MacInTouch Amazon link...

photography and color

Channels
Other
For those with an XRite i1 of the old mouse-shaped design that has been orphaned by lack of software updates*, the open source DisplayCal works very nicely with it. It works happily through a USB to USB-C dongle, and it made my jaw drop when I discovered I could use it on a Linux Mint MacBook Air and calibrate that screen as well. A new lease on life for a relatively old piece of hardware.

(*I think the last update was a 'Lion Edition' to get around the end of PowerPC code - there may have been software available after that but at a price I can't afford.)
 




This is not what I'm seeing. My iMac (Retina 5K, Late 2014) is running Sierra and the latest version of i1Profiler. This Mac often reverts from my custom monitor profile to the default iMac profile. Whenever I want to do color-critical work, I need to open the Displays preference panel and verify that my custom profile is selected. Another glitch: the Displays preference panel often shows duplicates in the list of installed (default or custom) profiles. I don't know what causes these problems, but I've seen the symptoms on multiple Macs, running calibration software from DataColor and X-rite.
That's not completely unheard of, and I've seen it before, but instead of "often", I'd put it more at "sometimes". I think you're right about Sierra - there's something a little bit wrong there, but it's hard to pin down, because it's sporadic.

"Duplicates": Also interesting, and, yes, it happens. There are various reasons for this.

1. There are multiple locations where profiles can be saved - although when running 3rd-party software, the wise thing to do is save into the user account location. Datacolor software started doing that several years ago (the latest versions of Spyder4/5/X software behave that way, and it can't be changed, for good reasons). XRite's software gives you a choice, but I would still stay with only saving the profile for the current user.

2. There's the issue of internal and external profile names. When you save a custom display profile, there's the file name itself and another internal copy of the name. It's the internal name (which may not be the same thing as the actual physical file name) that shows up in the Displays:Color profile list. What can get people in "trouble" is trying to rename a profile by changing the file name itself - that "works" at the Finder level, but it doesn't change the internal name at all, which will continue to be happily displayed in Displays:Color, unchanged.

3. Then there's the issue of file extensions. On the Mac, display profiles "can" have an extension (typically .icc), but it's not required and, internally, it doesn't look right for the profile name (which is shown in profile management lists in various system services and applications) to have .icc appended to it. Depending on the software you're using (and what version), your display profiles may have both external file names and internal file names without extensions, or your external file name might have an extension while the internal name (for lists) doesn't.

4. And if you try to rename profiles using at least some 3rd-party utilities, how that's dealt with by the application attempting to perform the "rename" can be an adventure. (I can tell you that System Preferences:Displays:Color uses only the internal names of the profiles when building lists, and if the same internal name exists at more than one physical location, it will show up as duplicates).

What you could do with duplicates (if they're custom profiles) is carefully delete the old ones and start over with fresh, new ones. It's tempting, but maybe counterproductive, to leave a litter of older display profiles behind as you recalibrate systems.

Datacolor's SpyderX software's SpyderUtility (which stays in the menubar as a background application) has a Profile Management command that lets you see your display profiles, switch between them, see more detailed info about them, and also has an intelligent rename function. (It also lets you delete profiles and, also, helpfully show them in the Finder, no matter what their physical location, and as you mouse over profiles names in the list it displays, it shows you the full file path to the moused-over profile below the list). All without going into System Preferences. That feature also exists for Spyder5 customers who upgraded to the "Plus" version of software over the past couple of years.

By the way, I know about these things, as I'm the developer for all of the Spyder consumer software for Datacolor. I've done a lot of coding and testing with this in various versions of OSX for the various Spyder applications, and also a certain amount of testing with other display calibration devices and software (XRite's, for instance; and also DisplayCal). I'm happy to answer questions like this about display calibration (and also printer profiling) and some of the system "plumbing" that goes underneath.

David Miller
Manager/Lead Developer, Consumer Graphics Software
Datacolor
 



Ric Ford

MacInTouch
Apple's ColorSync Utility, installed with the OS in the Utilities folder, can be used to change the internal name of a profile.
Yes, but it's a sick exercise in futility here... Here's what I just tried on macOS Sierra:
  1. Open /Applications/Utilities/ColorSync Utility.app
  2. Find the profile of interest in the sidebar and select it
  3. Click the Open button
  4. Select the appropriate data item (e.g. 'desc' 'desc' 98 Localized description strings)
  5. Select the appropriate field(s) (e.g. ASCII Name)
  6. Edit the value(s)
  7. Click the close button
  8. Respond to the alert dialog ("Save")
  9. The change is not displayed in the sidebar
  10. Quit the app
  11. Reopen the app
  12. The change has apparently been discarded.
  13. Go to System Preferences > Displays > Color. No sign of the change.
Am I missing something?
 


