Slashdot Mirror
Apple Sued Over 'Lacking' Macbook Display
qu1j0t3 writes "Business 2.0 reports that two MacBook owners have filed a class action lawsuit charging Apple with deceptive advertising, as well as misrepresentation and unfair competition over the use of the phrase 'millions of colors' to describe the capability of the LCD displays in MacBook and MacBook Pro computers. The article likens the complaint to an an angry forum thread, and is more than a little bit skeptical of the plaintiff's motives. Perhaps it's their uncanny attention to detail. From the filing: 'The reality is that notwithstanding Apple's misrepresentations and suggestions that its MacBook and MacBook Pro display millions of colors, the displays are only capable of displaying the illusion of millions of colors through the use of a software technique referred to as dithering, which causes nearby pixels on the display to use slightly varying shades of colors that trick the human eye into perceiving the desired color even though it is not truly that color.'
If this goes through then it opens the door to lawsuits against nearly all LCD manufacturers, since most displays these days (except for the highest end ones) are 6-bit with dithering...
This guy's the limit!
Your average display can only show 768 unique colors. However, our eyes blend them so that you see white instead of a combination of red, green and blue. Some displays support 16-bit shading, but I don't know if that's available in an RGB format or just a black-and-white format.
And while "dithering" is usually used to refer to pixels, as opposed to subpixels, the same principle applies.
This result of this case will have less to do with the technical merits of the display, and more to do with common practice and practical use.
tasks(723) drafts(105) languages(484) examples(29106)
How about providing the IMPORTANT part from the article in the summary, hmmm?
"At the heart of the case is plaintiff's claim that rather than delivering 16,777,216 colors with an 8-bit LCD, Apple chose a cheaper route, delivering the illusion of millions of colors using a 6-bit LCD and dithering."
2^24 = 16,777,216
2^18 = 262,144
Nothing wrong with 6-bit LCDs, but they shouldn't be advertised as 8-bit...
*gets ready to be torn apart by rabid mac fans*
ClutterMe.com - easiest site creation on the Net. Just click and type.
You're focusing on the software end. It's the hardware that isn't capable of displaying 16.7 million colors, although you'd be very hard-pressed to tell (I can only see slight banding in very specific cases on my 6-bit LG monitor).
This guy's the limit!
I was always surprised how bad the colours on my MBP look compared to the same colours on my 19" CRT. Now I know why they look so bad. Here is a better article about this case. Both articles don't mention iBooks and Powerbooks. Do they use 'normal' screens? I had an iBook once and I always thought the iBook had a (much) better screen than the MBP.
-- Cheers!
I mean sure that level of pedanticism is tolerated in some forums, but this is a court of law.
I'd say the other way round - being pedantic is more important in legal matters and advertising, than it is on a forum.
And this isn't being pedantic - surely it's long established that "number of colours" refers to the number of possible colours an individual pixel can display, and not using tricks like dithering? Otherwise, back in the 80s/90s when computers only had 256 colours or less, why didn't we see manufacturers claiming they could actually display thousands of colours? Why weren't the computers which could display thousands of colours back then advertised as displaying millions? And maybe the original black and white classic Macs should actually be greyscale, because you could dither the black and white pixels?
I'm curious, as I thought 24 bit displays had been standard on computers for well over a decade now - is it common for laptops to have an 18 bit display, or is it only Apple that have decided to take us back to the 1990s?
really? funny how I can see color banding when you look at photographs of blue skies on a 16,000 color display but you dont see the color banding when you go to a 24 bit color display.
you eye CAN see more than 16,000 colors. espically when it is looking at 3-4 colors all next to each other to show off color banding.
Do not look at laser with remaining good eye.
I have bad news for you. Your Powerbook has 262k colors.
The MacBook displays weren't a regression in quality, they were a continuation of an exceptionally long-standing practice.
Maybe we'll see 8-bit displays in some future revision, but until now, it's been a cost issue, and Apple was smart enough to realize that if the entire industry is using 6-bit displays, they can do it too.
I just checked the MacBook specs, and saw this under display: "13.3-inch (diagonal) glossy widescreen TFT display with support for millions of colors"
What exactly does "support for millions of colors" mean, anyway? In the world of (E|H)DTV monitors, "supports 1080i" generally means "can display a 1080 image, but only at 768" or somesuch. I look for words like "native resolution" to figure out what something is technically, actually, capable of.
