13" MacBook Pro - 6 or 8 bit color? Specs page True or False?

I don't think there is a notebook LCD panel made that isn't 6-bit. But pseudo 8-bit drive (a.k.a. Millions of Colors) can be achieved by means of something called dithering. There are a lot of different ways to do dithering, some very sophisticated and some not. Apple tends to have very good dithering algorithms. So while all Apple notebooks ever made have had 6-bit panels, they are mostly being used by all levels of professionals to good effect and without complaint.

Including the 17 inch? I'd heard that no one makes an 8 bit LCD smaller than 17 inches, so with no supplier, dithering is the only recourse. But, perhaps that was none smaller than 20 inches?

The 17" is also 6-bit.

The Sony TT line has an 8bit display:

http://www.sony.com.sg/microsite/vaio/products/vaio_tt/8bit.html

Check out this article by Rob Galbraith:

http://www.robgalbraith.com/bins/multi_page.asp?cid=7-10041-10146

While this guy is seriously into the sRGB color space, partly because of his DSLR POV, his tests show Apple at rough parity with the big Lenovo W700, a really great example of what 8-bit TN panels are capable of today, and show the Apple displays closing in on the vaunted Eizo ColorEdge CE240W ⚠.

The 17" UMBP uses (and has since day 1), an 8-bit TN panel, as do the 13.3" and the 15" present-iteration UMBPs. Some say that TN panels are only made in a 6-bit type. This used to be the case, but nowadays, some TN panels are made in an 8-bit variety. The Sony TT line uses 8-bit TN panels. Both the Sony and present Apple UMBP lines uses 8-bit TN panels. These panels are capable of displaying, after correct calibration, accurate representations of both the Adobe RGB 1998 and the ProPhoto RGB color spaces. Both of these are considerably broader than NTSC, which is a standard that's been around since the '50s, and any good color television can accurately and entirely display the NTSC color space since the '70s.

In the recent past, several Lenovo laptops could be ordered with S-IPS panels. These panels also offered true hardware 8-bit color. But the S-IPS screens, unlike the best LED backlit TN panels in the Apple UMBP and Sony TT lines of today, had relatively short '100%' image-quality lives, and were more power-hungry than the presently used TN panels in the Sony TT and UMBP Apple lines. Lenovo no longer offers any laptop with S-IPS panels for primarily these reasons, along with the obvious higher cost of good S-IPS panels. Lenovo does, however offer a great 8-bit TN panel in its gargantuan W700 (top-option choice), too.

In addition to all 3 of the presently offered Apple UMBPs, I own one of these Lenovo beasts for dedicated Windows clients. I strongly prefer the Apple hardware and software, and have for 20 years, for all the obvious reasons ;^).

Welcome to the discussions, polymath consulting.

Thanks for your post. I had thought that the increased gamut MIGHT have something to do with the increased color resolution of the screens. If this data is correct, that would explain it.

Very interesting article, polymath consulting. I'm glad to hear the new MBPs are comparing so well to industry benchmark displays for color accuracy. That's great. It just goes to show how little the 6-bit/8-bit difference matters against all other things that matter when dithering is done well.

The article you reference completely skirts the 6-bit/8-bit issue. I stand by my original guilty-until-proven-innocent claim that Apple used 6-bit panels in their notebook computers, not 8-bit panels. I have seen no hard evidence to the contrary and find plenty of evidence in favor of the 6-bit claim.

Ok by me. I just want to throw a few things into the ring, though.

One, there is no possibility of a 6-bit TN panel being capable of displaying ProPhoto RGB color space accurately, or even completely. The Sony and the Apple UMBPs (present-day) all accurately render the ProPhoto color space when calibrated correctly. This isn't my opinion- it's physically impossible for any 6-bit screen to display that broad a color gamut, period. The top-notch 8-bit TN panels made today aren't prohibitively expensive, nor are they particularly hard to procure, nor are they unduly power-consumptive; the reasons for using 6-bit TN panels don't exist for Sony and Apple- it just doesn't make sense to use inferior panels, when the kind of performance we see today in the newest crop of UMBPs is ONLY possible using 8-bit TN hardware, not 6-bit.

I say this withot having a panolply os studies handy to point to on the Web, nor any lengthy experiments from which to draw originating in my mountaintop lab- I DO have 20 years' experience in graphics, printing, digital imaging and digital video, and know which hardware is capable of rendering various industry-standard RGB output accurately- my livelihood depends on it, I charge upwards of $300.00 an hour to clients like Harvard, Middlebury College, Lonely Planet, Smith College Department of Digital Imaging, and the Yale School of Art for lending my expertise in printing archival gallery prints from digital files captured with everything from a $100k Leaf to an iPhone. There would be every reason for Apple to stick with a lower-cost, lower-quality 6-bit TN panel, if it were possible to even to render the much narrower color gamut (compared to ProPhoto RGB), of Adobe RGB 1998, most often still the seed RGB color space conversion of choice, from an additive color space (RGB) to a subtractive color space (CMYK), for serious commercial CMYK printing. The fact that 6-bit TN panels are physically incapable of accurately displaying either ONLY the entire NTSC , color space, most of the sRGB color space, probably less than 70% of the Adobe RGB 1998 color space and not more than 60% of ProPhoto RGB color space is what I've found in my experience. When one examines ProPhoto RGB and Adobe RGB 1998 color files on 6-bit TN panels, even professionally hardware-calibrated ones, posterization in the mid-tones, clipping in the highlights and blocking up of the shadows ramps to black are present, just to name a few of the sure signs of a 6-bit display trying, unsuccessfully, to render a color space beyond its gamut capabilities. When these identical files are displayed on properly hardware-calibrated 8-bit hardware panels, whether TN, S-IPS or any number of the current hardware 8 to 12-bit panel technologies, they are beautifully rendered onscreen, very accurate, and print, when a proper CMS is deployed across all the display, input and output devices in the printing chain, with a realism and depth ONLY possible in the broader color gamuts that are beyond the capabilities of 6-bit display hardware to obtain.

