Is the new MacBook Pro Display 10-bit colour?

[nol2001]

Does anybody know whether the new MacBook Pro Display is 10-bit colour?

I don't remember anyone at the event saying anything like 'it can show 1 billion colours'
The specifications on apple.com don't mention whether it is 8-bit or 10-bit.

 

[leman]

Most likely. It supports the same color profile than the 5K iMac, so I guess its also 10bit? Otherwise the entire 'wide gamut' talk is a bit pointless.

 

[nol2001]

Looking at the tech specs on apple.com it says for 15 inch rMBP:

Simultaneously supports full native resolution on the built-in display at millions of colors and:

• Up to two displays with 5120-by-2880 resolution at 60Hz at over a billion colors
• Up to four displays with 4096-by-2304 resolution at 60Hz at over a billion colors

so this suggests to me the 15 inch rMBP screen is only 8-bit 🙁

 

[maflynn]

so this suggests to me the 15 inch rMBP screen is only 8-bit 🙁

Yes, its my understanding that they only made the colors more vibrant, and the display brighter. Everything else with the display is the same as prior models, including the resolution

 

[nol2001]

I'm worried that they won't improve the screen again for around 4 more years.

I would like:

• HDR
• 10-bit
• Higher resolution (but still 16:10 ratio)
• 120fps

 

[maflynn]

I'm worried that they won't improve the screen again for around 4 more years.

I would like:

• HDR
• 10-bit
• Higher resolution (but still 16:10 ratio)
• 120fps

I don't know about 4 years, but is the market moving to 10-bit? If not then you may not see apple do it. Remember they're focusing this product on the consumer and do consumers really need 10-bit displays?

 

[x-evil-x]

I'm worried that they won't improve the screen again for around 4 more years.

I would like:

• HDR
• 10-bit
• Higher resolution (but still 16:10 ratio)
• 120fps

what do you need HDR for?

 

[SoyCapitanSoyCapitan]

Lots of modern monitors claim 10-bit but aren't. Lots of them claim wide gamut but they aren't telling you which colour profile they are measuring against. Some claim they can show '1 billion colours' even though a 4K screen only has about 7 million pixels and a typically photo has far less than a million colours. Marketing doesn't like to go into these details.

Then there's graphics cards. The driver would need to support 10 bit colour, and only a few Radeon do so on the Mac. None did before El Capitan.

Then there's the backlighting situation. Laptops have a backlight so close to the screen that the mid-tones tend to lose detail and there is too much contrast. There is a reason Eizo monitors are very thick - it's too make sure backlight produces an even soft light so that you can see every detail in a photo.

To understand gamut and palettes please look beyond those silly specs and marketing gimmicks. Visit specialist sites where photographers and print enthusiasts will let you see beyond the hyperbole.

 

[leman]

To understand gamut and palettes please look beyond those silly specs and marketing gimmicks. Visit specialist sites where photographers and print enthusiasts will let you see beyond the hyperbole.

Very fair point. Are you aware of any in-depth reviews of the new MBP 13" screen under these aspects?

 

[SoyCapitanSoyCapitan]

Very fair point. Are you aware of any in-depth reviews of the new MBP 13" screen under these aspects?

Not yet. Normally I don't pay attention to such small screens but felt that the marketing gimmicks surrounding the term 'wide gamut' needs to be addressed so that the public understand what they are being sold/told.

 

[leman]

Not yet. Normally I don't pay attention to such small screens but felt that the marketing gimmicks surrounding the term 'wide gamut' needs to be addressed so that the public understand what they are being sold/told.

Agreed. If the display itself can display a wider color range, but at the cost of significant banding... its not really what one expects.

 

[SoyCapitanSoyCapitan]

Agreed. If the display itself can display a wider color range, but at the cost of significant banding... its not really what one expects.

A DCI-P3 wide gamut display won't show banding but what will happen is the colours will shift if presented with an AdobeRGB image that contains colours that are not in the same range. So for example (rough idea) you shoot a desert, you edit the image on the MacBook Pro screen to adjust certain red tones, but then when you print it out you notice those reds were not what you saw on the screen. Hence why it is important to use a calibrated Eizo for colour work, because then you can soft proof and estimate correctly how the colours will display on print and on other devices.

 

[ihatetoregister]

I'm worried that they won't improve the screen again for around 4 more years.

I would like:

• HDR
• 10-bit
• Higher resolution (but still 16:10 ratio)
• 120fps

Give me OLED. *that* would have been a nice update.

 

[Mr. Wonderful]

I'm worried that they won't improve the screen again for around 4 more years.