I seem to remember something about permission and location.

Save the file to your desktop then copy it back and provide your admin password?
 


Yes, but it's a sick exercise in futility here... Here's what I just tried on macOS Sierra:
  1. Open /Applications/Utilities/ColorSync Utility.app
  2. Find the profile of interest in the sidebar and select it
  3. Click the Open button
  4. Select the appropriate data item (e.g. 'desc' 'desc' 98 Localized description strings)
  5. Select the appropriate field(s) (e.g. ASCII Name)
  6. Edit the value(s)
  7. Click the close button
  8. Respond to the alert dialog ("Save")
  9. The change is not displayed in the sidebar
  10. Quit the app
  11. Reopen the app
  12. The change has apparently been discarded.
  13. Go to System Preferences > Displays > Color. No sign of the change.
Am I missing something?
Ric,

I just went through the process (in High Sierra on this particular system, but it would be the same in Sierra). I think the problem may be: don't try to rename the current display profile (the one you've got selected). The profile is "open" and the system won't let you save it. Try it this way instead:

- Select a profile in System Preferences > Displays > Color. (Note that doing this makes it the current display profile immediately, and you'll also see a shift of color on the screen, assuming its internal calibration luts are "different" than what you previously had set as your display profile).

- Click the Open button to open it in ColorSync Utility.

- (Important): Back in System Preferences > Displays > Color, select and switch to a different profile.

- Now, back in ColorSync Utility, edit the "desc" tag and make the same changes to whichever internal variants exist. There may only be a name in ASCII Name: there may also be duplicates of the same in the Unicode and Mac Script name fields. If the name is in all 3, modify it in all 3, and make it the same, for all 3. You can shorten or lengthen the name if you want, but they should all match. (Technically all 3 fields "should" be filled but it's not a requirement; many profiles only have text in the ASCII Name field)

- Save, from ColorSync Utility's file menu.

- Back in System > Preferences > Displays, click and switch out of the Color tab, and then back in; that'll refresh the name list and you should now see your changes.

- You think you're done? You're not. You're wondering why? Re-read my longer post above and then keep going here. (spoiler: remember? Internal profile names are completely uncoupled from the external file name). (ah ha)

- So what you did above "works" but you only changed the internal name. Not that there's anything wrong with that, but prepare to be confused if you look at the physical profiles on the internal drive and the file name hasn't changed. See how problematic this can be?

- Last step then, to avoid future confusion and (if appropriate) hair pulling. In the Finder, locate the profile that you just changed the internal names for, and also rename the file, the usual way. Now you're "done" - you've changed the internal name (not simple) and the external name as well to match it. None of this is really that simple at all. How many typical users would want to do remember how to do this?

A few more comments.

- System Preferences > Displays > Color can be tricky, because every time you click to select a different profile in the list, it "becomes" your current display profile (the screen should immediately shift to whatever calibration luts are embedded in it). Don't forget this. If you want to select profiles and then open in ColorSync Utility, the very act of selection will change your display profile; you'll want to set it back to what it was previously set to. A way around this is to directly launch ColorSync Utility and use "it" to select profiles if you want to look at them in more detail and possibly change the internal names (in which case, profile selection doesn't also switch to it as the system profile).

- ColorSync Utility can be wonky. If you click the Devices tab in its main window, be prepared to wait for a while as it will become non-responsive (internally it'll be searching through all profiles on the system and separating them into categories based on device names, rather than as a simpler profile list). Eventually it "will" become alive again. Set ColorSync Utility to Profiles in the tabbed section at the top for more reasonable performance.

- This is one of the reasons why we added the Profile Management: Rename function to SpyderUtility in the Datacolor Spyder5 "plus" updated software; and also in the new SpyderX display calibration software. It's a useful tool that makes things like this much, much easier. In Profile Management: selecting a profile doesn't immediately set it as the current display profile; so you can simply select a profile in the list, click a button to rename it; change it however you like; and then save. Everything described above thing happens in a single action, no multiple steps required. The internal name(s) get changed and the file name gets changed, all in one shot. Even if you just use ColorSync Utility by itself to rename a profile (internal name), you should also rename the file itself in the Finder as your last step.

David Miller
Manager/Lead Developer, Consumer Graphics Software
Datacolor
 


Ric Ford

MacInTouch
Try it this way instead...
  1. I don’t believe I was editing the selected profile (but I didn’t try all different sorts of profiles after several failed attempts).
  2. All my experiences, alongside others’ notes, suggest that macOS 10.12 Sierra is completely broken and dysfunctional in this area. I don’t know if macOS 10.13 or later fix that.
  3. I’m sure you realize that these procedures are insane in terms of user interface and the original Macintosh design philosophy.
 


