Why Retina Isn’t Enough [Feature]

Why Retina Isn’t Enough [Feature]

The new Retina MacBook Pros are only Apple’s first step towards the living display of the future.

Apple’s new MacBook Pro follows the fine tradition of the iPhone 4 and third-gen iPad in that it has a super high-resolution Retina display: a 2880 x 1800 panel with an amazing 220 pixels packed in per inch.

It’s an incredible display. In fact, it’s such an incredible display that it actually has about one million, seven hundred thousand pixels more than it needs to satisfy Apple’s definition of Retina, leading some to claim that those pixels are all going to waste.

Nothing could be further from the truth.

Apple’s new MacBook Pros have absolutely great displays, but they need every single pixel they have, because the truth of the matter is that Apple’s got a long way to go before it catches its display tech up to the incredible power of human vision. And that’s a good thing, because it means we’ve got a lot to look forward to.

Editor’s Note: Throughout this article, we will be talking about two types of Retina displays: Apple’s Retina displays, and a theoretical Retina display that would have much greater pixel density. For the purposes of this article, the latter type of display will be called True Retina displays: all other mentions of Retina displays can be taken to refer to Apple’s technology.

What Apple Means By Retina
Why Retina Isn’t Enough [Feature]

No it isn’t, Steve.

At this point, everyone knows that Steve Jobs, the master showman, fudged facts a little bit when he introduced the iPhone 4’s Retina display back in 2010.

What Steve Jobs said at the time was this:

It turns out that there is a magic number right around 300 pixels per inch that, when you hold something around 10 or 12 inches away from your eyes, is the limit of the human retina['s ability] to differentiate the pixels.

Apple called such a display a Retina display, and when the new iPad was announced in March as having one, Tim Cook clarified a bit on their definition of Retina:

You may recall that with an iPhone held at a normal distance your retina can’t discern individual pixels. When the iPad is held at a normal distance [15-inches] it’s the same result.

In other words, because the average person holds a tablet further from their face than their phone, the new iPad didn’t need to have pixels that were as small as the iPhone’s, meaning the new iPad could get away with a pixel density of only 264 pixels per inch. Likewise, people sit even further back from their MacBook Pros, meaning their Retina displays only need 220 pixels per inch.

So far, so good. There’s only one problem: Steve Jobs said that the human eye, viewing a display from 12 inches away, can’t discern individual pixels if the density is over 300 pixels per inch. Except that this “magic” number is wrong. The real number is closer to nine hundred pixels per inch. Apple’s Retina displays are only about 33% of the way there.

Why Apple’s “Retina” Isn’t True Retina
Why Retina Isn’t Enough [Feature]

This eye chart is what Apple uses as a basis for its Retina displays.

Apple uses Retina as a marketing term, and it’s a great one. But it also implies that there’s nowhere else to go from here when it comes to resolution, which simply isn’t true. Ten years from now, we will all own Macs, iPhones and iPads with screens so crisp, looking at the iPhone 4S or new MacBook Pro will be like looking at a 1024 x 768 CRT from 2002. And that’s something to be excited about.

To understand why there’s so much more to be done with display resolution, you need to understand how Steve Jobs came up with his initial “magic number” for Retina: in short, he based it on a person having 20/20 vision. Seems reasonable, because colloquially, 20/20 vision is synonymous with having perfect eyesight.

The only problem? 20/20 isn’t perfect eyesight at all.

Apple’s definition of Retina is based upon the vision of seniors

When we talk about a person having 20/20 vision, what we’re actually referring to is how well they can read a standard Snellen eye chart, the kind you see hanging in optometrists’ offices all over the world. If a person has 20/20 vision, it means that a person standing 20 feet away from such a chart can read what an average person could see from the same distance. This is considered standard vision.

But while 20/20 vision might traditionally refer to “standard vision,” most research suggests that normal vision is actually much better than 20/20. In fact, people with normal vision usually won’t see their eyesight degrade to 20/20 until they are 60 or 70 years of age!

Got that? Apple’s definition of Retina is based upon the vision of seniors.

What Is The Resolution Of The Human Eye?
Why Retina Isn’t Enough [Feature]

How small is the smallest pixel your eye can see? Smaller than you think.

So what are the limits of vision? How small would a pixel actually need to be for it not to be disernible from other pixels with the human retina? To discuss this, we first need to explain a technical term: arc minutes.

Why Retina Isn’t Enough [Feature]Look at the image of the letter E to the right. It’s made up of 15 pixels, in a 5×3 grid: 11 black pixels and four white pixels. Imagine you are getting further and further away from the letter. At a certain point, you’ll no longer be able to see the white lines between the black pixels on the E.

