iPhone 4's "Retina Display" Claims Challenged 476
adeelarshad82 writes "Of the many things that buyers might need to know about the new iPhone, Raymond Soneira — president of DisplayMate Technolgies — added one more to the list. Soneira challenged Apple's claims that Apple's new iPhone contains a so-called 'retina display.' According to Soneira, the resolution of the retina is in angular measure, 50 cycles per degree, where a cycle is a line pair, which is two pixels, so the angular resolution of the eye is 0.6 arc minutes per pixel. So, if you hold an iPhone at the typical 12 inches from your eyes, that works out to 477 pixels per inch. At 8 inches it's 716 ppi. You have to hold iPhone 4 out about 18 inches before it falls to 318 ppi. So the iPhone has significantly lower resolution than the retina."
The Question Is: +1, Helpful (Score:1, Informative)
Will the iPhone AND Apple ( The Company)
BLEND [youtube.com]?
Yours In Anchorage,
K. Trout
higher res description (Score:5, Informative)
Re:This looks like a typical straw man argument. (Score:3, Informative)
>>>Your retina doesn't even have pixels!
Yes it does. It has light sensitive spots which can be considered the equivalent of pixels (picture elements), same as a CCD has. True the eye is biological and the CCD is mechanical, but the basic principle is the same..... millions of these pixels make-up the image we see.
Re:I agree, *however* (Score:3, Informative)
No offense but you have issues. Either your integrity or your dexterity are in serious doubt.
Re:bad vision (Score:3, Informative)
Recommended by my eye doctor for one: he was adamant that keeping my point of focus at least 20 inches away, and looking into the distance at least once every 20 minutes, is important for keeping the muscles that focus my eyes healthy.
Re:I agree, *however* (Score:3, Informative)
Either that or he insists on using a third-party mouse with inadequate driver support for Mac OS X, so what he thinks is helping him is really causing his problem.
Maybe for perfect vision, but not for 20/20. (Score:5, Informative)
For a human eye with excellent acuity, the maximum theoretical resolution is 50 CPD[32] (1.2 arcminute per line pair, or a 0.35 mm line pair, at 1 m).
...A resolution of 2 arcminutes per line pair, equivalent to a 1 arcminute gap in an optotype, corresponds to 20/20 (normal vision) in humans
If my math is correct then this is 60% worse than the 'excellent' eye; so the figure of 477 ppi at 12 inches is 286.2ppi; so well within the retina display's capability.
Re:In other news... (Score:5, Informative)
3. Nor have you ever played Nethack [wikia.com]
Re:I agree, *however* (Score:5, Informative)
No offense but you have issues. Either your integrity or your dexterity are in serious doubt.
I'm simply a more distinguishing user. Try the google search below. Note: I develop OSX kernel extensions and I'm writing this from the WWDC right now - Apple broke the API's all of the "fix" programs you will find below use to try and fix the acceleration curve.
http://www.google.com/#q=mac+mouse+acceleration+fix [google.com]
Focal distance (Score:5, Informative)
"Yeah, that's the thing: You can't really talk about this sort of issue with pixel density alone. You can only talk about it as a function of both pixel density and viewing distance."
No, actually it's possible to simply say that the human eye cannot discern individual pixels. Just like we can't discern individual molecules, no matter how close we hold the object to our eyes. There is an average minimum focal distance for the human eye, and if the object is held closer than that to try and discern more detail then it will become out of focus. If the DPI exceeds the human eye resolution at the typical minimum focal distance then the claim is valid.
Re:Slight Misfire above.... (Score:5, Informative)
Also remember, the colour resolution of the eye is far poorer than the b&w resolution of the eye, and the aim here is about colour.
I'm not entirely sure what you mean, but the fovea responsible for your "high resolution" sight contains almost exclusively cones, which are colour sensitive. Most of them detect red and green light, so the resolution in monochromatic red or green isn't that far below white light.
The rod cells on the other hand can only distinguish between black and white, but they are much sparser giving significantly lower resolution. (Their advantage is that they are extremely light sensitive, almost down to detecting a single photon. This is why you have no colour vision when it is dark. Another interesting consequence is that you are blind in the center of your visual field when light conditions are bad, since the fovea lacks rods.)