1. I can try it with Sierra over the weekend. :-)

2. Perhaps, but these are the kinds of low-level obscure [areas] that very few users will ever want to navigate. A simpler solution is to use whatever display calibration product you have, and build new profiles, vs. trying to change the names of existing ones.

3a. Totally agree. The bit about internal profiles names in the 'desc' tag being separate and uncoupled from external file names goes decades back, to how the ICC profile spec was implemented. And they "did" have their reasons. For instance, it wasn't uncommon (at least back then) for external profile "file" names to be cryptic and indecipherable by humans, but for the internal profile names to be descriptive and readable. Even now this happens - for instance, the actual file names for the "generic" base profiles that Apple provides for known displays. Let's say you have a BenQ display. The default profile that OSX assigns to it, the first time you plug it in, shows up as "BenQ" in OSX lists and popups (the internal name) but the actual file name is something quite different (in /Library/ColorSync/Profiles/Displays where these "special" profiles are kept at the system level, the actual profile is named "BenQ PV270-585404F2-C5E9-3E1A-F1C1-D9589FEE8094.icc" - too long and unfriendly to show to humans in a list, but the extra hex digits are added to the file name as a unique identifier.

3b. The entire thing about internal profile names vs. external file names sounds simple at first, but from a coding standpoint, how you try to deal with this thing as a developer ends up being much more complicated. Once you try to implement a function to fully rename profiles and do so "intelligently" without causing problems, it's not straightforward at all.
 


Ric Ford

MacInTouch
The entire thing about internal profile names vs. external file names sounds simple at first, but from a coding standpoint, how you try to deal with this thing as a developer ends up being much more complicated. Once you try to implement a function to fully rename profiles and do so "intelligently" without causing problems, it's not straightforward at all.
Hmmm... I could imagine a developer creating a simple utility to just solve a few of these sorts of problems and distributing it at a low cost or free/shareware as a way of marketing the developer's other products. :-)
 


Props to David Miller. Good description. (Sorry I was late coming back to the party.) I make my own profiles, so dealing with them has become second nature. I use ColorThink Pro for manipulating them (and all its other wonderful features), but it's not cheap.

My biggest, and very longstanding complaint regarding profiles is that there is no way to categorize them. For example, when I print and need to choose the correct profile, the selection has to be made from a list of well over 100 entries, most of them entirely irrelevant to printing (such as scanner profiles, monitor profiles and so on.)

Why Apple could not have set up the system so simply use folders, like everything else in the OS, is beyond me...
 


Thanks, Tracy! I think you can create your own subfolders inside the Profiles folder in your account, and profiles contained in those subfolders will be accumulated and listed by macOS, along with those at a higher level. But even then, that separation and grouping by folder only exists at the Finder level. If you go into System Preferences > Displays > Color, all of the display profiles are flattened into the list that's displayed, and you don't see the folder structure.

If the same internal profile name exists at multiple locations, it will show up as "duplicates" of the same name in the list, even though each "duplicate" is most likely to be: a discrete profile at a unique folder location. ("Most likely" because even that's not guaranteed; remember, the list shows internal names; there could be more than one physical profile file, in the same folder on the driver, with the same internal name but with different file names).

Depending on the software you're using to print, the list of choices can be filtered and narrowed while the popups are being filled. For instance, display profiles could be excluded from a popup when it's known you're only looking for printer profiles, or vice-versa. Your mileage will vary depending on the application you're using.

For example: in the SoftProofing feature in Spyder5Elite+, and SpyderXElite, I implemented the ability to not only select a profile for soft proofing from the main popup list (which defaults to all types of profiles), but also the ability to additionally filter the main popup contents by using a second popup to specify that the main list will be filled with display profiles only, or printer profiles only.

(It's more likely that someone would use this feature to soft proof through a printer profile, but it's also technically valid to cross-proof an image onto a display profile of a different gamut; for instance, maybe your system has one display only that's wide gamut, like a modern MacBook Pro or iMac (P3 wide gamut), but you'd like to see what the image would look like on a smaller gamut display, such as sRGB; you can select sRGB as the proofing profile, turn soft proofing on, and see the effect, such as highly saturated reds on the P3 display would become less saturated and "duller" on sRGB).

In the case of a system that has a large number of profiles installed, it's user-friendly for software that fills color management popups to provide a way to thin the list down. There are other, more advanced things that could be done, too, such as filtering based on user-typed characters in a text field, printer manufacturer ID, etc., and these kinds of specials could be done on the fly at the time the popup menu list content is built or refreshed.