Stop. If you have 20/20 vision, it’s generally accepted that, from this distance, each pixel on the E is now one arc minute in size.

This is how Apple builds its Retina displays. According to Apple’s definition, a pixel on a display must be no larger than one arc minute as seen from an average viewing distance by a person with 20/20 vision in order to qualify as Retina.

But as we’ve seen, 20/20 vision doesn’t mean perfect vision, and it arguably doesn’t even mean good vision. The pixels are too big. How small do we need to make them to get a True Retina display?

Unfortunately, there’s no universally accepted limit to what the human eye can see. According to DisplayMate’s Dr. Raymond Soneira, the industry’s go-to guy on display technology, the resolution of a perfect human retina is 0.6 arc minutes per pixel. So for the iPhone 4S, for example, to be truly Retina, it would need to have pixels that are 40% smaller than it currently has.

Soneira’s resolution isn’t universally accepted though. Others disagree and think the eye is much more powerful than that. J. Blackwell of the Optical Society of America determined back in 1946 that the resolution of the human eye was actually closer to 0.35 arc minutes. Again, this means that for an iPhone 4S to have a true Retina display, it would need pixels that were 65% smaller than it currently has.

For the iPhone 4S to have a true Retina display, it needs 65% smaller pixels.

Some studies push the limit of human vision even further. However, at the end of the day, it’s generally agreed that whatever the limit of human sight is, it can’t discern individual pixels smaller than 0.3 arc minutes, with most studies hovering between 0.3 and 0.4 as the limit.

In other words, for a display to truly, indisputably match the resolution of the human eye and be considered true Retina, it needs to have pixels that are no larger than 0.3 arc minutes in size from an average viewing distance.

This might seem boring, but once you know this, you can actually work out what Apple’s screen resolutions would have to be to be true Retina. This is where it gets fun, because now we can get a glimpse at the True Retina Macs of the future!

What “True Retina” Apple Displays Would Look Like

Given the above, here’s a breakdown* of what resolutions every current Apple display on the market would have if they stayed the same size, but were upgraded with True Retina displays.

ModelScreen Size (Inches)Average DistanceRetina Res (Apple)True Retina ResPPI
iPhone3.512960 x 6402772 x 1848952
iPad9.7152048 x 15365968 x 4486769
11-inch MacBook Air11.6222732 x 15365184 x 2916513
13-Inch MacBook Air13.3222880 x 18005872 x 3670520
15-Inch MacBook Pro15.4242880 x 18006096 x 3810467
21-Inch iMac21.5283840 x 21607408 x 4168395
27-Inch iMac27285120 x 28809120 x 5130388

Incredible, right? According to our math, an iPhone with a true Retina display would have more pixels than today’s 27-inch iMac!

But so what? What do all of these numbers mean in practice? Why should you care if you’re already happy with your Apple Retina Display?

Planet Earth, Seen In True Retina
Why Retina Isn’t Enough [Feature]

Only True Retina displays can do the breathtaking beauty of our planet justice.

The image above is NASA’s famous “Blue Marble” super high-res image of Earth. It’s probably the most beautiful and most detailed photo of our planet ever taken. And short of flying off into outer space, the only way you can even come close to taking in all its detail at a single glance is if you have a True Retina display.

The big boxes overlaid directly over the Earth in the image above represent the various resolutions of future Macs with True Retina displays. Anything inside a box would be what you would see on the specified True Retina Mac or iOS device display. The smaller boxes at the bottom indicate the resolutions of Apple’s current Mac and iOS line-up.

As you can see, with a True Retina 27-inch iMac or iPad, you can pretty much see the entire planet at once, with each pixel mapped one to one with NASA’s original. On a 13-inch True Retina MacBook Air, you’d take in almost all of North America. On a True Retina iPhone, you’d get the whole West Coast.

Compare that to Apple’s current models. With the 2012 Retina MacBook Pro, you get, at best, the same amount of detail you’d get on a True Retina iPhone. On an 11-inch MacBook Air, you’d only be able to see most of Texas. And on an iPhone 3GS, you’d be lucky to see all of Delaware.

If you want to see this image in its original super high resolution, you can download it here.

Why It Matters
Why Retina Isn’t Enough [Feature]

Imagine the day when you can pull out a magnifying glass to check out a picture on your iPad.