Re:Real Ratina Display (Score:5, Informative)
For what it's worth, the highest resolution photographs are typically printed ~300ppi. This is the standard used by glossy magazines (Playboy is the canonical reference mag here). Higher than that, most people don't see any difference at all.
Years ago I remember reading a study on this that claimed most people could not reliably differentiate between images printed above 280ppi when asked to pick the image with more detail. However, a significant fraction of people were able to differentiate higher resolutions when asked to judge things like: "which image seems to jump off the page and seem more 3d?"
I don't buy Apple because I don't support their need to own the entire hardware and software stack. However, I'm thrilled that they've put out the first device with a screen that is this hi-res. I hope that by this time next year, there are no phones made with screens under 300ppi.
Re:This looks like a typical straw man argument. (Score:5, Informative)
Yes, but the density, color sensitivity, and light sensitivity vary across the retina. The eye has a nice hack where the high resolution is only on-center, and we point that spot at whatever interests us. The total "pixels" on the retina are far smaller than the on-center resolution would suggest.
Also, displays in general do a remarkably inaccurate job of rendering colors, they just choose colors that our eyes see the same as the originals (but a species with cones centered on different frequencies might think out displays odd). Most absorbtion spctra, emission spectra, and the ends of the monochromatic spectrum can't be displayed, but what is diplayed looks right in all but the last case (which has annoyed many a physics professor - you simply can't put an accurate spectrum on an electronic screen).
Re:This looks like a typical straw man argument. (Score:5, Informative)
Sorta, but the eye's color sensing mechanism works on an opposing color system because the biological pigments in the cones of the retina don't just respond to one frequency of light, they have a bell-curvish response centered on one frequency, and the curves overlap. The M and L cones almost entirely overlap, while S cones are way off in the blue region, though light that only stimulates S without any M is typically seen as a violet color. When you "see" green light, it just means the M cones are stimulated more than the L cones, whereas deep reds will trigger more of the L cones, but also some M cones.
What most people think of "pure" green is right around where the response curves for M and L meet in the middle. Yellow is where light peaks on the L cones but is still stimulating the M cones about half-strength. Both L and M overlap S on one tail end, on the other there is a very tiny range where the L cones are the only ones responding and that color is interpreted as a brownish color. Light that can stimulate only S and L cones without really triggering M cones is interpreted as magenta-ish.
Some theories posit that trichromatic vision is a genetic mutation where the M cone gene was copied and mutated to result in a slight shift. If it were a truly independent adaptation, you might expect it to be much further away, about the same distance S and M are, which would give humans near infrared vision. (Dogs/etc that have bi-color vision only have the mammal's original S and M cones, so their brain gets the blue vs yellow and light vs dark signals. A few mammals have only rods, resulting in true monochromatic vision).
Also, the retina ends up sending differential signal pairs to the brain: red vs green, blue vs yellow, and light vs dark, which has a huge effect on how the brain processes visual information. The naive expectation would be that it would just send the output of the three cones and the intensity, but that's not how it works. Not to mention the real-time color correction and processing, edge detection, shadow compensation, three-dimensional processing, etc.
To sum up: Any attempt to compare raw pixels is idiotic by definition. A corollary to that is the only way to measure the quality of a display device is subjectively.
Re:Real Ratina Display (Score:4, Informative)
Who said anything about holding it out?
Holding it DOWN.. at about navel level.. that's where I tend to use my droid. That's about 18 inches.
Re:Real Ratina Display (Score:4, Informative)
You did read the WHOLE section, right? You didn't just immediately stop at the first number you saw in the article? You did get to this part...
A resolution of 2 arcminutes per line pair, equivalent to a 1 arcminute gap in an optotype, corresponds to 20/20 (normal vision) in humans.
The iris, well, irises. Depending on the level of background light, the resolution changes dramatically. The claim that this screen is in that area is by no means a stretch.
