6 ms·

Why DCI-P3 and not Adobe RGB? While they're roughly the same gamut size, with P3 having more crimsons and Adobe more aquas, Adobe actually covers most of the C

Why DCI-P3 and not Adobe RGB? While they're roughly the same gamut size, with P3 having more crimsons and Adobe more aquas, Adobe actually covers most of the CMYK color space, while P3 does not (it is lacking in the aquas, like sRGB), and Adobe has a larger gamut in the highlights (due to its less saturated, and therefore brighter, red primary). Not to mention the fact that Adobe RGB has long been established (albeit not terribly widely used) in the graphics/photo industry. As another poster mentioned, even Apple couldn't find any P3 images to use as wallpapers, so settled on Adobe RGB ones instead. What's their goal? Are they aiming for better compatibility with digital cinema?

sliverstorm 11y ago
It always seems to me, Apple loves dodging the established choice. Firewire, Thunderbolt, I swear the list is long but I can't remember everything right now.
- techenthusiast 11y ago
  DCI-P3 actually is the relevant film industry standard here. No consumer displays can do Rec.2020 gamut yet, so P3 is a nice halfway step that is comparable to sRGB, unlike Adobe RGB. And it's "future-proof," because both can use the SMPTE ST-2084 EOTF. Both the P3 and Rec.2020 gamuts are part of the UHD specification.
- wtallis 11y ago
  Firewire and Thunderbolt both introduced important capabilities that no established choice could even come close to. Apple doesn't wait around for the beige box vendors to come to a consensus on technological advances.
  - sliverstorm 11y ago
    You're upset about questions of merit. I'm just saying it hardly surprises me.
    - coldtea 11y ago
      Parent answered with just the facts. You deduce they're upset from nothing at all. >I'm just saying it hardly surprises me. Which doesn't convey any information -- except your non-surprised status. It doesn't surprise you because it's something that had to be done, it doesn't surprise you because Apple tends to innovate in those areas, it doesn't surprise you because Apple wants to lock people in, it doesn't surprise you because Apple doesn't play well with other's standards, etc, we don't know which.
- xenadu02 11y ago
  When Steve Jobs came back to Apple he did a Q&A session at WWDC. Someone asked him why Apple was moving away from unique (proprietary standards) and Steve said it was pointless to do your own interfaces/standards if the existing ones were good enough and that Apple would only do its own thing when no existing solution would work. Apple created lightning before USB-C existed. In fact I'd wager USB-C wouldn't exist if Apple hadn't created lightning (it costs money to introduce a new standard after all, and the USB-IF is made up mostly of cheap-ass PC makers who will do anything to save a penny... also why USB-C isn't a solid lug like Lightning). Apple created Firewire 400 when USB was stuck at 12. Thunderbolt wasn't Apple's creation, it was Intel's extension of the PCI bus to an external interface. Intel is the one who pushed it as the next great thing and promised to roll it out, then pushed onerous licensing terms and took their sweet time rolling it out to all their chipsets. Even FaceTime couldn't be open-sourced due to patent trolls winning a lawsuit against Apple over the peer-to-peer protocol it used, resulting in Apple having to redesign it to use a server-based scheme (in other words the reason we don't have cross-OS, cross-device video calling right now is due to patent trolls. Think about that for a minute). Why hasn't Apple updated the Thunderbolt Display for Retina? Because the DisplayPort spec can't handle the resolution and hasn't been updated for it yet (at least that was the cast last time I looked). I'll also note Apple contributed the mini-DisplayPort spec back which is why anyone can use it now.
  - jkestner 11y ago
    Good list. In fact I'd wager USB-C wouldn't exist if Apple hadn't created lightning There are rumors that Apple led the design effort for USB-C. http://daringfireball.net/linked/2015/03/14/apple-usbc http://daringfireball.net/linked/2015/03/14/apple-usbc
    - techenthusiast 11y ago
      I discovered through relevant contacts that that's not actually true.
- sjwright 11y ago
  You are probably misremembering because Apple frequently dives into new standards early. Mini- and micro-SIM cards come to mind. (I assume you wouldn't include Windows, or the choices made by Microsoft for Windows as being an "established" choice.) The only substantial "dodged established choice" I can think of recently is the lightning connector over micro-USB. But that's fair enough, as micro-USB is a hateful connector. Paying a few dollars more per cable is relatively inconsequential.
techenthusiast 11y ago
Yes, DCI-P3 is the film industry standard and is part of the UHD specification.
- isomorphic 11y ago
  Right, the article even mentions this: > DCI-P3 will be the gamut to have when UltraHD content rolls around, and Apple choosing it instead of Adobe RGB was a well planned move.
- TD-Linux 11y ago
  No, broadcast and Blu-ray 4k specs specify the BT.2020 color space. Of course, you could argue that neither of those are relevant to an iPad.
  - techenthusiast 11y ago
    The UHD spec supports both P3 and Rec.2020. With SMPTE ST-2084, it's thankfully not an issue at all.
gulpahum 11y ago
Apple doesn't use the exact DCI-P3, they use "Display P3". DCI-P3 has gamma 2.6, while Display P3 has gamma 2.2 (the same as sRGB). I wonder why they didn't go with the correct gamma? Does sRGB gamma compatibility help here so that they can easily display sRGB images on the screen? Or is gamma 2.6 bad on lcd displays?
- johntb86 11y ago
  OpenGL has native support for sRGB blending, so maybe keeping the same gamma allows them to use the same hardware to accelerate Display P3 blending. Though I think sRGB's gamma is technically the much more complicated { cs / 12.92, cs <= 0.04045 cl = { { ((cs + 0.055)/1.055)^2.4, cs > 0.04045 and not exactly 2.2.
- onedognight 11y ago
  It's easy to convert gammas; the worst you get is banding when if you lose resolution and fa r movies this can be done offline, but when two color spaces don't align you'll have to truncate to the intersection where converting making the lager gammuts pointless. So given that nearly all big movies have been filmed and color corrected in P3 for years now, it makes sense to match their gamut, but match common display gammas, so that common sRGB content needs little conversion to be displayed.
- colanderman 11y ago
  Thanks, I didn't know that. Do they actually use sRGB "gamma" (which has a linear region near 0) or do they use a straight 2.2 gamma like Adobe RGB does?
  - gulpahum 11y ago
    I didn't know that Adobe RGB has a straight 2.2 gamma. :) I don't know the details about the Apple's gamma curve, but these guys have measured it. Perhaps it gives some answers. (I'm not sure how to read that graph..) http://www.displaymate.com/Gamma_35.html http://www.displaymate.com/Gamma_35.html
CamperBob2 11y ago
There's an interesting comment on the thread from someone calling himself "melgross" who sounds like he knows what he's talking about. ( http://anandtech.com/comments/10265/understanding-the-97-ipad-pros-true-tone-display/497051 http://anandtech.com/comments/10265/understanding-the-97-ipa... ) Always hard to be sure but he seems pretty credible. The TL,DR version of melgross's comment is that Adobe modeled their RGB color space after the old Kodachrome film gamut, which was optimized for colors found in natural outdoor scenes rather than human flesh tones. As such, it's not ideal for general production use, and many professionals have been unsatisfied with it. Another interesting link from a comment disagreeing with melgross is here: http://www.astramael.com/1 http://www.astramael.com/1
- colanderman 11y ago
  I think there's only shades of truth there. Regardless of where skin tones lie (which generally is within sRGB!), Adobe does cover most of CMYK (whatever failings CMYK may have). Spoony's comment further down, which links to this: http://www.astramael.com/1 http://www.astramael.com/1 is much more on the spot and provides a very interesting comparison of the three with actual examples.