It’s true that many people with poor or average eyesight, an iPhone 4S display held from a mere 12 inches away might appear indistinguishable from a True Retina iPhone with triple the pixel density. Likewise, it’s true that some people with a Retina MacBook Pro, from two feet away, might not be able to tell the difference between that and a True Retina MacBook Pro. Some, then, would argue that this is all pointless. Hyper-resolution for hyper-resolution’s sake, aimed at a small subset of anal retentive individuals lucky enough to have above average and even perfect vision.

Such an argument is faulty. For one thing, no one sits a uniform average distance away from their devices. When you text on your iPhone, you might hold it at twelve inches, but if you’re squinting at it in the middle of the night to answer a phone call, you might hold it six inches away. And while you might write an email on your MacBook Pro at 24 inches, you might lean in on the edge of your seat during an exciting movie or game to closer to 18.

The truth is our devices suck us in as close as we can be to them. And if our displays are ever going to reach their ultimate form — living paper — they need to have enough resolution that we can zoom in to see more detail just like we would with a piece of paper: not by pinching or zooming in with your mouse, but by getting as close to the screen as we can get. Imagine the day when you can pull out a magnifying glass to check out a picture on your iPad, and instead of seeing pixels,
you see a whole world of perfectly resolved detail that you didn’t even know existed.

But there’s another reason why everyone should care. Back in 2010, Japan’s NHK Science & Technology Research Laboratories discovered that across the board, the smaller the pixels, the more real an on-screen object seemed to the person viewing it, even if they didn’t have perfect vision. In fact, this effect is so profound on individuals that it eventually reaches a point where you have pixels small enough that on screen objects are completely indistinguishable from real objects.

That is the future of displays: screens so perfect that FaceTiming with a friend is visually indistinguishable from having a conversation in person with them, where watching a movie is like peering out a window to another world, and where playing a video game is like having a real-life adventure. That is why we need True Retina. And that is where Apple is ultimately going to take us.

* – The math behind how we did these calculations is a little complicated, but a brief word of explanation. If we assume that the resolution of the human eye maxes out at 0.3 arc minutes per pixel (see above), we can then determine how many pixels a display needs to have (given average viewing distance) to be a true “Retina” display to everyone, even if they have better than 20/20 vision.

To do this, we need to first calculate the area size a display takes up in a person’s field of view, or in other words, how big a square of light is actually shining on their eyeball at a given time. This is measured in degrees of apparent size, which we can then multiply by 200 (60 arc minutes per degree divided by 0.3 arc minutes) to come up with the minimum number of pixels we need for True Retina.

This might all seem complicated, but once you understand this math, it’s incredibly useful for doing things like figuring out how big a TV you need for your living room, or what row you should sit in a movie theater. For further explanation, consult this page.

Many thanks: The help of reader Neven (@randomoneh) was invaluable in gathering a lot of the data for this piece together. In addition, I’d like to thank my friend Gil Forsyth for helping me with some emergency last minute trig problems. You guys are much, much smarter than me.

  • Tallest_Skil

    Thank you for posting something that absolutely needed said.

    This is why people who tell me that “I can’t see a difference between 720 and 1080″ are blind fools.
    This is why I advocate for going to Super Hi-Vision, and why the TV resolution world doesn’t need to wave their collective penises around anymore (meaning the customer doesn’t get screwed or lied to) once we get there.
    We can see it.
  • crateish

    Why Retina Is Enough:

    The average consumer doesn’t notice the difference between a Retina and a non-Retina screen.
  • randomoneh

    Great job!

    @ crateish,

    Please read the article. Especially interesting part:

    “But there’s another reason why everyone should care. Back in 2010, Japan’s NHK Science & Technology Research Laboratories discovered that across the board, the smaller the pixels, the more real an on-screen object seemed to the person viewing it, even if they didn’t have perfect vision. In fact, this effect is so profound on individuals that it eventually reaches a point where you have pixels small enough that on screen objects are completely indistinguishable from real objects.”

  • Evan Meichtry

    Everyone I know as well as myself can’t see the pixels on the iPhone even when holding it really close because then the eyes seem to lose focus. Anyone else feel like that? Also on an iMac I can see this “super” retina actually mattering.

    • jbo5112

      I see the pixels on the iPhone just fine, as well as subpixel artifacts on my 386 ppi Note 3. It’s odd to have a pentile arrangement. My horizontal lines are jagged and my diagonals are rather smooth. With a little practice, you can probably train your eyes to focus on something closer.

  • JaredABurnett

    This is the best article I’ve ever read on an Apple blog site. This concept is something that I think most of us had thought a lot about and I’m glad that it has finally been addressed. Thank you John.

  • john4043

    hmmm love this piece!

  • Arnoud van Houwelingen

    Great article .. This is what journalism is all about!

