- This topic has 11 replies, 4 voices, and was last updated 4 years, 4 months ago by
.
-
Posts
-
Hello,
I am learning about color calibration but need help.
I am looking at two monitors, one to buy. One has 100% Adobe RGB, 125% sRGB, 90% DCI-P3 % and the other monitor has 87% Adobe RGB, 100% sRGB, and 90% DCI-P3.
The first monitor only is factory calibrated for HDR DCI-P3 and SDR sRGB. It has lots of calibration controls.
The second monitor is set up the same.
I understand how to get the DCI-P3 or the sRGB, they have controls to switch between the two. But how would I access the Adobe RGB color gamut? Do I need to calibrate for that? If so , how?
AdobeRGB mode is a gamut emulation mode, arestricted version of native gamut. If there is no OSD control for “AdobeRGB mode” and monitor does not have some tool for HW calibration… you can’t do it.
Do you need it? Usually no. Use displayCAL with native gamut “user mode” OSD (or “custom” or whatever name this manufecturer uses), the OSD mode at native gamut where you can tweak RGB gains for white, etc.
Resulting profile is enough for color managed apps to render images (for example AdobeRGB images) to whatever gamut your display has… or to cut where your display cannot show those color.To echo what Vincent wrote: you can get the bigger gamut unit (all other things being equal?) and put it in DCI mode, profile it and assuming color managed apps with all the usual caveats you can see / work in Abobe RGB as DCI coverage includes Adobe.
Strictly speaking the “DCI P3” mode is probably the Display P3 variant, a standard created by Apple, which uses DCI-P3 gamut and sRGB curve and whitepoint.
DCI stands for Digital Cinema Initiative and P3 is a wider color gamut spec made for theaters with a 2.6 gamma, which assumes controlled very dark lighting and huge screen (where shadow bits count for a lot) and a whitepoint bias typical of the powerful lamps used in large projectors (and maybe some other psychology of cinema — idk).
Adobe RGB is a variant of sRGB released in 1998 to cope with printing industry standard offset press color. It extends green to get better cyan ink representation. It was a reaction of Adobe customer base to shortcoming of sRGB to their industry.
Note that pulling one primary can have a noticeable effect on color near the other primaries, because it works via triangulation. So Adobe RGB looks overall more colorful than sRGB but the extension is in green primary only.
P3 pulls all three primaries for a more uniform total gamut extension, and is an excellent step to better displays and future with more color, and a fair compromise with what already works well. It’s in a lot of new mobile gear with Apple leading way across all products.
As aside, UHD TV spec went nuts w BT.2020 and tried to grab for the entire visible gamut. Great for creating hallucinatory simulations maybe—I joke—but is also difficult to achieve, largely unnecessary and prone to problems, not the least of which is incredibly toxic waste from the chemicals required to make the displays. And outside R&D no one has one that can do it, so there’s this obnixious idea of color “containers” where UHD TV industry is forgiven indefinitely into future for not meeting its own standard. Cinema (old film) guys are way more practical.
Back on track: With true P3 mode and color management you will get the Adobe RGB coverage. And if you need to you can switch to sRGB mode for legacy conditions.
Don’t get an Adobe RGB spec monitor unless you know exactly why you need it. That’a passe. You want P3 capable with an sRGB mode.
You’ll have to profile both ways if you want to switch back and forth and this means swapping switching display controls and OS profile together. Or being super aware of how software is playing into your work load.
This P3 stuff is nearly baked, so now’s an ok time to head that way with new gear purchase, assuming the monitor can be switched back to sRGB.
Stuff that caught my attention are along lines of Dells with PremierColor, e.g., UP2716D or BenQ SW271.
Dell reviews have early complaining about panel uniformity., but this is a common problem. The last two digits in the Dell model numbers are the design year, so that’s 2016 era design. There’s something to be said for a production line that’s a couple years old.
The BenQ has been written up as a true “reference” display with perfect color out of box.
Buyer beware. This is anecdotal. I suggest these as examples of gear that’s well designed, not insanely expensive, and reputable. I haven’t seen it myself.
To echo what Vincent wrote: you can get the bigger gamut unit (all other things being equal?) and put it in DCI mode, profile it and assuming color managed apps with all the usual caveats you can see / work in Abobe RGB as DCI coverage includes Adobe.