David Miller
Manager/Lead Developer, Consumer Graphics Software
Datacolor
 


... To my surprise, viewing a few good Nikon D750 raw files in Preview, I really couldn't distinguish much difference between them. I don't know if Preview is affecting results or if I need better tests, but I expected to see a more obvious distinction. (More to come, I guess.)
Preview isn't the best (or even a good) test using raw files. A raw converter (Apple's, Adobe's or any of the dozens of others) generally uses the sRGB setting from your camera when first displaying an image. Since the purpose of a "demosaic" program (a.k.a. raw converter) is to allow you to make the adjustment, and the file itself isn't actually an image yet, they need some setting to start with, and they usually choose what matches the monitor on your camera: sRGB.

Upshot: you didn't see much difference, because the default setting for all was the same: most likely sRGB. (The point is the sameness, not the color space, although, as you noted, sRGB is the smallest.)

To see a difference, you should not use raw files but some JPEG or TIFF color-checker file, such as many of the files on this excellent resource:
Keith Cooper said:
Hope this helps.
 


Ric Ford

MacInTouch
... To see a difference, you should not use raw files but some JPEG or TIFF color-checker file, such as many of the files on this excellent resource:
Thanks! I'm finding some other resources (but still looking for additional collections of ICC profiles). A Gamut Test File (among others here) seems quite useful:
Here are some other resources I found:
On the flip side, I'm finding it very difficult to come up with any good examples or tests to show discernable differences between the supposedly superior wide-gamut iMac 5K display vs. any other decent display in real-world use, including digital photography. (Specific suggestions are welcome.)

Lastly, some perspectives on sRGB vs. Adobe RGB vs. Apple's Display P3:
 


On the flip side, I'm finding it very difficult to come up with any good examples or tests to show discernable differences between the supposedly superior wide-gamut iMac 5K display vs. any other decent display in real-world use, including digital photography. (Specific suggestions are welcome.)
I have taught this stuff for over a decade now, and my own prints are in museums. I'll note this to begin with: it can be confusing, and there is a ton of simply wrong information on the "fount of all wisdom" (a.k.a. Internet).

First, I'd note that literally seeing the difference is, in fact, difficult without practice. The brain makes all kinds of adjustments automatically. (See here:)

Because of that, as your eyes switch between physical monitors, seeing the difference is almost impossible. If you monitor allows it (as some do) and you can switch the monitor itself between Adobe RGB and sRGB, it will still be subtle, but easier to note.

For this kind of test, as for printing or photo work, your monitors (both/all of them) must be both calibrared and profiled. A hardware "puck" is the only serious tool to use. I use the X-Rite i1 Display Pro for my monitors. Their software is good, but there are also third-party apps, such as basiCColor Display. Most of the time both calibration and profiling happen in the software that does the work: first the monitor is calibrated (i.e.: R+G+B = true gray), and then it's profiled for individual "errors" across a range of colors. Failing to do this is like taking two untuned race cars to the track and comparing their lap times.

For full benefit of a wide-gamut monitor, you need a full 10-bit path: the monitor, the card driving it; the software you're using. Most consumer hardware is still 8-bit, which pretty much limits it to a smaller range.

You also want to be looking at 16-bit images, for the same reason.

Finally, the wider color spaces are (with practice) visible on the monitor, but the effect is subtle. They are mainly used for photo and film color manipulation. While the wider space is more useful in video work, since the difference can be seen when projected, it is less advantageous to printed output, since no printed image can come close to the full gamut of the larger spaces. The gamut of a given paper & ink is always smaller than the color space.

So why bother? Because the larger color space (Adobe RGB and ProRGB) allow for more "breathing room" for tonal and color adjustments. Such manipulations are evenly spread, like peanut butter, in a larger space, where as in sRGB, the tonal adjustments are like stair-steps, or checker boards: not smooth but segmented steps.

Yes: it's still not going to "fit" on the paper, but the process of reducing it to fit has better information to work with.

Bottom line? To eyeball just one monitor vs another, they must both first be calibrated and profiled. Even then, you're going to find it subtle.

Finally, if you are using your computer for looking at family photos; playing games; and "tuning" snapshots in Preview, you really don't need to spend the cash on a wide gamut monitor (although you really should still use the hardware puck, in my opinion).

If you're printing your images or editing video, it's more up in the air. If, like me, you are doing it for a living, then there's no question you need the more expensive gear.
 