And if Apple can show that EVERYONE in the industry is doing exactly the same thing, with similar advertising language, then it's probably not going to go anywhere. It's sort of a visual equivalent to the silly GB vs GiB argument, though at least in that case hard drive manufacturers have started better explaining their side of the equation....
From the PDF ( http://www.engadget.com/videos/PDF/apple_macbook_l awsuit.pdf ) :
''The extent to which a particular make of computer is capable of "dithering" is a function of the sophistication of the programming of the software. For example, in the case of the MacBook and MacBook Pro, because of the uniqueness of these computers to be able to run both Apple's OS operating system, and the PC's Microsoft Windows operating system, it is possible to compare the quality of the display between the two operating systems. In the case of the display that the MacBook and the MacBook Pro produces using the Microsoft Windows operating system at all resolution levels is superior to the display that those same computers display using Apple's OS operating system.''
Some nasty grammar in there, but in summary: no such problem when running Windows on the same machine?
If that is true, then it is indeed an Apple software problem. Note that software shouldn't be in charge of this sort of thing in the first place. LCD displays themselves handle incoming 8bit values on a 6bit displays in one of three ways:
A. 'as is', 91 becomes 92.
B. 'dither', 3 out of 4 pixels are 92, the other one is 88, averaging to 91
C. 'frame rate control', 3 out of 4 refreshes it draws the pixel as 92, the other one is 88, averaging to 91.
B&C are both common, and both have pros/cons. But either way, the software shouldn't be doing anything there (arguably, a driver might - i.e. if the monitor specifically allows you to specify which method to use, what dithering pattern, etc. by means of driver control).
D'oh! I got teh numbers wrong: the human eye can discern about 350,000 colors (warning: MS Word file).
My point, though, was that it's a silly lawsuit. As someone who spends hours in Photoshop doing color correction I know the monitor is, at best, a blunt instrument. That's why we have matchprints and digital color proofs and the like.
I am a believer of momentum and curves.
But dithering only works well at minimum distances, and it can still create problems with gradient lines, if the algorithm is of sub-par quality or the picture happens to have something in it's nature that doesn't quite dither properly.
Dithering can add a 'blur' factor, since it's not an exact representation, which also causes a loss in image accuracy.
and as a counter point to all my previous points...
Aren't all images on computers dithered anyway, since each "pixel" is actually 3 sub-pixels, each with a different color?
34486853790
Connection too slow for X forwarding? Try "ssh -CX user@host"
This got me wondering how many bpp my own Viewsonic Pro series monitor can display. I was surprised to find that it wasn't listed in the product specifications -- neither as bits nor total number of colors.
A little further digging brought me to this article which gave some good insight about the differences. Some highlights:
The displays are only capable of displaying the illusion of millions of colors through the use of a software technique referred to as dithering, which causes nearby pixels on the display to use slightly varying shades of colors that trick the human eye into perceiving the desired color even though it is not truly that color.
Isn't that going to get laughed right out of the courtroom? I mean sure that level of pedanticism is tolerated in some forums, but this is a court of law. Surely the judge is going to say a dignified version of "What the fuck are you on? Get the fuck out of here."
Check this image:
squares
Imagine you're told your new expensive laptop, by *Apple* (a brand mostly known as going for component quality since designers work with it), will have have display quality like the second square.
But instead you see display quality like the first square.
If you think it's a laughing matter, I want to see you persuade those unfortunate Mac owners join the laugh.
Except that it's a 6-bit display. 2^(3*6) is a quarter of a million.
Also FatPhil on SoylentNews, id 863
The dithering done on 6-bit LCD panels is in the time domain. A pixel will flicker between two different shades at a frequency high enough to be almost invisible, creating the illusion of a shade in between. (I say "almost" because some people can see the flickering, including me. It's easier to see if your eye is moving around the screen instead of staring at a point.) The 256-color displays of days gone by dithered in the spatial domain, so their dithering was always visible. The only way it created the illusion of continuity was if you sat far enough back that you couldn't see the individual pixels.
It's an interesting distinction that I'm not sure how it would hold up in court. I should point out however that many light sources we think of as continuous do the exact same thing to produce the illusion of continuous light output. Fluorescent lights, lights on some new cars, the backlights on many cell phones and PDAs, all of them flicker.
