim
-
I'd shutter to think that in 10 years, the gamut of output devices the masses are using isn't wider than today.
4K HDTV coming pretty fast. 4K will use the Rec. 2020 spec, which has a wider gamut than anything currently on the market.
(http://upload.wikimedia.org/wikipedia/commons/thumb/b/b6/CIExy1931_Rec_2020.svg/300px-CIExy1931_Rec_2020.svg.png)
Wikipedia article on Rec. 2020 (http://en.wikipedia.org/wiki/Rec._2020)
From the Wikipedia article: "In coverage of the CIE 1931 color space the Rec. 2020 color space covers 75.8%, digital cinema covers 53.6%, the Adobe RGB color space covers 52.1%, and Rec. 709 covers 35.9%." Which smokes my NEC PA241W.
There are 4K TVs on the market right now. (http://www.tigerdirect.com/applications/SearchTools/item-details.asp?EdpNo=7674736) (but I don't know if this one is actually Rec. 2020.)
I extracted most of the above info from a recent thread on Slashdot:
High End Graphics Cards Tested At 4K Resolutions (http://slashdot.org/story/13/04/30/2151231/high-end-graphics-cards-tested-at-4k-resolutions)
-
^ That particular one is not quite ready for prime time, according to a PCMag revue (http://www.pcmag.com/article2/0,2817,2418007,00.asp). Looks like that Mt. Everest-like, Rec. 2020 green peak and vision-spanning blue to red base is gonna have to wait for better technology.
(http://www4.pcmag.com/media/images/383338-seiki-se50uy04.jpg?thumb=y)
-
Even if you get a " sooper dooper " monitor then you have to get " sooper dooper " profiles and a " sooper dooper " bang up to date printer. if you are outputting for the web then viewers won't have a " sooper dooper " monitor as good as yours. It is all problematic and to say the least expensive. :)
-
The photography industry is tiny compared to the consumer electronics industry. The vast middle ground of photography uses the leftovers of the consumer electronics industry. The first generation of P&S cameras used sensors that were originally designed for camcorders, which matched 4:3 CRT TVs. Current consumer computer monitors use panels that were designed for HDTVs. etc.
3D was the last big thing for consumer electronics and I think everybody agrees that that was a bust. The consumer electronics industry likes to sell everybody all new everything. The last time they pulled this off was with (approx. 2K) flat screen HDTVs/Blu-Ray. If they want to repeat selling everybody all new everything they need to make a compelling case for 4K.
I thought that 4K was ambitious enough that it would be years before anything tangible hit the market. Which was why I was surprised that 4K products are starting to trickle out. Don't judge 4K/Rec. 2020 to be a total failure because the first generation products are a bit lacking. The first $5,000 3 megapixel DSLRs weren't very impressive either.
If (and this is a big if) the entire consumer electronics industry succeeds in rolling out 4K that meets the Rec. 2020 standard, then the massive gamut I displayed in my first post will be the new normal--everybody will be using this gamut. sRGB will be a distant memory, just like 4:3 CRT TVs are now.
We can only hope.
-
Even if you get a " sooper dooper " monitor then you have to get " sooper dooper " profiles and a " sooper dooper " bang up to date printer. if you are outputting for the web then viewers won't have a " sooper dooper " monitor as good as yours. It is all problematic and to say the least expensive. :)
If you get a sooper dooper display like the SpectraView, you're getting a super duper display profile!
You might need profiles for output, maybe not. Epson for one supplies superb canned profiles for their papers. Other's can as well. YMMV.
But yes, sending images through the web means that you have to accept some poor slob who's got a 1994 CRT display that can't hit 67cd/m2 and the most neutral image will appear with a color cast.