Ric Ford

MacInTouch
First, I'd note that literally seeing the difference is, in fact, difficult without practice.
Thanks, Tracy. Good to know (wish I'd known earlier!).
... For this kind of test, as for printing or photo work, your monitors (both/all of them) must be both calibrared and profiled.
For what it's worth, I did buy a hardware profiler/calibration puck and have used that and all kinds of monitor tests pretty constantly.
... For full benefit of a wide-gaumet monitor, you need a full 10-bit path: the monitor, the card driving it; the software you're using.
Is there a clear way to determine this? Is a 2017 iMac 5K Retina fully 10-bit for instance? What if I hook up a 10-bit LG 4K display over Thunderbolt to a 2015 MacBook Pro running macOS 10.12? How about a 2018 MacBook Pro with an eGPU housing a Radeon 580?
You also want to be looking at 16-bit images, for the same reason.
I developed some Fuji X-T3 raw files into 16-bit TIFFs using Capture One, outputting with several different color profiles (sRGB, Adobe RGB, DCI(P3). The scene was loaded with bright, saturated natural colors and a full range of values (green grass and foliage, red and yellow and purple flowers, plus whites and stone and shadows on an overcast day). I couldn't see any differences on either the iMac 5K or Viewsonic VP2770 among the different files, though they had the desired color spaces when the EXIF data was checked.
So why bother? Because the larger color space (Adobe RGB and ProRGB) allow for more "breathing room" for tonal and color adjustments. Such manipulations are evenly spread, like peanut butter, in a larger space, where as in sRGB, the tonal adjustments are like stair-steps, or checker boards: not smooth but segmented steps.
Thanks, this makes sense to me — like editing audio at higher "resolutions" to preserve quality during effect changes.
If you're printing your images or editing video, it's more up in the air. If, like me, you are doing it for a living, then there's no question you need the more expensive gear.
I can understand that, but can even you actually see the differences that will ultimately appear in print on your screen during soft proofing?

And, lastly, will Apple's exhorbitantly priced Pro Display XDR have any real, visual, major effect on your photographic processes, for instance?

(Bonus question: Have you ever tried the Dell 8K display?)
 


I can understand that, but can even you actually see the differences that will ultimately appear in print on your screen during soft proofing?
Generally speaking, no. Most people simply aren't trained well enough to see very subtle differences, and then we have the whole issue of color blindness to talk about, too.

As Tracy said, even something as simple as background or adjacent tonal values can skew your brain into seeing something that isn't there.

For several years at the turn of the century I had some of the best color experts in the world at the time sitting downstairs from me and worked with them regularly as I tried to improve the image quality of the magazine I ran. They simply would not "evaluate" color without it being done in a controlled situation (e.g., their very carefully set up workstations, which were in a very light-controlled room and shaded for glare, among other things).

When I work with students, the first thing I start with isn't color, it's tonal recognition. If you can't recognize basic tonal issues, you won't be able to recognize color issues.
 


Thanks, Tracy. ...
Re hardware profilers: There are inexpensive ones, which are not much better than nothing, and more expensive ones, which are visibly superior. The generally accepted best monitor-only one is the Xrite i1 Display Pro, but it's over $200. (I also use an i1 Photo Pro 2, which is about $1800. Even with the more expensive product, which I use for creating my own paper profiles, I choose to use the i1 Display Pro for my monitors.) There is a product called "Spyder" (I believe) that is about 1/2 the cost of the Display Pro, but you get what you pay for.

Is there a clear way to determine if you have a 10-bit path? Yes, fortunately, and it's pretty easy. Choose a test image that is just a luminosity (B&W - no color) gradation shift from black to white. This is easy to create in Photoshop: just drag out a gradation from the left edge to the right edge. You can also find these easily on many of the site you mentioned, above. Make it a large area, and be sure that it runs horizontally, not top to bottom, then just look at it occupying most of your screen area. If you see vertical bands, like steps, of tone then you have an 8-bit system (although you cannot tell which component is causing it to be 8-bit from this test.) If, on the other hand, the tone is smooth across the width, you have a 10-bit system.

If your monitors show banding, then regardless of the color gamut of your images, it will be almost impossible to see any difference.

Re: your experience with the images: the place it's "easiest" to see is in the greens. There is a lot to conflate here. For one thing, the backlight is probably different between the monitors, so the color temperature of the white will not be the same. (It's very hard to even get two monitors of the same make and model to match, much less different ones.)

Your example of audio editing is a perfect corollary.

Can I actually see a difference? Well, if I were to walk into a room full of monitors, I certainly couldn't pick out the ones with larger color gamut at first glance.
I can understand that, but can even you actually see the differences that will ultimately appear in print on your screen during soft proofing?
Can I tell the difference? Yes, I can, or, frankly, I would not have paid all the extra bucks for pucks, monitors, display cards and other hardware. (On the other hand, I've been doing this for a couple of decades now, so I'm pretty sensitive to it.)