The vast majority of LCD panels are 6-bit, and use dithering to generate 16.2 million colors. True 8-bit panels are usually fairly expensive, and only used on high end LCDs designed for graphics work. The fact that you hadn't noticed this is a pretty good argument that this type of dithering isn't really false advertising.
From one of the comments below TFA: "Out of 28 notebook LCDs manufactured by Samsung, only 2 can display 16.7M colors natively, a 15.4-inch panel with a lowish resolution of 1,280 x 800 (part number LTN154X5) and a 19-inch panel (part number LTN190W1). The rest, 26 LCDs, are 6-bit and can display 262,144 colors natively, without dithering, and millions of colors with dithering. [...] At LG.Philips, all of the 15 notebook LCDs are 6-bit and can display 262,144 colors natively, without dithering, and millions of colors with dithering."
So it seems virtually no laptop LCD can display 16.7M colors without dithering. It's a problem which affects the whole industry, and all laptop manufacturers seem to be, well, somewhat "optimistic" in their advertising claims.
Which doesn't make it better that Apple does so, too, and as far as I'm concerned, the suit is well justified.
no, a CRT uses variable intensity for each of the three colors, and each pixel can be one of ~infinite colors on an analog display
Snowden and Manning are heroes.
WTF? Pantone (and all printing) is about ink colors .. subtractive color, not additive like your monitor. Your screen was never and will never (with ANY current technology) show you printed colors accurately, it just approximates. What's more, 99% of Pantone colors are out of both RGB and CMY(K) gammut. Products such as Photoshop and Illustrator have RGB conversion libraries supplied with them to approximate Pantone colors in RGB on your monitor.
Printing in spot color is all about ink recipes. Your printer will have a Pantone book that tells him exactly what weight of what ink he needs to _add_ together to create the color you asked for. And that requires a LOT more than just cyan, magenta and yellow.
Feel free to complain about Apple, or anyone else's LCD monitors, but don't get additive and subtractive confused. And never expect to see Pantone colors accurately on your monitor. Buy a Pantone color chip book if you need something to show your clients.
Take a look at the huge difference between the specs on the same displays from their US site and European site. The European site has the actual specs listed. Apparently something about false advertising was preventing from misrepresenting what they were trying to sell. (The European site doesn't even attempt to mention "16.7 million colors" for some of the displays!)
Parent poster describes how with a restricted number of colours, though the human eye can't see banding in the original, if you then process that image the quantisation of colours in the original can lead to banding in the processed version.
Yes, but that has nothing at all to do with this story.
There would be an issue if a 24-bit image was downsampled to 16-bit (for example) in order to display it on a 16-bit screen and then resaved (or processed) at the lower bit-depth. But it's not. Any transformations done in the image editing program are at the bit depth of the image, not of the display.
So if the eye can't see the deficiencies of the display before manipulating the image, it won't see them afterwards either.
Well, the image sent to the display is handled internally as 24bits for well over a decade as you pointed it. With CRT displays, this was correctly handled since this is an analog display.
With LCD, another layer of digital conversion has been added. Even if the computer is handling the screen buffer in 24 bits internally, LCDs have starded from the good old 256 colors (8bits) then progressively switched to 4096 (10 bits) and 65k (16 bits).
Most LCDs are currently 18bits (6bits per channel) as stated in others comments. To display more than the 262k colors available, you can use temporal and spacial dithering. Spacial dithering is when you display a 50% gray next to a 52% gray when you want to display a 51% gray. Temporal dithering is when you display a 50% gray then a 52% gray then a 50% gray, etc... to display the same 51% gray.
This dithering can be handled at the application level, the OS level, the graphic card driver level, the graphic card hardware level or at the panel electronic level. In fact, this might be handled at all those places at the same time with varying results :-(
Sig (appended to the end of comments you post, 120 chars)
VGA and DVI claim to deliver 8 bits of precision per channel; these Apple LCDs are only capable of displaying 6 bits of precision. Yes, there is loss in any computer display due to the nature of color triad-based pixels. This problem goes beyond that, and introduces unexpected dithering artifacts into the image. It is entirely possible to create images that look great at 6 bits per channel, but since the downsampling is done at the display rather than by the system, and because the user was not expecting the reduced precision, it oftentimes looks like crap (quick example I've seen is the top toolbar in Firefox; the slight gradient doesn't look pretty when it gets dithered).