Maury
Re:Real Ratina Display (Score:5, Informative)
What does that translate to in terms of halftone printing? There's a world of difference between 90000 dye-sublimation continuous tones per square inch, and 90000 little squares that can be exactly black, cyan, magenta, or yellow. That's one reason why a "300dpi" magazine like Playboy still looks richer and better than the 1200dpi output of a color laser printer, and why an inkjet printer almost always produces better-looking continuous-tone images (ie, photos) than any laser printer. A 1200dpi color laser printer uses most of its resolution to get better interpolation. An inkjet printer that sprays magenta ink over yellow ink produces a muddy orange as long as the yellow ink is still wet. A laser printer that prints magenta over yellow will end up with... magenta. Likewise, a true laser printer can (in theory, at least) do more with 300dpi than a "LED" laser-like printer, because the laser's brightness and beam diameter can be modulated a bit, so you can simulate real halftone patterns a bit more easily. In contrast, a LED laser-like printer is going to charge rectangular areas of constant dimension, so your resolution is literally *it*.
It's kind of like trying to argue about the true resolution of a recent-vintage DLP light engine. In the old days, a DLP TV with 1280x720 resolution literally had 1280 x 720 little micromirrors on the DMD (well, more for overscan purposes, but it was basically a 1:1 correlation between a single micromirror and a single rendered pixel on the screen). Then, someone (Samsung?) figured out that if you used a brighter light and modulated their movement at a higher rate, you could use one mirror to illuminate a pair of adjacent pixels. Then the whole definition of native DLP resolution kind of went to hell, because nobody knew what a pixel of resolution on a DLP TV meant anymore.
If you really want to get depressed, try shopping for a HD video camera that's more than a hundred bucks, but less than $10k. There's a huge gray area in between, and the liberties that some manufacturers (not necessarily the lowest-end Chinese imports, either) take with their advertised resolutions is borderline fraudulent. There are cameras with interlaced sensor modules that claim to be progressive by virtue of double-buffering a pair of fields internally and outputting them sequentially. There are cameras that alternate the sensors red-green-blue-green-red-green-..., then count a red-green pair as one pixel, and the adjacent blue-green pair as another pixel (hey! instant resolution-doubling makes the marketing department happy). It's sad, but in 2010 we're still reduced to taking digital photographs of black and white angled lines and using the same metric people had to use a hundred years ago for lack of a better way to describe camera resolution. 10 years ago, if you bought a camera with 1280x960 resolution, you knew damn straight it had 1280 clusters of red, blue, and green sensors horizontally, and 960 of 'em vertically. New cameras, alleged to have near-gigapixel resolution, commit frauds that basically amount to counting the number of discrete sensors sensitive to any wavelength of light, then play games with interpolation algorithms to see just how high they can claim their resolution is without getting indicted by state attorneys' offices for false advertising.
Re:300dpi is magic number, like 20kHz on CD (Score:5, Informative)
Not to be pedantic, but CD audio quality is 44.1 kHz
44.1 kHz is the sampling frequency, 20 kHz is the audio signal frequency. According to the Nyquist-Shannon Sampling Theorm in order to accurately capture a signal you need to sample at least at twice the rate of the highest frequency you want to capture. That means you should sample at a minimum of 40 kHz to accurately capture 20 kHz signals.
Now, you want to overshoot a bit because of how the filters work so you should choose a sampling rate that's a bit higher than the minimum necessary. They chose 44.1 kHz partially for this reason, but also because of the reason found on this site [columbia.edu]:
From John Watkinson, The Art of Digital Audio, 2nd edition, pg. 104:
In the early days of digital audio research, the necessary bandwidth of about 1 Mbps per audio channel was difficult to store. Disk drives had the bandwidth but not the capacity for long recording time, so attention turned to video recorders. These were adapted to store audio samples by creating a pseudo-video waveform which would convey binary as black and white levels. The sampling rate of such a system is constrained to relate simply to the field rate and field structure of the television standard used, so that an integer number of samples can be stored on each usable TV line in the field. Such a recording can be made on a monochrome recorder, and these recording are made in two standards, 525 lines at 60 Hz and 625 lines at 50 Hz. Thus it is possible to find a frequency which is a common multiple of the two and is also suitable for use as a sampling rate.