[…]
Back on track: With true P3 mode and color management you will get the Adobe RGB coverage.
No, that is not correct, not even close to truth. There are a lot of colors outside P3 but inside AdobeRGB, printable cyans, tuquoises and greens, also there are colors in P3 outside AdobeRGB.
Displays with native gamut taht covers both worlds to a great extent are usually photo monitors, like GBLEDs, some QLEDs (benq) and WLED PFS phosphor displays in its AdobeRGB flavor (like newer CGs form Eizo or NEC PA271Q).
WLED PFS have several variants with green wavelegths diferences in spectral power distribution. The biggest is the one use in those photo monitor (AdobeRGB AND P3). After that comes apple p3 screen “close to exactly P3”, no more no less. After that comes 95% P3 WLED PFS like in those gamming or multimedia displays (popular in IPS and VA variants).That fake fact about P3 > AdobeRGB was popular in the days of the first P3 displays from Apple. Their legion of paid bloggers started a “campain” so P3 and its lack of some printable colors where viewed as a no concern for photo hobbyist and traditional casual Apple user “you gain A loosing B” and that was false.
The truth is that since U2410/PA241W amost 10 years ago all widegamuts for photo, inclusing those old CCFL widegamuts cover a huger percent of BOTH colorspaces.
Years later when GB-LED appeared they had typical 99% AdobeRGB coverage (full typical offset coverage and some greens fro “fineart printers”) and about 94% DCI-P3 coverage.
Then we moved to WLED PFS, in AdobeRGB flavor covering 98-100% AdoberGB and 98% DCI-P3.So P3 screen > AdobeRGB screen is a fake tale pushed on by some apple friendly blogs and web. Our AdobeRGB displays covered both worlds from a long time ago, and as time goes on both coverages get bigger… but P3 displays like the ones used by Apple are just P3, so no AdobeRGB printable colors.
Anyway Thomas’ display has a slightly uncommon gamut: “One has 100% Adobe RGB, 125% sRGB, 90% DCI-P3 %”. Maybe those numbers come from factory calibrated OSD modes (=AdobeRGB OSD mode). I would say that native gamut provided by backlight should have bigger P3 coverage unless it’s an old Widegamut CCFL, but IDNK actual model name of his display
Ah yes, I am wrong.
P3 leans towards red. Adobe: green. They exclude each other in specific ways.
My opinion comes from examining the intent of Adobe RGB to usage conventions.
Looking at coverage of sRGB and P3 to FOGRA39 (an estimate of an average press) you see that P3 does very close. And there are other important intentions of P3 beyond press. P3 is a different compromise.
If you run ICC Profile Info and compare gamut basics in a very perceptually oriented plot, such as DIN99c, you gain a good sense of P3 being a very fair compromise over Adobe RGB and printing, especially when you also consider that gamut extension, like many qualitative aspects, is a pursuit with rapidly diminishing returns.
VIncent is completely correct that Adobe RGB does stuff with green that P3 cannot do. And he is not wrong to call me out. So I stand corrected. P3 is not a perfect substitute for Adobe RGB for those who cares about perfect overlap in their domain .
While I admit that he’s correct, I don’t happen to share Vincent’s joy for the purely numerical differences.
Why do you want the extra gamut? A user who requires perfect conformity with a history of prepress standards may be required to go with Adobe RGB to comply with his profession. Is that you?
Ask yourself what you value, and examine the trends. Today I think it would be inappropriate to suggest to a new user to buy an Adobe RGB complete display over a P3 display. And if the user knows they need Adobe RGB’s cyan extension, say for a certain very rarified prepress workflow, then they aren’t asking for this kind of help.
A claim that P3 can work for all but the most detail-oriented prepress users is also correct.
Since the time of the creation of Adobe RGB, inkjet printers for photographers can do things with reds and yellows that are just as desirable to photographers. I will argue more desirable, because the extended gamut is more likely to be encountered in the real world. P3 is better suited to these colors.
And P3 is a major trend in commodities color. Conformance to industrial printing presses is not a major trend.
Any one who wants to explore this a little can fire up DisplayCal’s ICC Profile Info and do gamut plot comparisons.