  • david_fabian

    I frequently sit 18″ away from my 27″ iMac, especially when working with Photoshop. So I do see the draw of a “True Retina” display for pros.

  • baby_Twitty

    John Brownlee is on fire… 

  • Antonio14_

    great post! excelent job but the tittle is wrong, retina display is enough but (like the post says) is not really retina. 

  • DilbertAsended

    Another great article.

  • Ronin

    Big swing and a miss.

    Soniera is no doubt an authority on the electronics and physics behind the displays but the specs and details of the display are not in dispute.  
    The question revolves around the acuity of the human eye.  This is a biology question, not an engineering or math question.
    Assuming the standard visual acuity or 20/20 vision means that a 20/20 eye can discriminate two lines or two pixels separated by 1 arcminute (1/60 degree).  Your discussion or pixel size is of little use.  A 20/20 human eye that can discriminate two points separated by 1 arcminute/cycle at a distance of a foot can, therefore, also discriminate two points 89 micrometers apart which would be about 287 pixels per inch.  The “math” isn’t complicated.
    This work was done by the appropriate expert for this question: a retinal neuroscientist named Bryan Jones (not the resurrected former Rolling Stone).
    So your assumptions are wrong, your cited expert is wrong, and your understanding of the question is wrong.  But if you’re are getting paid by the word and click you probably don’t care.
    Bottom line:  All Apple’s claims about their Retina displays are legitimate.

    See Jones’ work here: http://prometheus.med.utah.edu/~bwjones/2010/06/apple-retina-display/

  • randomoneh

    Big swing and a miss.

    Assuming the standard visual acuity or 20/20 vision means that a 20/20 eye can discriminate two lines or two pixels separated by 1 arcminute (1/60 degree).  Your discussion or pixel size is of little use.  A 20/20 human eye that can discriminate two points separated by 1 arcminute/cycle at a distance of a foot can, therefore, also discriminate two points 89 micrometers apart which would be about 287 pixels per inch.  The “math” isn’t complicated.

    You’re missing the point. Angular resolution of 60 ppd may be fine for those with 20/20 vision (but not even close to the limit, search for NHK in the article and follow the link), but majority of young people have much better vision (again, read the article).

  • John Brownlee

    Big swing and a miss.

    Soniera is no doubt an authority on the electronics and physics behind the displays but the specs and details of the display are not in dispute.  
    The question revolves around the acuity of the human eye.  This is a biology question, not an engineering or math question.

    Exactly. It’s a biology question, as the article makes abundantly clear. It really seems like you didn’t bother reading or understanding the article, because you make like I’m citing Soneira as my sole expert, when, in fact, I actually cite more than three different research papers on vision acuity. And despite what you say, I don’t even use Soneira’ definition of the limits of human vision!

    Assuming the standard visual acuity or 20/20 vision means that a 20/20 eye can discriminate two lines or two pixels separated by 1 arcminute (1/60 degree).  Your discussion or pixel size is of little use.

    Again, it really seems like you don’t understand the core concepts here. Pixel size is INCREDIBLY important. It’s of paramount use. That’s the whole point of a Retina display! If a pixel is smaller than a certain measurement in arc minutes from a certain distance away, then it is a Retina display: if it isn’t, it’s not a Retina display. It’s totally simple. Saying pixel size doesn’t matter is just silly.

     A 20/20 human eye that can discriminate two points separated by 1 arcminute/cycle at a distance of a foot can, therefore, also discriminate two points 89 micrometers apart which would be about 287 pixels per inch.  The “math” isn’t complicated.

    Again, I have a hard time understanding how you could have read the article, saying this. I spend at least 500 words talking about how 20/20 vision isn’t the limit of human vision (and explaining exactly why) and another 500 words or so explaining that multiple sources say that the necessary resolution needed to reach the limits of human visual acuity is closer to 800-900ppi from 12 inches away.

    So your assumptions are wrong, your cited expert is wrong, and your understanding of the question is wrong.  But if you’re are getting paid by the word and click you probably don’t care.

    Bottom line:  All Apple’s claims about their Retina displays are legitimate.

    No, I’m sorry, you’re wrong, in every particular. I don’t cite one expert — I cite no less than three , and the one you’re singling out (Dr. Soneira), I didn’t base the article’s logic upon. Furthermore, I’m a salaried writer, so I’m neither “paid by the word” nor “paid by the click.” And, as this article makes abundantly clear, Apple’s claims about Retina displays aren’t legitimate.