The allowable sampling rates in a pseudo-video system can be deduced by multiplying the field rate by the number of active lines in a field (blanking lines cannot be used) and again by the number of samples in a line. By careful choice of parameters it is possible to use either 525/60 or 625/50 video with a sampling rate of 44.1KHz.
In 60 Hz video, there are 35 blanked lines, leaving 490 lines per frame or 245 lines per field, so the sampling rate is given by :
60 X 245 X 3 = 44.1 KHz
In 50 Hz video, there are 37 lines of blanking, leaving 588 active lines per frame, or 294 per field, so the same sampling rate is given by
50 X 294 X3 = 44.1 Khz.
Re:Real Ratina Display (Score:2, Informative)
There really aren't first to the really high rez phone business though - my nexus one's amoled screen is about as high rez as the new iPhone's display.
Re:Real Ratina Display (Score:5, Informative)
What does that translate to in terms of halftone printing? There's a world of difference between 90000 dye-sublimation continuous tones per square inch, and 90000 little squares that can be exactly black, cyan, magenta, or yellow. That's one reason why a "300dpi" magazine like Playboy still looks richer and better than the 1200dpi output of a color laser printer...
If you're actually interested:
"300dpi" is something of an oversimplification. Images are sent down at 300dpi. The printing plates are usually imaged by laser at 2400dpi, but each halftone cell takes up more then one "dot". Print resolution is measured in "lines per inch", and ranges from ~85 lpi for newsprint to over 200 lpi for higher end printing. I'd guess that playboy prints much closer to the 200lpi end of the spectrum.
A "1200" dpi inkjet (usually more like 1440dpi) will be able to print 1440 dots per inch, but multiple dots are needed to make each halftone cell. In effect, even the best consumer level inkjets are half the resolution of an offset press.
As for laser printers, if you look at the industrial level digital presses (many of which are really glorified laser printers), they produce print that is much closer to the level of an offset press, but then again they can cost well into the six figures, so I guess you get what you pay for.
Re:Real Ratina Display (Score:3, Informative)
You have obviously never hung around places where really rich people do. My wife works with a lot of people that are in the $1M+ salary range so occasionally I have had the chance go to these things. You will meet the slutiest gold diggers there you've ever seen. It really is a different world when you're rich. I'm not sure I would like it that much.
Re:Real Ratina Display (Score:4, Informative)
Sure that one never got me laid but I knew which one I would prefer and which value I'd appreciate more.
The Amiga wasn't meant to get you laid. It was meant to fill in the void of the lack of a girlfriend, since Amiga literally means "girlfriend".
Re:Real Ratina Display (Score:3, Informative)
In my day, wealth was when you had a sports car. And I mean Italian or German, not some Japanese thing. Isn't that significantly more expensive than any iPhone you can buy?
Getting Laid, a cost/benifit analysis (Score:3, Informative)
OK, an Iphone costs A$790 for the 8GB 3G model (cheapest model in AU) which is now been superseded twice. Lets assume you, no matter how fat and ugly you are, you will get laid once before the device is superseded or rendered non functional. A flight to the Philippines with Tiger Airways from Perth costs A$550, a hotel room as little as A$30 and a sexy Pinay lady is about 1500 PHP (A$40).
Iphone A$790,
PI: A$620,
With the difference, you could get an additional 4.25 lays in the Philippines for the same money.
In addition to this, in the Philippines you'll be getting laid by a girl.
Re:Real Ratina Display (Score:1, Informative)
Nokia N900 has had this kind of display since last november. Apple are late.
Re:Real Ratina Display (Score:3, Informative)
All chicks aren't like that. I've been dumped for unemployed losers; it's a matter of chemistry. Any woman who only goes for rich guys is a whore.
Re:Real Ratina Display (Score:2, Informative)
Funny, but inaccurate. It means "female friend", but only in the literal sense of a friend who is female, but not what we mean by a "girlfriend".
Not that any computer back then was seen as fashionable - indeed compared with the stereotype of a 286 DOS user, Mac user, or an early 1990s Linux user...