I myself went through a rigorous examination of the impact of gamut on my style of landscape and urban photography. I created a special working space that starts with sRGB but is as much smaller than sRGB as P3 is greater than sRGB. I call it “Not-P3”. I ran a wide assortment of my photos through it an looked for differences both visually and via soft-proof and gamut warning. Much to my surprise, I could not visually detect the restriction over sRGB, except in the most contrived and saturated reds and oranges. And moreover, when soft-proofing, I saw no case when greens were limited in all kinds of outdoor settings with greenery. With Not-P3 I found that there is an obvious value of sRGB in reds. As I explored P3 more with a next gen phone, I was not able to easily come up with any ordinary scene was benefited by extension past sRGB. And I never found a scene where sRGB was insufficient to greens and cyans.
To quote myself quoting someone else, “absence of proof is not proof of absence.” But I discovered through my own careful exploration that the limits of sRGB only appear in certain rarified circumstances, such as flowers and sunsets, and man-made color. And maybe special effects in fantastical movies.
So yes, I am not right.
While it is fair for Vincent to call me out for being not right, I am also not wrong. and can see why Apple would make the claim in light of trends.
I stand corrected. But I am also not wrong, and I stand by the thrust of my previous opinion.
@Wire, it’s more easy: why should I choose between P3 and AdobeRGB when I can have both of them since 2010? (or at least since 2013 GB-LEDs)
The whole debate “P3 vs AdobeRGB” pushed on years ago by apple fanboys and their blogosphere is a nonsense. Get a display that can display the two, then choose as profile/colorspace for your content the one that best suits you: sRGB, AdobeRGB, P3-D65, DCI-P3 or eciRGBv2.
We can do that (and we do, without the “can” word)… all we but Apple users with P3 displays which are limited… to P3, so no choice for them, or a very limited one at best.
What I would love to see is some inter-vendor standarization of what ATI/AMD exposes in their AVIVO engine since 15 years ago: a lut-matrix-lut running in GPU.
This way we can get on GPU a system global HW-like calibration with simple gamut emulation, akin to Eizo CS/Dells/Benq etc HW calibration (thanks to dithering).
That would be cool 😀24 bit color is not quite enough to do bigger gamuts. I feel this constraint is the biggest factor in the compromises that inform engineering.
My take-away from investigation what color spaces can do for me is that sRGB was a very good design cut.
sRGB left stuff on the table perceptually, but it did so out of reasonable compromises, the most important being scarce bits. Keeping in mind that indexed, dithered color was a popular graphics approach at time sRGB was conceived, and bits were thousands of times more expensive than they are now.
In hindsight, the cut was slightly too conservative, but that’s arguing with history.
I feel Adobe RGB was always a red-herring. It commits a lot of bits to get access to a relatively small range of colors that are only meaningful to an industrial sector with a specific technical concern. From standpoint of perception, It puts gamut where it’s less needed. Printers and monitors are such different creatures of physics that it’s very tough to compare them perceptually. Due to the nature of our devices, we are trapped behind the power of the electro-luminescent display, and that’s never going to change.
The cinema guys made a better cut given the commodities state of the art. Obviously DCI is not Apple’s idea. Apple is following another medium, recognizing that scarce bits in a 24 bit RGB pipeline (or less if you consider preponderance of Ycbcr 4:2:0 in video content) are better spent on other colors, and that the system has to perform for legacy content.
Meanwhile, UHDTV is still being cooked.
I understand how if you are a gearhead, you might not like the DCI compromise for various reasons. I don’t follow how Apple fanaticism fits into any critique of this design. Please say more.
And please very much, I would love to hear other opinions that inform the engineering.
I agree very much that a more adaptable display interface would be great, but I also see why the industry is taking a long road to get there, and may never get there.
What I notice is the incredible complexity of this specific domain, and how the tools seem to be increasing the complexity, when the original promise of the ICC was consistent color everywhere. How wrong they were. And they’re mostly to blame!
But then again, why was this ever supposed to be easy?
Again that is not true or a half truth at best:
24 bit color is not quite enough to do bigger gamuts. I feel this constraint is the biggest factor in the compromises that inform engineering.