  • Wbeaudot

    Nice article but you are wrong in thinking that:

    1) the display has to match the retina resolution (the retina is a receiver and the display an emitter, so according to the well-known Nyquist-Shannon sampling theorem a display compatible with the retina resolution would only need to have half its resolution),

    2) visual acuity is limited by the retina resolution alone (optics is actually the main limiting factor).

    See my original blog on “Retina Display”: http://www.kybervision.com/Blog/files/AppleRetinaDisplay.html

    By the way I’m one of the other vision scientists who has strongly disagreed with Dr. Soneira and supported Apple’s claim.

  • John Brownlee

    Nice article but you are wrong in thinking that:

    1) the display has to match the retina resolution (the retina is a receiver and the display an emitter, so according to the well-known Nyquist-Shannon sampling theorem a display compatible with the retina resolution would only need to have half its resolution),

    Thanks for writing. If you’re right, and a display doesn’t need to match the resolution of the retina, but only half that, then a minimum of 0.6 arc minutes per pixel would seem to be the magic number according to what I wrote above… which is still a great deal better than Apple’s 1 arc minute per pixel minimum.

    Also, in your article, you state that 20/12 is the limit of human visual acuity, and indicates 20/20 is an acceptable norm. The research I cite above challenges that assertion, and the NHK data says that at far greater pixel densities than Apple is using, on screen objects (given ideal brightness and contrast) look indistinguishable from real objects. Surely, then, there is still a great deal of room for improvement.

  • randomoneh

    Nice article but you are wrong in thinking that:

    1) the display has to match the retina resolution (the retina is a receiver and the display an emitter, so according to the well-known Nyquist-Shannon sampling theorem a display compatible with the retina resolution would only need to have half its resolution),

    2) visual acuity is limited by the retina resolution alone (optics is actually the main limiting factor).

    1 & 2 – Article doesn’t take a route of calculating Retina display resolution by figuring out the resolution of human retina (cone size in foveola, cone spacing…) nor does it ignore optics.

    Instead, it uses measured values (20/xx, not theoretical) for final calculations.

  • Ronin
    Big swing and a miss.
    Soniera is no doubt an authority on the electronics and physics behind the displays but the specs and details of the display are not in dispute.  
    The question revolves around the acuity of the human eye.  This is a biology question, not an engineering or math question.

    Exactly. It’s a biology question, as the article makes abundantly clear. It really seems like you didn’t bother reading or understanding the article, because you make like I’m citing Soneira as my sole expert, when, in fact, I actually cite more than three different research papers on vision acuity. And despite what you say, I don’t even use Soneira’ definition of the limits of human vision!

    Assuming the standard visual acuity or 20/20 vision means that a 20/20 eye can discriminate two lines or two pixels separated by 1 arcminute (1/60 degree).  Your discussion or pixel size is of little use.

    Again, it really seems like you don’t understand the core concepts here. Pixel size is INCREDIBLY important. It’s of paramount use. That’s the whole point of a Retina display! If a pixel is smaller than a certain measurement in arc minutes from a certain distance away, then it is a Retina display: if it isn’t, it’s not a Retina display. It’s totally simple. Saying pixel size doesn’t matter is just silly.

     A 20/20 human eye that can discriminate two points separated by 1 arcminute/cycle at a distance of a foot can, therefore, also discriminate two points 89 micrometers apart which would be about 287 pixels per inch.  The “math” isn’t complicated.

    Again, I have a hard time understanding how you could have read the article, saying this. I spend at least 500 words talking about how 20/20 vision isn’t the limit of human vision (and explaining exactly why) and another 500 words or so explaining that multiple sources say that the necessary resolution needed to reach the limits of human visual acuity is closer to 800-900ppi from 12 inches away.

    So your assumptions are wrong, your cited expert is wrong, and your understanding of the question is wrong.  But if you’re are getting paid by the word and click you probably don’t care.

    Bottom line:  All Apple’s claims about their Retina displays are legitimate.

    No, I’m sorry, you’re wrong, in every particular. I don’t cite one expert — I cite no less than three , and the one you’re singling out (Dr. Soneira), I didn’t base the article’s logic upon. Furthermore, I’m a salaried writer, so I’m neither “paid by the word” nor “paid by the click.” And, as this article makes abundantly clear, Apple’s claims about Retina displays aren’t legitimate.


    You still don’t get it.  Pixel size is not the issue here because they are easily small enough to justify the retina claims of Apple when viewed at the usual distances.  The question is what can and can’t be perceived by a human retina so it’s retinal neurobiology – not engineering and physics – that matter.  And it’s not really even a controversial question for those who know about it: it’s already been answered.