Thom Hogan (one of my digital heros, for all of his contributions and knowledge—thanks Thom) is absolutely correct regarding tonal recognition, and it applies significantly as a needed adjustment for monitors as well. There are test images that show a series of numbers from 1 - 60 (-ish) against a black background. Black is RGB 0,0,0. The number one in this test is 1,1,1; followed by two at 2,2,2 and so on. The point is that 99% of the monitors out there cannot, without very careful adjustment, allow you to distinguish any black below level 6 or so. "1" to "6" just look like pure black.

I can, in fact see them all.

There is a Photoshop layered image that I think is excellent for tuning the luminosity and contract of monitors, provided by Aardenburg. It remains my go-to for such adjustments. You can find it here: Aardenburg Imaging MonitorChecker. (Photoshop is required for this to be used.)

You asked:
I can understand that, but can even you actually see the differences that will ultimately appear in print on your screen during soft proofing?
Well, that's sort of a non-sequitur - the correct answer is simple: no. You will never see on your monitor screen what your final print will look like, as it's physically impossible. Monitors are direct light, and photos are reflected light. "Ne'r the twain shall meet." Going back to the "old" days, for exactly the same reason, your slides will always have more pop and gamut than a print from them.

What you can do is get "close enough" to take advantage of "your brain messing with you" and allow you to interpret properly enough to make a print that "matches" (given your level of skill with Photoshop, of course).

I'm at the point in my own skill level where I can get a darned close first print, and usually only require one more to reach my artistic vision. (Note that is not the same thing as a "match" to the monitor, which, as noted above, simply isn't possible.) That "two-print" is probably true 60% of the time. The remaining 40% takes one or two more.

To my way of thinking, the bottom line in photography is the print, nothing but the print. The equipment used, the technique, and all the other shiny objects are irrelevant. It's only the print you hold in your hand that is ultimately important. What you want to do is get there with the least effort, and that's where the tools come in. That's why I have a more expensive puck; wide-gamut monitor etc.

Finally, you asked if the
...exhorbitantly priced Pro Display XDR have any real, visual, major effect on your photographic processes?
I've not seen it, but I expect that given the gamut of a physical print, the answer is likely "no."

Video/film editing on the other hand, the answer would be "yes" because the gamut of the output is so much larger. Yes, I do video editing, but not on the major-studio level, so, no, I won't be buying one. (Donations accepted, however :-)
 


Ric Ford

MacInTouch
Thanks, Tracy for all that information and perspective, and I understand and agree with a great deal of it, including the difference between reflected and direct light, which was partly why I wondered about the rationale for purchasing expensive wide-gamut displays.
... To my way of thinking, the bottom line in photography is the print, nothing but the print. The equipment used, the technique, and all the other shiny objects are irrelevant. It's only the print you hold in your hand that is ultimately important. What you want to do is get there with the least effort, and that's where the tools come in. That's why I have a more expensive puck; wide-gamut monitor etc. ...
This brings up the big issue, and it's fascinating: how do any of us want to deliver our images, who is going to view them, and how? I started in the darkroom with silver emulsions and chemicals and will forever love great silver prints. But now I, and everyone else to the greatest extent, view images almost exclusively on electronic displays with all the attendant advantages and issues. I suppose we, collectively, want a computer system that can help us effectively deliver excellent content in all possible media.
Video/film editing on the other hand, the answer would be "yes" because the gamut of the output is so much larger. Yes, I do video editing, but not on the major-studio level, so, no, I won't be buying one. (Donations accepted, however :-)
That's a great point, because viewing cinema video is so different from viewing still images, and TVs are different from computer displays (generally). I suppose "wide gamut P3" (DCI P3) makes sense for Apple customers who are producing professional video, as it's cinema-based rather than print-friendly like Adobe RGB. I haven't a clue about needs on that front and would be curious to hear how video professionals view these same gamut issues we've been discussing re still images.
 



Thanks, Tracy for all that information and perspective, and I understand and agree with a great deal of it, including the difference between reflected and direct light, which was partly why I wondered about the rationale for purchasing expensive wide-gamut displays.

This brings up the big issue, and it's fascinating: how do any of us want to deliver our images, who is going to view them, and how? I started in the darkroom with silver emulsions and chemicals and will forever love great silver prints. But now I, and everyone else to the greatest extent, view images almost exclusively on electronic displays with all the attendant advantages and issues. I suppose we, collectively, want a computer system that can help us effectively deliver excellent content in all possible media.

That's a great point, because viewing cinema video is so different from viewing still images, and TVs are different from computer displays (generally). I suppose "wide gamut P3" (DCI P3) makes sense for Apple customers who are producing professional video, as it's cinema-based rather than print-friendly like Adobe RGB. I haven't a clue about needs on that front and would be curious to hear how video professionals view these same gamut issues we've been discussing in relation to still images.
Yet another reason I read MacInTouch daily. I can barely operate my phone, I have a fancy digital camera I run on auto, and here I am getting a master class in Color Management from two of the foremost people in the field! Can't get this stuff just anywhere!
 