Not, that’s false. “Displayable” colors fit properly in 24bit, but calibration needs dithering. Same for gradients, 8 bit path to display is enough, even 16bit/10bit gradients. Lightroom and CaptureOne are the proof.
eciRGBv2 is big enough to cover a huge amount of fine art works that AdobeRGB and P3 do not cover… but small enough to be encoded in 8bit. It was designed that way.I understand how if you are a gearhead, you might not like the DCI compromise for various reasons. I don’t follow how Apple fanaticism fits into any critique of this design. Please say more.
It is false that I do not like P3. If indirty uses it for movies, I’m fine. I said that I DO NOT NEED a “just P3 display” like the ones from Apple or those multimedia gamer monitors because WITH CURRENT PHOTO WIDEGAMUT I have BOTH: I have AdobeRGB and I have P3.
Hence the whole debate is pointless.
If I buy a CS2730 or even better a WLED PFS I have AdobeRGB and a minimum of 94% of P3, up to almost full coverage if it was an AdobeRGB WLED PFS.
If you buy an iMac.. you do not have both.
It is extremely easy to understand. We had both to a great extent since almost 10 years ago.It’s just that Apple tought that P3-D65 would be the new “sRGB” of the 2020s and it’s pushing it on, not beause it’s better, or worse, just because if you make the new “defacto” internet colorspace you can color manage “less”, or provide an extremely oversimplified color management engine for basic 1 to 1 P3-D65 to screen and a basic sRGB to P3-D65 coordinates. But such “extremely oversimplified color management engines” break when you require them to to something more complex, like 3TRC profiles, LUT-table profiles on UI and such color issues that force you,apple users to use a simplified profile in DisplayCAL and the other calibration tools out there.
And again… I’m not against P3-D65 as new internet sRGB of the 2020s. Good, it’s fine.
I just said that I want an screen that can show both: AdobeRGB and P3 and if in a few years there is a new LED widegamut variant that can cover full eciRGBv2 that would be extremely cool and I would want it. “RGB” displays with gamuts much bigger than that… hmmm, not cool = observer metameric failure. Up to full eciRGBv2 it’s fine for me right now.
Re eciRGB, look at what a great cut NTSC was…
In 1953!https://en.wikipedia.org/wiki/RGB_color_space
http://www.brucelindbloom.com/index.html?WorkingSpaceInfo.html
Give it another 65 years and it’ll all get figured out.
Yes, sRGB was too conservative. But the tradeoff was fair: don’t waste precious bits on color that rarely occurs. Again, to argue about this is to argue with history.
I like your regard of Display P3 as sRGB 2020. That’s the way I think about it. As to a debate re Display P3, I personally have never seen any brouhaha about it.
Yes wouldn’t it be wonderful if we could have it all. Well, in that light, P3 is a better compromise that Adobe RGB, and to say so is not controversial nor to engage in debate. Adobe RGB was not engineered to be a great general purpose space. As you’ve indirectly pointed out SMPTE figured the best cut out about the same time as the cold war was going big. Adobe RGB was engineered to solve a gamut glitch for prepress. The fact that we even think about it occurs for no other reason than this specific point of history. Without it, the concern would not be on the table. So it it were abandoned it would mean nothing except to the printing industry. Do we talk about “Adobe fanboys”? No!
I don’t regard “Display P3 as sRGB 2020”, even though I have yet to sink my teeth into it. From what I read in this thread, seems to me Apple is trying to make sRGB “obsolete” but only for political reasons (to distance themselves from Microsoft/HP) or, for marketing reasons, trying to seduce Hollywood… I notice my iPhone11 pictures are coming out tagged as “P3” (?) which is new to me, having used a Windows phone for the longest time…
To come back to the original poster question, I’d say buy what he can afford. NECs and Eizo’s are good brands and both surprisingly well factory-calibrated. Sadly, I’m a little bit behind for not having kept up with the likes of BenQ SW271 and the NEC PA271Q…
Thanks for the lively discussion 🙂
My two cents….
Nothing wrong with following Hollywood, those guys should know something about color that works for consumers.
sRGB left gamut on the table. It’s time to reclaim. But we cant’t because ICC scheme didn’t pan out. Too complex; can’t simulate well enough. Hardware has to be involved. It’s still about industrial lowest common denominators.