    You are not in the field so maybe that is part of the problem for you.  Your citations are flawed for numerous reasons and are outside the consensus as you attempt to apply them.  Additionally, 3 or 4 citations don’t make a consensus in science.  Maybe on a blog but not in a scientific discipline.

    I repeat: Simply stated, a 20/20 human eye that can discriminate two points separated by 1 arcminute/cycle at a distance of a foot can, therefore, also discriminate two points 89 micrometers apart which would be about 287 pixels per inch.  The “math” isn’t complicated.

    So your assumptions are wrong, your cited experts are wrong or wrongly referenced, and your understanding of the question is wrong.  Please leave the science to the scientists.

    Bottom line:  All Apple’s claims so far about their Retina displays are legitimate.
  • randomoneh
    Bottom line:  All Apple’s claims so far about their Retina displays are legitimate.

    You haven’t disproved a single claim from this article.

    If you simply convert Snellen chart result of 20/20 to arcminutes per pixel – yes, you’re going to end up with 1 arcminute per pixel figure [287 ppi @ 12''] (even though at this point contrast is equally important factor to consider + NHK’s study implies that humans are able to perceive quality beyond their Snellen-tested acuity).

    We haven’t seen an asterisk at Apple’s website saying “Doesn’t apply to those with vision better than 20/20.” As John said, they implied there’s nowhere to go from here.

    Apple’s a great company, they have enough great products that we don’t need to be biased and try to defend their every move.

    I have nothing more to tell you but this – try to read everything again.

  • Ronin
    Big swing and a miss.
    Assuming the standard visual acuity or 20/20 vision means that a 20/20 eye can discriminate two lines or two pixels separated by 1 arcminute (1/60 degree).  Your discussion or pixel size is of little use.  A 20/20 human eye that can discriminate two points separated by 1 arcminute/cycle at a distance of a foot can, therefore, also discriminate two points 89 micrometers apart which would be about 287 pixels per inch.  The “math” isn’t complicated.

    You’re missing the point. Angular resolution of 60 ppd may be fine for those with 20/20 vision (but not even close to the limit, search for NHK in the article and follow the link), but majority of young people have much better vision (again, read the article).

    No.  You are wrong.

    Of course there are outliers for any metric.  20/20 is an universally accepted norm and average.  You saying the new MBP Retina display is not a retina display for someone with better than 20/20 vision is like me saying the iPhone 3 display IS a retina display for someone with 20/500 vision.  It’s not “retina display for people with way better than average acuity,”  it’s “retina display.”
    Apple’s retina display claims are legitimate.
  • ApplePr0n

    All I will say is that I work for a VERY well known consumer electronics company in the U.S. and we have the Retina Display MBP’s out on display and the screen is gorgeous, I mean it has to be seen to be believed

  • EllisJohnnie1

    just as Martha responded I’m amazed that a mother able to make $5078 in a few weeks on the computer. did you see this web site(Click on menu Home more information)  http://goo.gl/ZiRbh


  • Thomas Manheim

    There is one important point on the variation of eye performance- dark adaption.
    The eye pupil diameter varies from 2mm to 7mm. You are diffraction limited to about 1 minute
    when the pupil is only 2mm in diameter.
    To get the high resolution stated you need to turn the brightness of the
    screen down to very nearly zero and hide in a dark room. You then need perfectly non
    astigmatic eyes at maximum pupil diameter- rare.

    Looking at anything at all luminous on the screen closes the pupil and degrades the resolution.

  • Ari Kolbeinsson

    Even if everything in the article were correct, of which I am not convinced, then we WON’T be seeing these super-high resolutions. Why?

    Well, the law of diminishing returns. The difference between 72/96ppi and 300ppi is pretty huge and very noticeable. No research suggests that the difference between 300ppi and 500ppi is anything that you would notice, and 500ppi to 900 ppi is WAY not noticeable. Maybe the images seem somehow more real, but convincing a consumer to pay a much higher price for something that he can’t see is not a winning proposition.

    Then there’s the pixel pushing requirement. The next ten years in GPU development would have to go mostly towards pushing these higher resolutions if the resolution were to be usable. This is a solvable problem, but still needs solving.

    There are two related problems that might be tougher. Feeding the screen (data and electricity) would be much tougher, and the other problem is one of light (also about electricity). The new MBP screen is dimmer than the previous MBP screens even though it has a bigger backlight and draws more power. This is because the grid of “wiring” for all those little pixels take up space and block light. Increasing the pixel density means reducing the light that passes through. This is a very real problem and does not look like it is solvable for 900ppi using today’s tech. The new MBP has a huge 95Wah battery to compensate for this, yet only gets seven hours battery life. The same battery in an older MBP should give over 10 hours of life. Take the screen to 440 PPI and you would probably need to at least double the battery size as well as figure out a pretty good cooling system for the backlight. The power draw would be immense.