"Because Science" is one step from "Because old book". Try "Because of my experiment testing my falsifiable assertion".
This is not the case of all 6-bit panels. Some use actual dithering while others use FRC (Frame Rate Control), which is what you described.
Nitpick, and you shall be nitpicked yourself:
God is a smiter.
God is about to smite Bob.
God is smiting Bob.
God smote Bob.
God has smitten Bob.
Bob has been smitten.
The adjective is always the past participle.
You may want to revise your numbers again.h t.pl?kw=&file=answers%2Fstructure%2Fans_011.html
http://ask.yahoo.com/20041227.html
http://www.hhmi.org/cgi-bin/askascientist/highlig
On my MacBook, which should have a 6-bit display, the left and right squares look quite different to me. I believe that's a good indication that the time-based "dithering" used on the MacBook is not nearly as bad as space-based dithering, at least for people who are unable to see flicker significantly above 60 Hz.
Though it seems hard to believe, there is some chance that Apple does have a serious mistake. There is an explanation as to why it might be "better under Windows" and it is not good for Apple. A sure way to compare would be to take a screen shot from one system and display it on the other, so that differences in graphics and font rendering do not enter into it.
Apple has had a history of using a gamma correction table, which was always a mistake. I thought they eradicated this in OS/X but perhaps it lives on. Some ill-informed people actually think this makes the image better but it is always a bad idea on current hardware.
The reason is that the hardware interface to the monitor is 8 bits (per channel). If you have an 8-bit-per-channel image, and the gamma correction table is anything other than 1:1, then two or more different 8-bit shades are going to get mapped to the same 8 bit number sent to the monitor, due to the pigeonhole principle. It also means some possible 8-bit outputs are not going to get produced. It is possible the diterhing of the LCD is amplifying this effect. For instance if many of the "pure" values are the missing ones, then there is going to be far more dithering.
Both Windows and Linux just dump the 8 bit images you send to the graphics api to the screen buffer with no change. Though this sounds more primitive, it turns out it is the right thing to do. Color correction and profiling has to be done by software, not by hardware and drivers.
Look, if you want to get all big-dick-swinging about this, send me your client list, a link to your portfolio (including samples from all of the international ad campaigns you've worked on and the awards you've won) and your fee list and we can see who's the biggest, baddest, most calibratingest knowingest motherfucker of them all. If you don't want to do that, just try to be less of a dick when you post.
I've been doing this since Photoshop 1.0 (and PageMaker and Illustrator 88. . .) so I am sure of where I speak. Am I the bestest retoucher in the universe? No, because no matter how good you are, there's always someone better. The best guy I've ever worked with was a portrait painter in a former life (he painted the portrait of Reagan which hangs in the White House) and working next to him was a revelation: he could paint photorealistic images in Photoshop with the same effort I take to tie my shoes.
That aside, no monitor in the world will accurately show you low values of yellow and cyan. Sure, something will be on the screen, but to really check and make sure your whitest areas are 2/2/2/0 (or 5/5/5/0 or whatever standard the shop you work in uses for non-specular highlights) you need to use both the info palette and a matchprint/high end digital color proof. That's just the way it is, and all of the people I've worked with in all of the years will say the same thing. Trusting the monitor--any monitor--is asking for a surprise when things come off the press. It's just the nature of conversion from additive to subtractive color models.
I am a believer of momentum and curves.
The more I look, the more values I find. I think we can agree on three things:
1) My original quote of 16,000 was wrong;
2) The human eye can see many more colors, between a range of 350,000 and several million;
3) Only the Flying Spaghetti Monster knows what the real number is.
I am a believer of momentum and curves.
re:"If the human eye thinks its getting shown millions of colors, it is."
Not so much. Depends on what you're looking at. If I use the gradient tool in Photoshop to create a smooth field of color - LCD screens (all LCD screens I've seen to date - including my Apple Cinema Display 22") I get banding (albiet less severe because of higher rez and better dithering) that I haven't seen since the early 90s. Now - if Apple were clever - they could translate "support for millions of colors" to tube-based external displays which the MacBook does. In fact I use mine closed (to preserve screenlife when it's not on my lap) hooked up to a tube monitor in offices that I do design work for. One client has one of my old monitors and others have piles of them that accumulated when they were replaced by LCDs.