I would like:

• HDR
• 10-bit
• Higher resolution (but still 16:10 ratio)
• 120fps

You're already getting WCG and a higher contrast ratio, which is two of the three pieces of the puzzle for HDR. You're also getting a variable refresh rate with the new MBP. That said, I'm disappointed that it won't be 10-bit as well, but it's still going to be a huge upgrade over the prior screens.

Not specifically referring to the above poster, it's funny to see people ask and complain about specs for the new MacBook Pros, yet not understand what they are, and why they want them.

A great example is the complaints about the 2133MHz RAM, when on the 13", it actually makes a ton of sense for a high-performing, well-balanced system for many workflows, due to the 64MB L4 cache. (http://www.anandtech.com/show/6993/intel-iris-pro-5200-graphics-review-core-i74950hq-tested/3)

The calls for OLED screens are similar. In many cases, you're actually better off with LCD screens.

 

[ihatetoregister]

Not specifically referring to the above poster, it's funny to see people ask and complain about specs for the new MacBook Pros, yet not understand what they are, and why they want them.
[...]
The calls for OLED screens are similar. In many cases, you're actually better off with LCD screens.

I HAVE an OLED screen. TV.
Nothing comes even close. Nothing.
In a blind test you would always pick this one. Everyone is just amazed at it.
So I know you were not referring to me in particular, but I know why I wanted it, and I really don't think you're "better off with LCD screens".
Also, much lower power consumption, which could be nice you know, on a laptop.

 

[dallas112678]

Looking at the tech specs on apple.com it says for 15 inch rMBP:

Simultaneously supports full native resolution on the built-in display at millions of colors and:

• Up to two displays with 5120-by-2880 resolution at 60Hz at over a billion colors
• Up to four displays with 4096-by-2304 resolution at 60Hz at over a billion colors

so this suggests to me the 15 inch rMBP screen is only 8-bit 🙁

Over a billion colors suggest 10 bit. 8 bit is 16 million colors.

EDIT: Nvm, was looking at the multiple displays. 8 bit it is then.

 

[dbackeus]

Could someone clarify the relationship between the new displays capability of 25% more color via DCI-P3 vs not supporting 10bit color depth?

I'm trying to understand what possibilities I actually have to use the "wider" colors on my new MBP and currently second leman's sentiments on the topic:

It supports the same color profile than the 5K iMac, so I guess its also 10bit? Otherwise the entire 'wide gamut' talk is a bit pointless.

Ultra HD "HDR" Bluerays are out now with 10bit color depth. What is the (technical) reason for not being able to enjoy this content on a new MBP display vs an external display which would appear to support the exact same color space (DCI-P3)?

 

[Sanpete]

Could someone clarify the relationship between the new displays capability of 25% more color via DCI-P3 vs not supporting 10bit color depth?

I'm trying to understand what possibilities I actually have to use the "wider" colors on my new MBP and currently second leman's sentiments on the topic:

In a few words, color gamut is about the range of colors, 10-bit is about the number of gradations within that range.

 

[dbackeus]

In a few words, color gamut is about the range of colors, 10-bit is about the number of gradations within that range.

That sort of makes abstract sense but still not clear about how it affects my computers ability / inability to make use of the wider colour gamut.

As a concrete example, here is a "Ultra HD" youtube video which has both 8-bit and 10-bit versions available:

When I watch that in the browser then youtube will automatically select the 8-bit version for me. However using youtube-dl (https://rg3.github.io/youtube-dl/) I can manually select to download the 10-bit version of the file. If I then play that file in a good colorspace aware media player such as mpv (https://mpv.io/) and compare it side by side to the 8-bit version the 10-bit version appears to have more beautiful and accurate colours to me (the 8-bit version has a very pastelly smudgy look with slightly more saturation while the 10-bit version appears to have deeper shadows and more natural colors in general).

Now I'm imagining that the 10-bit version would contain enough color information for the software to make better use of the DCI-P3 colourspace on my display, versus the 8-bit version which I'm supposing is standard sRGB...

Does that make sense or is the benefit of the 10-bit version just a matter of subjective opinion (ie, I'm actually not seeing a wider range of colours, just a different experience of contrast / saturation of sRGB colours)?