    But then again, I think that the primary problem is that almost no one will be willing to shell out for screens with higher ppi than around 300. They just wouldn’t see enough of a benefit. 200-300 reaches that magical “good enough” level.

    It hasn’t been mentioned here yet but the human retina isn’t made up of equal and square pixels. In fact it uses cells that are packed into all kinds of shapes and have a variance in  size. Our brains make up for this, as well as making up for all kinds of strange vision artifacts. It also introduces a few of its own, as do the eyes themselves. And working with a backlit screen limits our optics as well (through our eyes contracting the pupil) and reducing the effective resolution we can see.

    I will stand by the vision experts who have commented here in that some of the stuff in the article seem to be slightly misunderstood or overstated, and that Apple’s retina display advertising seems legit (using their criteria. Standard vision can be assumed as it is, well, standard..). My background is in cognitive science (including neuroscience, vision, and visual processing)

  • richarddas

    Interesting report, but I think you’re missing the bigger picture: high density, so-called “Retina” displays, are really an engineering milestone on the way to glasses-free 3D displays.

    When you can serve four pixels in the physical space where you used to serve 1, and place a diffraction lens on top of it of equal size, you gain the capability to serve different images in different directions (namely one image to the left eye, and one image to the right eye).

    Add an accelerometer into the device so that the screen “knows” it’s orientation in space, and a camera to the front so that it “knows” where the User’s eyes are located, and you have glasses-free 3D. This will come to mobile devices first, since they have the benefit of single-viewer interaction and somewhat standard/knowable positions for the User (as opposed to, say, a television screen which has multiple-viewers and a wide range of viewable positions).
  • carloborja

    John Brownlee, this is probably the best article I’ve ever read on CultOfMac so far. I wish to hear more of you on the CultCast.

    You are a geek that deserves respect and recognition.

  • ElliottClaude

    what Randy said I am impressed that a stay at home mom able to make $8511 in 1 month on the internet. did you look at this website(Click on menu Home more information)  http://goo.gl/WgLfD


  • Dan Kaminsky

    Oh, it’s so much more complicated than you’re giving it credit for.

    Humans can fairly easily distinguish 2400dpi and beyond.  Don’t believe me?  Make the dots black and white.
    As it turns out, it is much easier to detect edges when they’re high contrast.
    It’s when they’re not high contrast that things get messier, and frankly, this is why we can throw out most of the data posted thus far.  If you aren’t describing the contrast ratio under a given luminance level, you aren’t actually describing anything.  The reality is that the reason Retina is so close to the edge of visibility is precisely because it’s a full color display, meaning the deviation across pixels ends up nice and smooth.
    Pretending there’s some crazy frontier in which things look real just because we have more pixels to blend across is…a little too hopeful.  What’ll come down the pike is 3D — not this binocular delusion that people think creates 3D, but actual, light field projection in which the eye receives light at each appropriate focal depth.  That’s the revolution we’re waiting for, not 4K on the phone.
  • Brian Maday

    At my age, I reached RETINA DISPLAY at 1024×768 about 10 years ago! At least now I know why my magnifying glass is not dusty. Great explanation – makes me feel better about nature’s tendency to allow the human body to wear out over an extended period.

  • justmeblue5ft3

    Okay! You build one and give it to me and I’ll post a review. Seriously!

  • Jason Mark Ericson

    Very interesting and thoughtful article. However, I have kind of a philosophical issue with it. Most of your claims about Apple’s Retina Display being not high enough resolution are based on research that challenges the accepted norms…but there are still accepted norms. For Apple’s purposes, why should they bother putting in the effort when the most mainstream research on the topic backs up their assertion that Retina Display is high enough for the average person to no longer discern pixels? Unless you’re suggesting that the research making these challenges will become the accepted norms over time?

    Bottom line: the research you’re citing has not yet been corroborated enough for Apple (or anybody) to make any changes. Two cents!

  • Tseug

    Yes this is a great article, in depth and scientific. I do prefer these sorts of long, deeply analyzed articles than the more flippant updates that many other sites provide.

  • graig_smith

    i am slightly nearsighted. so my vision is best up close. i have noticed i can see small things that a lot of other people can’t see. i can totally see the pixels on the iPhone 4. yes they are tiny. and yes i have to get pretty close to the screen to see them. do i need more? no.. any more pixels would be a waste at that PPI.