“P3 “is Apple’s Display P3, an adaptation of DCI-P3. for cinema. Yes, it’s about Hollywood. UHD TV shot self in foot with insane expectations. TV came from movies. Can’t be rearranged.
P3 offers a well-rounded bigger gamut that’s not so far out as to be ludicrous to a simpler world. P3 is a useful enhancement. Sure it could go farther, but if it did the discrepancy would be too great for the color challenged.
NTSC knew. CIE is 1933 . My god.
Apple is smart to push displays far enough to have a great look for their users without totally alienating others. Microsoft—in other domains, not color—just gave finger to rest of world.
So yea, P3 is sRGB 2020
Wire, I think that you didn’t catch the whole point despite the amount of messages about this subject. To summarize:
-eciRGBv2 is meant to be used for print.
“Industrial print” usually covers smaller gamut (and AdobeRGB covers most of it), “Fineart print” (some people do not like this naming) covers huge colorspace volume like some Epson SC , Canon Prixma Pro and their “roll” bigger brothers like Image prograf & such.
DCI-P3 or P3-D65 is not close to cover what eciRGBv2 covers. It’s a fact.
It was designed that way instead of using ProPhotoRGB or their gamma variations (ProStar etc) because it is big enough to cover huge gamut for printing but small enough to work in 8bit (which is not recommended on ProphotoRGB).
It’s the same as your reasoning for P3 birth: an industry has some needs and decided a common way to work thet covers that needs-ICC scheme works very good. Dor example Little CMS works as intended and ACE works an intended.
If Apple never really cared about it when they move to macintel… it’s their problem, not ICC problem. ICC scheme works so well that it is equivalent to LUT3D, that is the reason you can make LUT3Ds with DisplayCAL choosing a ICC source and ICC destination.-As said many times before ***I/We/You*** have no need to choose a display “limited” by our current field of interest. Choose one that covers ALL!
Eizos do, NECs do (despite PA271Q fail in some units), even low cost vendors like Dell (UPs), Benq (SWs) do (although we can argue about QC and poor color uniformity) .
Apple displays DO NOT offer this… so they are pretty limited: in colorspace, in white point and in glossyness. Others… don’t.It’s like somebody soaked in the rain (Apple user) yelling at the clouds (this whole colorspace debate) while people around it walk to their destinations with an umbrella (current widegamut photo monitors covering most of all these colospaces).
Most of the umbrella men pass by without noticing this man yelling at the clouds, some of them point him for an umbrella shop and if soaked man keep yelling at the clouds, they sigh and continue their path.-If you need to “simplify” your image processing pipeline, or even do not use color management & color transformations on the fly with ICC, the HW needed is pretty simple and relatively cheap: a lut-matrix-lut.
Eizos do (and some even offer LUT3D), NECs do (with LUT3D too), Dell, Benq and these all low cost vendors do (and some even offer a public SDK).
Apple displays DO NOT have it, you have to rely in their bugged oversimplified ICC color engine unable to deal with accurate ICCs for UI.***
The problem is not in the industry (whatever it is): they choose colorspace meant for the content they produce.
The problem is not on display side (if we exclude QC/color uniformity). We had almost full AdobeRGB/P3 and a big amount of eciRGBv2 for a decade.
The problem is not caused by HW not be able to simplify this. We have relatively low cost lut-matrix-lut HW that can be used in “cheap” widegamut display since 6 years ago (U2413) and before that lut-matix or LUT3D HW on “pro” devices like CGs for a decade.
The problem is:
-color mangement at UI in Microsoft. Their answer is, if you need it (if you have a widegamut display) use HW, and we have it and we do use it. Not the best solution, but a working one.
-color mangement at UI in Apple. Their answer is we care only about “out of the box experience” for causal user (their actual market) so as long as their oversimplified color engine works for them, they do not care about other users needs.
And that’s why most umbrella men feel pity about the soaked men yelling at the clouds when they notice them.***
So again, Wire. GB-LED covers both worlds (printing, web, cinema). WLED PFS (AdobeRGB flavor) covers both worlds. Even lowcost chinese QLED from AUO covers both worlds (even better than GB-LED)
If apple does not cover both worlds… it is not an “AdobeRGB” problem… it’s a problem limited to apple users. Get an umbrella!