sliverstorm 11y ago

It always seems to me, Apple loves dodging the established choice. Firewire, Thunderbolt, I swear the list is long but I can't remember everything right now.

techenthusiast 11y ago
DCI-P3 actually is the relevant film industry standard here. No consumer displays can do Rec.2020 gamut yet, so P3 is a nice halfway step that is comparable to sRGB, unlike Adobe RGB. And it's "future-proof," because both can use the SMPTE ST-2084 EOTF. Both the P3 and Rec.2020 gamuts are part of the UHD specification.
wtallis 11y ago
Firewire and Thunderbolt both introduced important capabilities that no established choice could even come close to. Apple doesn't wait around for the beige box vendors to come to a consensus on technological advances.
- sliverstorm 11y ago
  You're upset about questions of merit. I'm just saying it hardly surprises me.
  - coldtea 11y ago
    Parent answered with just the facts. You deduce they're upset from nothing at all. >I'm just saying it hardly surprises me. Which doesn't convey any information -- except your non-surprised status. It doesn't surprise you because it's something that had to be done, it doesn't surprise you because Apple tends to innovate in those areas, it doesn't surprise you because Apple wants to lock people in, it doesn't surprise you because Apple doesn't play well with other's standards, etc, we don't know which.
xenadu02 11y ago
When Steve Jobs came back to Apple he did a Q&A session at WWDC. Someone asked him why Apple was moving away from unique (proprietary standards) and Steve said it was pointless to do your own interfaces/standards if the existing ones were good enough and that Apple would only do its own thing when no existing solution would work. Apple created lightning before USB-C existed. In fact I'd wager USB-C wouldn't exist if Apple hadn't created lightning (it costs money to introduce a new standard after all, and the USB-IF is made up mostly of cheap-ass PC makers who will do anything to save a penny... also why USB-C isn't a solid lug like Lightning). Apple created Firewire 400 when USB was stuck at 12. Thunderbolt wasn't Apple's creation, it was Intel's extension of the PCI bus to an external interface. Intel is the one who pushed it as the next great thing and promised to roll it out, then pushed onerous licensing terms and took their sweet time rolling it out to all their chipsets. Even FaceTime couldn't be open-sourced due to patent trolls winning a lawsuit against Apple over the peer-to-peer protocol it used, resulting in Apple having to redesign it to use a server-based scheme (in other words the reason we don't have cross-OS, cross-device video calling right now is due to patent trolls. Think about that for a minute). Why hasn't Apple updated the Thunderbolt Display for Retina? Because the DisplayPort spec can't handle the resolution and hasn't been updated for it yet (at least that was the cast last time I looked). I'll also note Apple contributed the mini-DisplayPort spec back which is why anyone can use it now.
- jkestner 11y ago
  Good list. In fact I'd wager USB-C wouldn't exist if Apple hadn't created lightning There are rumors that Apple led the design effort for USB-C. http://daringfireball.net/linked/2015/03/14/apple-usbc http://daringfireball.net/linked/2015/03/14/apple-usbc
  - techenthusiast 11y ago
    I discovered through relevant contacts that that's not actually true.
sjwright 11y ago
You are probably misremembering because Apple frequently dives into new standards early. Mini- and micro-SIM cards come to mind. (I assume you wouldn't include Windows, or the choices made by Microsoft for Windows as being an "established" choice.) The only substantial "dodged established choice" I can think of recently is the lightning connector over micro-USB. But that's fair enough, as micro-USB is a hateful connector. Paying a few dollars more per cable is relatively inconsequential.