    “No, I’m sorry, you’re wrong, in every particular. I don’t cite one expert — I cite no less than three , and the one you’re singling out (Dr. Soneira), I didn’t base the article’s logic upon. Furthermore, I’m a salaried writer, so I’m neither “paid by the word” nor “paid by the click.” And, as this article makes abundantly clear, Apple’s claims about Retina displays aren’t legitimate.”

    experts. yeah i can see the pixels. but not at a normal usage distance… this is why i think apples claims are legitimate. Could they make them smaller? yes. they could make them small enough so that i couldn’t see the pixels. But it wouldn’t make the devices better. if someone came out with a device tomorrow with your “super” retina pixels per inch. no one would care, because it’s excessive. apple took their retina displays to where they needed to go to make it look great and not any further. why go further if it’s great for how your using it?? going with more pixels than necessary would just increase processing demands on the device and lower battery life.

  • technochick

    You are basically your whole ‘facts’ on disputed research and data you didn’t bother to gather or math you could actually do yourself. Including from someone that isn’t, it seems, an expert on the field he deigned to comment on. And yet you include him as one of your experts.

    classy John, real classy.

  • magmashift

    Even if displays reach such high resolutions, you’ll still be able to discern objects displayed on them from real-world objects. Why? Because, for onscreen images, you’re still limited to what the camera lens collecting the images is “seeing” and cannot use the focusing dynamics of your eyes. For example, when looking at the real world, I can precisely focus on any object in the scene, regardless of distance. When I focus on a near object, distant objects appear blurry and vice-versa. Not true when viewing an image on a display, where I’m limited to seeing the particular focal length that the camera’s lens was set to when the image was captured (or whatever the moment-to-moment settings are while it’s being captured, in the case of video).

  • inkshawn

    Well written piece. Problem is the average non-techie consumer does not value sharper display enough to pay a significant premium for it. So many think that the screen just “looks nice” and don’t value that enough to fork out for. When asked to compare, it is intriguing to see how many people struggle to pick out the difference with the colour vibrancy of the New iPad making more of a difference and the sharpness.

    It is only when some apps really highlight this improved resolution that makes all perceivable difference.

    The cost premium of delivering the number of pixels being discussed in this article leave its core points in the realm of fancy.

  • LittleLion

    cool! One day we can build huge displays to create real life roadrunner situations!

  • iraszl

    I think you’re wrong. There is truly nothing beyond 330ppi. You can test it easily even without a screen that is 5-600ppi.

    First, you take a low resolution 160ppi screen and look at it at 30cm and then at 60cm. You will clearly see that at 60cm the screen is smaller but you see a sharper screen with equal amount of detail as at 30cm.

    Now, take a retina display and test it at both sizes. You will have a different experience from the previous test. At 60cm you will not feel like you’re seeing a sharper image. You will feel that it’s the same sharpness ast at 30cm but at 60cm you will start to lose detail. That’s because at 60cm you’re simulating a screen with more than 330ppi and your eye can’t tell the difference between pixels anymore.

    Another simple test is to shoot a texture and then open the image and put it next to the texture. You will see that it’s equally sharp as reality. Your eye can’t see more details than on a retina screen.

    People who worked in design and print for years know this as a simple fact.

    The only supposed benefit of a 500-600ppi screen would be that you can look at details with a loupe or a microscope, but since you can pinch zoom to see detail that’s a really unnecessary feature. :)

  • Circus clown

    What an anal retentive mouthfoam. :DDD Grow up kid.

  • TekBoi

    For years I proclaimed that screens were far too low in resolution. While I agree that Apple definitely pioneered the tech revolution of higher displays, and the mentality that everything needs to have a higher resolution, they almost never lead the way.

    Back in 2009 when I had the terrible HTC Touch Pro 2, I was boasting the VGA resolution, 259ppi (at the time, the iPhone had a terrible 165ppi). And while the retina Macbook is amazing at 220ppi, it’s lacking when compared to the 262ppi, touch screen, Razorblade. Also, while the iPhone lingers in the 300~ ppi range, Oppo has reached 500+ ppi with the Find 7.

  • Lego KingsCastle

    That’s why I’m getting the lg g3 if the iphone 6 has some stupid 900×1700 or whatever

About the author

John BrownleeJohn Brownlee is a Contributing Editor. He has also written for Wired, Playboy, Boing Boing, Popular Mechanics, VentureBeat, and Gizmodo. He lives in Boston with his wife and two parakeets. You can follow him here on Twitter.

(sorry, you need Javascript to see this e-mail address)| Read more posts by .

Posted in Featured stories, News, Top stories | Tagged: , , , , |