This brings up the big issue, and it's fascinating: how do any of us want to deliver our images, who is going to view them, and how? I started in the darkroom with silver emulsions and chemicals and will forever love great silver prints. But now I, and everyone else to the greatest extent, view images almost exclusively on electronic displays with all the attendant advantages and issues. I suppose we, collectively, want a computer system that can help us effectively deliver excellent content in all possible media.
Right you are, Ric. Technology will continue to advance, so this, too, shall pass.

The question of image quality occupies the minds of many in the photographic art world, as it did mine. The steamroller is unstoppable, and the direction is projected light (computer monitors, TVs and so on).

We have a generation (and the next and the next) who have never lived in a world without visuals from glowing boxes. Nuance, subtlety and depth are unknown, to the hand-wringing of old timers like me. But I think the worry is misplaced. Typewriters didn't kill literature, nor did word processors. The vast majority of the printed word isn't great art, but that doesn't prevent great art. Most music isn't destined to live for decades, but that doesn't stop great music from being written.

Cellphone cameras and monitors are not the death of photography.

In my decade of work with the Center for Photographic Art, I have seen the worry come and go. Folks who have ever only seen photographs on Instagram and Facebook come in the gallery, and are stunned by the power of the photographic prints on the gallery walls. They will stand and stare at a single image for minutes (instead of the three seconds on Instagram.) Young folks are more than ever taking that experience, and looking to apply it to their own images. Some go so far as to eschew digital altogether, and switch to film to better understand the experience, art and craft of photography.

This is most heartening to me, as I see so many "I wish I'd thought of that!" images.

The print isn't going away, and neither is Instagram. I can only smile at the bright future: artists will continue to starve; schlock will continue to sell; moments of brilliance will shine.

in my opinion, nothing has really changed after all.

your milage may vary.
 


As others have said, thanks, Tracy & Ric for the master class in photography!

I'm only a user of my iPhone camera and photo manipulation apps, but found this discussion fascinating! Thanks again.
 


Ric Ford

MacInTouch
I've just done a bunch of experiments with Tracy's help, so just a few quick notes, for what they're worth:
  • I've been comparing the 2017 iMac 5K wide-gamut display against a Viewsonic VP2770 I bought several years ago and have been using as a workhorse display. It turns out, though, that the VP2770 is actually an excellent display with 10-bit support, full sRGB coverage, and excellent overall visual quality (plus hardware controls completely absent from iMacs). I see virtually no difference between this non-retina 1440p/QHD display and the 5K Retina wide-gamut display of the 2017 Mac 5K in very extensive visual testing.
  • Tracy has a neat trick for testing for 8-bit vs. 10-bit display: looking for posterization with a grayscale gradation 16-bit TIFF (as he described above). I discovered, though, that if the TIFF is saved as an RGB file rather than grayscale, there was a very, very subtle blockiness — on both iMac and Viewsonic displays.
  • I finally managed to come up with a photo that demonstrated differences in color spaces, but it sure was subtle — only one particular brown value changed noticeably out of a full range of values from whites to greens to reds to earth colors — just one, and everything else looked virtually the same. This was a Nikon raw file saved from Capture One as a 16-bit TIFF in sRGB, Adobe RGB, DCI (P3) and "Nikon D750 Generic" (as well as CMYK, which looks obviously different vs. the others).
  • Tracy pointed to something on diglloyd.com and I discovered that Lloyd Chambers has done a lot there with color, etc. In particular, this is a great page that explains and demonstrates the critical issue of embedding ("tagging") color spaces in photos: Web Browser Display of Image Color: Color Space and Gamut
 