So the big question is - if you're hot to trot on color - why aren't you lugging around a 60lb Sony Trinitron like I am?
Pussy.
EVERY manufacturer who advertises 16.2 million colours uses a 6-bit display and dithering. Apple has a setting of "millions of colours" in their preferences, and they stuck with that description when they used a 6-bit display.
:-) and it's bright-red. Now apply a pulse-wave-modulation to that LED, and you will be able to *see* a smoothly-varying intensity between 0 and 1 (black and bright-red) even though the LED is only switching fully on or fully off. The pulse-train is controllable by a digital system, so you can electronically vary the effective brightness of this 1-bit system.
The point is that this is temporal dithering, not spacial dithering. There's no point in saying "my eye can see the difference when there's dithering in an image" because that's not what's taking place.
Consider a nominally 1-bit system, a single red LED. Apply zero power, and it's black. Apply constant 5v (with appropriate resistors
In this, the 1-bit output is temporally dithering its on/off state to give the illusion of a multi-bit system. Scale this up to a 6-bit system, and it's easy to generate the illusion of an 8-bit system. To the human eye there is no difference, we don't have the refresh-rate to catch the LED off or on, we just see the aggregated results of very fast controlled flickering.
Simon
Physicists get Hadrons!
Search Wikipedia for 'tetrachromatism' for more info.
Sick of WoW? Try the thinking man's MMORPG: EVE Online
for your clicking convenience...
The human is a blocked tetrachromat
Tetrachromacy
If you mod me down, I shall become more powerful than you could possibly imagine.
"the Apple monitors would be counterproductive at best."
You mean the notebook monitors, right? Because the Apple Cinema Displays are quite good and are even SWOP certified for your soft proofing needs. For LCD's this is pretty damned good, but you do pay for it...
"The greatest obstacle to discovery is not ignorance - it is the illusion of knowledge." - Daniel Boorstin
The left square has banding, the right square does not.
This is on my MacBook Pro display, btw.
Episode 137: Chain of Command, Part II
"What use is power to the Keeps of Balance?" -Disnt of Nightmare LpMud
(By the way, who made you king and gave you the power to decree how conjugations shall be demonstrated, yea, even unto the end of time?)There is no conditional in "God is about to smite Bob". There is also no passive there.There is nothing wrong with using two prepositions in a row. Furthermore, "God will smite Bob" does not mean the same thing as "God is about to smite Bob" - the latter implies that the smiting will happen in the very near future, while the former merely implies that it will happen at some point in the future.Um, no, it doesn't. There is no passive involved. You don't actually know what a passive is, do you?He did conjugate the verb, and he didn't turn it into a modifier. And "God smites Bob" does not mean the same thing as "God is smiting Bob": the latter emphasises the fact that the smiting is an ongoing action at the present moment, while the former merely specifies that it happens without making any real statement as to when (are you saying "God smites Bob every Thursday", or "Here is God. God smites Bob. See God smite"?)No, it is not a passive.Where do you get these ideas?Congratulations! You have successfully identified a real passive. That's one out of four, which I'm afraid is not a pass mark round here.
(BTW, the question of who smote Bob can generally be inferred from context: something like "God is on a rampage. Bob has been smitten, and so has Fred" is sometimes better style than "God is on a rampage. He has smitten Bob, and Fred too".)The irony is killing me.
(Cue half a dozen posts telling me that I'm misusing the word "irony". Come on, don't disappoint me here!)
The method of arriving at 16.2 million colors is actually simple.
...). The interpolation method is usually capable of representing 3 values between each color, but because the numbers on the end (... 244, 248, 252) are missing the final value, 255, there is no final value to dither in. So those final 3 values (253, 254, 255) are missing. Thus, you only have 253 values for each dithered component (since we include 0). 253 * 253 * 253 = 16194277 ~= 16.2M.
On a true 8-bit display, the value range for each component is from 0-255. Because the 6-bit display can only display 1/4 of those numbers (0-63), it must dither for the rest in the form of (0, 4, 8, 12, 16
It's really stupid. Just say 262144 colors.
http://www.thinkwiki.org/wiki/TFT_display#Flexvie
TO START
PRESS ANY KEY
Where's the 'ANY' key? I see Esk, Kitarl, and Pig-Up...