techenthusiast 11y ago

Yes, DCI-P3 is the film industry standard and is part of the UHD specification.

isomorphic 11y ago
Right, the article even mentions this: > DCI-P3 will be the gamut to have when UltraHD content rolls around, and Apple choosing it instead of Adobe RGB was a well planned move.
TD-Linux 11y ago
No, broadcast and Blu-ray 4k specs specify the BT.2020 color space. Of course, you could argue that neither of those are relevant to an iPad.
- techenthusiast 11y ago
  The UHD spec supports both P3 and Rec.2020. With SMPTE ST-2084, it's thankfully not an issue at all.

gulpahum 11y ago

Apple doesn't use the exact DCI-P3, they use "Display P3". DCI-P3 has gamma 2.6, while Display P3 has gamma 2.2 (the same as sRGB). I wonder why they didn't go with the correct gamma? Does sRGB gamma compatibility help here so that they can easily display sRGB images on the screen? Or is gamma 2.6 bad on lcd displays?

johntb86 11y ago
OpenGL has native support for sRGB blending, so maybe keeping the same gamma allows them to use the same hardware to accelerate Display P3 blending. Though I think sRGB's gamma is technically the much more complicated { cs / 12.92, cs <= 0.04045 cl = { { ((cs + 0.055)/1.055)^2.4, cs > 0.04045 and not exactly 2.2.
onedognight 11y ago
It's easy to convert gammas; the worst you get is banding when if you lose resolution and fa r movies this can be done offline, but when two color spaces don't align you'll have to truncate to the intersection where converting making the lager gammuts pointless. So given that nearly all big movies have been filmed and color corrected in P3 for years now, it makes sense to match their gamut, but match common display gammas, so that common sRGB content needs little conversion to be displayed.
colanderman 11y ago
Thanks, I didn't know that. Do they actually use sRGB "gamma" (which has a linear region near 0) or do they use a straight 2.2 gamma like Adobe RGB does?
- gulpahum 11y ago
  I didn't know that Adobe RGB has a straight 2.2 gamma. :) I don't know the details about the Apple's gamma curve, but these guys have measured it. Perhaps it gives some answers. (I'm not sure how to read that graph..) http://www.displaymate.com/Gamma_35.html http://www.displaymate.com/Gamma_35.html

CamperBob2 11y ago

There's an interesting comment on the thread from someone calling himself "melgross" who sounds like he knows what he's talking about. ( http://anandtech.com/comments/10265/understanding-the-97-ipad-pros-true-tone-display/497051 http://anandtech.com/comments/10265/understanding-the-97-ipa... ) Always hard to be sure but he seems pretty credible. The TL,DR version of melgross's comment is that Adobe modeled their RGB color space after the old Kodachrome film gamut, which was optimized for colors found in natural outdoor scenes rather than human flesh tones. As such, it's not ideal for general production use, and many professionals have been unsatisfied with it. Another interesting link from a comment disagreeing with melgross is here: http://www.astramael.com/1 http://www.astramael.com/1

colanderman 11y ago
I think there's only shades of truth there. Regardless of where skin tones lie (which generally is within sRGB!), Adobe does cover most of CMYK (whatever failings CMYK may have). Spoony's comment further down, which links to this: http://www.astramael.com/1 http://www.astramael.com/1 is much more on the spot and provides a very interesting comparison of the three with actual examples.