For reference here are the terminal commands for downloading the various versions:

youtube-dl tO01J-M3g0U -f 315 --postprocessor-args='-strict unofficial' -o 4k-8bit.webm
youtube-dl tO01J-M3g0U -f 337 --postprocessor-args='-strict unofficial' -o 4k-10bit.webm
youtube-dl tO01J-M3g0U -f 308 --postprocessor-args='-strict unofficial' -o 1440p-8bit.webm
youtube-dl tO01J-M3g0U -f 336 --postprocessor-args='-strict unofficial' -o 1440p-10bit.webm
# the reason for postprocessor args is to make 100% sure that any post processing software like ffmpeg doesn't butcher the color profile metadata, shouldn't actually happen with these particular commands since we don't have to merge any audio data but can be good to know for further experimentation

Attempting to play the downloaded files in the web-browser or the VLC player gives a very washed out look, I'm assuming this is due to not respecting the colorprofile metadata in the files, feel free to correct me if I'm wrong about this 🙂

 

[leman]

Now I'm imagining that the 10-bit version would contain enough color information for the software to make better use of the DCI-P3 colourspace on my display, versus the 8-bit version which I'm supposing is standard sRGB...

Do the 8-bit and 10-bit versions of the video use different color profiles? I guess its the matter how the information is interpreted...

 

[evec]

Most 10bit lcd monitor actual is 8bit+AFRC and is not real 10bit panel,
many monitor received 10bit color for better produce work on AFRC and not true 10bit.
also 10bit panel not mean should better, it just mean 10bit and may not accuracy as some good 8bit.

Also, internal panel though eDP, control dithering and FRC directly by graphic chips for better efficiency, so Apple actual provide some class of 8bit + AFRC as like many 10bit (8bit panel+AFRC) monitor in internal panel.

 

[dbackeus]

Do the 8-bit and 10-bit versions of the video use different color profiles? I guess its the matter how the information is interpreted...

Yes, inputting the files to ffmpeg you can see that the 8-bit ones use "bt709" (https://en.wikipedia.org/wiki/Rec._709) and the 10-bit ones use "bt2020nc" (variant of https://en.wikipedia.org/wiki/Rec._2020).

I'm not sure exactly where or how the conversion to my displays DCI-P3 profile takes place (whether in video player / drivers etc). But doesn't it seem logical that a conversion process from 10-bit Rec 2020 video would be able to make good use of the DCI-P3 spectrum while an 8-bit Rec 709 video would not?

 

[leman]

Yes, inputting the files to ffmpeg you can see that the 8-bit ones use "bt709" (https://en.wikipedia.org/wiki/Rec._709) and the 10-bit ones use "bt2020nc" (variant of https://en.wikipedia.org/wiki/Rec._2020).

Ah, well, then this explains it. The BT.2020 has a much wider color space and thus will be shown in more details on a more limited DCI-P3 screen. In contrast, BT.709 offers a narrower selection of colors than what the DCI-P3 display can do, so it is not surprising that it appears bleaker in comparison.

I think its great that software finally starts taking color spaces seriously! It is such an easy way to vastly improve the quality of the graphical content, its incredible that this has been mostly ignored for so long.

 

[Sanpete]

That sort of makes abstract sense but still not clear about how it affects my computers ability / inability to make use of the wider colour gamut.

8-bit doesn't limit the range, i.e. the extremes, of color you can see. @leman is right to point out the different color profiles in the videos. That's what would explain at least the most obvious differences you see in the colors in the two versions of the video. It appears you were watching the videos with the MBP, in which case the different color spaces account for all the differences you saw, because you can't view 10-bit color with the MBP. It's reduced to 8-bit no matter the source.

The 2020 color space is even bigger than the P3 space the MBP uses, so your MBP will show only what it can of that, but it will be more than the sRGB of the 8-bit version of the video. Rec. 2020 in 8-bit has the same color range as 2020 in 10-bit. The color range and the bit-depth are independent.

By the way, as I understand it, Safari is able to handle the P3 color space, but some other browsers aren't. There are webpages that allow you to check that out.

 

[dbackeus]

Ah, well, then this explains it. The BT.2020 has a much wider color space and thus will be shown in more details on a more limited DCI-P3 screen. In contrast, BT.709 offers a narrower selection of colors than what the DCI-P3 display can do, so it is not surprising that it appears bleaker in comparison.

Yeah that makes sense 🙂

Just to be clear though, it doesn't actually appear bleaker. The 709 version appears over-saturated compared to 2020 version. Possibly youtubes encoding process pushes up the saturation in an attempt to retain the "vividness" of the HDR video while downsampling from 2020 to 709. Curious how the 2020 version would look going through an officially supported viewing mechanism on a proper HDR television, maybe slightly more vivid due to even higher brightness etc?

I still have one last subjective test to do - try to play the 10-bit version on an older Macbook Pro without DCI-P3 support and see how that compares.