Re hardware profilers: There are inexpensive ones, which are not much better than nothing, and more expensive ones, which are visibly superior.
I'd tend to disagree here. While, no doubt, the better profilers do a far better job, doing even a basic, not-so-great profile is far better than not using one. The one exception here is the recent iMac monitors, which come pretty well adjusted from the factory and don't seem to drift much.
There is a product called "Spyder" (I believe) that is about 1/2 the cost of the Display Pro, but you get what you pay for.
One problem is that the lower-cost profilers tend to have simplified software, and that allows users to miss a critical step if they're not aware of it. That step would be: what is the actual monitor brightness set at?
Re: your experience with the images: the place it's "easiest" to see is in the greens. There is a lot to conflate here. For one thing, the backlight is probably different between the monitors, so the color temperature of the white will not be the same.
Right, and that's the other simplification that often comes with low-cost profiler and software.
There are test images that show a series of numbers from 1 - 60 (-ish) against a black background. Black is RGB 0,0,0. The number one in this test is 1,1,1; followed by two at 2,2,2 and so on. The point is that 99% of the monitors out there cannot, without very careful adjustment, allow you to distinguish any black below level 6 or so. "1" to "6" just look like pure black.
On my current monitor, I can't quite get to those lowest values. I suspect that may be due to the reflectivity off the monitor glass itself, another thing you have to consider. I should point out that for years we've been recommending that "black" should be at 20,20,20 (or maybe 10,10,10 in some cases) when you print. Why? Because of the way ink works and spreads.
Well, that's sort of a non-sequitur - the correct answer is simple: no. You will never see on your monitor screen what your final print will look like, as it's physically impossible.
Correct. But you didn't mention Soft Proofing, which is something that you really have to learn how to do (even if you still have to make adjustments to get that print "matched"). In the publishing business, we soft-proofed on our monitors, then soft-proofed again on our in-house printer (because the way inks work on printing presses is different than inkjet/laser/etc.).

The primary thing I learned is this: follow a known set of values all the way through the process. That generally means a ColorChecker chart of some sort. There are "tricky points" at every step of the way, and if you get something wrong at any one of those points, everything downstream becomes a mess, and possibly not "fixable."
Monitors are direct light, and photos are reflected light.
And don't forget that your source material is reflected light. ;~) Your brain gets in the way, because it's always trying to self-correct. A white T-shirt looks white to you in any kind of light because of that correction, but not to a device that captures what actually is being reflected. There was a time when I could walk into any situation and state the Kelvin setting, without measuring, within +/-10 MIRED, so you can train yourself to recognize what is happening in the world you're trying to capture.
To my way of thinking, the bottom line in photography is the print, nothing but the print.
This clearly is going to change. To some degree, it has already changed, as the way we "make" a print is different today than we used to, and it's going to continue to change in the future.

The problem here is that the still photography world isn't driving "standards" the way the video world—particularly Hollywood—is. The camera makers, in particular, seem to have no concept of how the world of imaging has changed since film, which is ironic, as they've made far more digital cameras than film cameras.
Video/film editing on the other hand, the answer would be "yes" because the gamut of the output is so much larger.
Oh dear, you've opened a can of worms there. In actuality, until some recent changes, video was actually quite more restricted than photography. A few years ago I was privileged to see early remastering of a number of Pixar films for the current best digital projection abilities, and it was a real eye-opener. Wow, did the gamut change. So did the dynamic range. And with those two things, other things appear to change, too. You have to be careful of how you grade, because you can start bringing out too much perceived detail via contrast and other changes that happen, and our brains are trained to see "movie" as a softish presentation.
 


Thanks, Thom. Actually, Ric's question mentioned soft-proofing, which is why I didn't specifically mention it again. That said, you're of course correct.

And regarding soft-proofing: using it successfully makes it all the more mandatory that your monitor is properly calibrated and profiled. If you have problems with gamma, then Gamma Control can be useful, but you must use it judiciously! (Leave color alone; only 0.0n adjustment in gamma. If you need more than that, something else is wrong.)

From my own workflow, Soft Proofing allows the following technique to be used to make better (read: "matching") prints:

1) Adjust the image to your liking in Photoshop (sorry, Ric ;-)​
2) Save it to disk.​
3) Duplicate that file on disk.​
4) Load them both back in, using a side-by-side split window.​
5) Apply soft-proofing to one of them (say the right-hand one).​
6) Leaving soft-proofing on, adjust that right-hand image to match the left-hand one (usually some vibrance, contrast/saturation, luminosity and/or curves adjustment layers will do).​
7) Turn off soft-proofing and save the right-hand file.​
8) Save a copy of the file flattened for printing, and use that to print.​
9) If you need to adjust (you will ;-) return to the file from step 7 and tweak away.​

For those unfamiliar with soft-proofing, it is simply running the image thru the paper profile you'll be using for printing. Obviously the tweaking is only good for that particular paper, which means you need to keep the the file created in step 2, and repeat the process if you plan to print on a different paper.

For black and white images, I have a printer dedicated to Jon Cone's Piezography inks. That replaces all the color cartridges with shades of gray ink, up to a recently released (and pretty spectacular) High Density black which allow for excellent d-max. I should note that Piezography will show up every flaw in your B&W images and your technique, and learning to use it is "non-trivial." The results, however, can be stunning.
 


An interesting innovation in printing has appeared in the past few years: rather than printing on paper, photographers are printing on aluminum. The result is a much more lively print (due, I speculate, to the reflected light from the aluminum). Thoughts?
 


Amazon disclaimer:
As an Amazon Associate I earn from qualifying purchases.

Latest posts