I've training countless clients how to obtain top-notch high-end printing results with the Apple UMBP 17" laptops, and now the present-day 13" and 15" UMBPs as well, because much of the printing of even medium-level comping requires Adobe RGB 1998 and ProPhoto RGB capable displays, scanners, printers, digital SLRs, etc., so the actual color-rendering capabilities of Apple laptops are crucial for me to know BEFORE I go and train clients how to obtain the results they require to do their jobs.

Stick with the 6-bit theory if you like- and I agree, there seems to be more than a few allegations of 6-bit TNs being used in the Apple laptops in years past which were absolutely true, including some models of iMacs. But, with regard to the presently shipping Apple UMBPs- all of them- my experience with real-world files, real-world clients' printing needs, and many years of developing professional color-management systems for gallery-quality archival printing in both RGB and CMYK, along with extensive higher-end CMYK commercial work with any number of publishers proves, to me and my clients, anyway, garbage in=garbage out. The days of Apple reverting to the PC practice of crap components in their laptop displays is now over, once again. Whether they stick to 8-bit display hardware, well- let's hope so! ;^)

Best,

Charlie

I must say that I also missed any specific reference to an 8 bit panel, BUT from your description, that's because it was inferred from the fact that a 6 bit panel could NOT display the gamut defined on that page. So, either Apple has found a magical valley filled with 6 bit panels that can display a gamut that physics doesn't allow, OR they're using 8 bit panels.

Thanks again for your insight!

8 bit displays are only 256 colors.
16 bit displays are 32,767 colors.
24 and 32 bit are millions of colors.
If you want to really get picky, open a colorwheel in a graphics program such as GraphicConverter or Photoshop with only 256 colors displayed. You will actually be able to see the individual 256 colors in blocked arced regions of the wheel. Switch to 16 bit, 32000 colors, you'll see a smoothing effect between each color in a much narrower band of colors. Switch to millions, it is indistinguishable which color is which. It doesn't matter if you use a MacBook, Powerbook G3 or better, or a G3 desktop or better Mac. All support millions built-in. This dithering effect may be more of a colormatching issue, than an actual pallete issue. The colors are there. They are more likely picking from a different hue range based on different colormixing rules. You can lookup color in the http://developer.apple.com/ guidelines.
So no, Apple isn't lying. The pallete table though is different from PC to Mac. And hence why the Color in the Displays System preference offers different color palletes to choose from.

I think the 8-bit here is in relation to the number of bits per channel per pixel. For older panels, they used 6-bits for each Red Green and Blue. In order to "fake it out", the panels would perform dithering. These newer panels use 8-bits for each Red Green and Blue, so it actually produces the range of colors without requiring dithering.

Quoting Kyn Drake:

+So, either Apple has found a magical valley filled with 6 bit panels that can display a gamut that physics doesn't allow, OR they're using 8 bit panels.+

There is a conceptual error in this thought, and Charlie is making the same mistake. The amount of color space (gamut) enclosed by a color rendering device, like a display, is defined by endpoints, not points in between. You could have a display that rendered only 1-bit per color channel and it could still encompass the same gamut as an 8-bit, or 10-bit, or n-bits per pixel display. The difference would be in middle tones; all those in-between colors. But as far as the color gamut goes, the red would be just as saturated, and so too green and blue, so the gamut, namely the endpoints of the color space, would be identical regardless of the number of bits per pixel.

So there is no physics that interferes with the notion of a high gamut display with a low number of bits per pixel. Again, that would only affect the number of shades or, if you prefer, the quality of gradients within the color space, not the absolute size of the color space enclosed, which is the device's gamut.

What I understood from Charlie's post is that he has witnessed, with measuring equipment, that the panels currently being used by Apple have both a wide color gamut PLUS the ability to display all the intervening midtones.

Of course I'm no color expert, and, since there appears to be no one that has a part number which references a specific data sheet which says one way or the other, it makes sense to me that if one were to make measurements that appear to support the premise of an 8-bit panel, then we've either got an 8-bit panel or an extremely cleverly designed 6-bit panel that, in every way, matches the quality of an 8-bit panel.

Either way, I enjoy the discussion, even if I may be misunderstanding bits of it, I'd rather be set straight here 🙂

Well... still no official answer from Apple... I found this link about the same subject... Is the assertion (based on the rendition of the pattern) correct or not ?

On my MB Air (1st revision from 2008) i do see the line (and even another horizontal one about 20% from the bottom of the pattern)... though i calibrated my pannel with an X-Rite Eye-one display tool

On my iPone 3GS i do see the line...

I'll check tonight at home on my DELL 2405 display (which is a PVA or MVA pannel... that, i assume is a 8-bit per channel display)...

Keep you posted...

Message was edited by: Gavroche1973