You are here: » Vision Research » Vision in the European Focus » 2012 » How simple vision concepts can explain high-technology products by Pablo Artal

What have in common the Apple “Retina” display and our retina?

image artal banner

Probably, only the name as explained in Pablo Artal blog

Dear Prof. Artal.
First of all thank you for your very interesting blog. In my opinion it is tricky and unfair for Apple to use the term "retina" for their screens. It is not only the number of pixels to be compared with the eye resolution (photoreceptors).
Some of our photoreceptors are not connected 1 to 1 with "signal" transmitters. Some of them are undergoing a kind of amplification by being connected to more than one "signal transmitters". I hope I gave you my point quite simply. Please correct me if I am wrong. Alex. Murcia, Spain.

I received this very timely question from Alex Pennos, a PhD student that recently joined my lab in Murcia from Greece (I hope not trying to escape from the difficult Greek situation to land in a probably worst Spanish situation!). Several other readers of my last post on the new iPad display also mentioned this issue about the convenience of the marketing name that Apple is using for their new displays. So, I had no choice but to write something else on the name of the Apple display. The last entry described a comparison of the resolution of the displays in the old iPad2 and the new iPad using very simple concepts of Visual Optics. I should mention to you that I discovered the "magic" of the word "iPad", because this, usually quiet, blog received more visitors in a day than in a full month after this was posted.

The question from Alex made a lot of sense. I believe many people (especially vision scientists) are not happy with the name used by Apple: “Retina”. Are they suggesting that these displays have significant resemblances to our own retina? I will try to give you here some information so you can decide yourself on this issue, using a few simple Physiological Optics facts.

The latest versions of the iPad (new) and iPhones (4 and 4S) are equipped with a high resolution display, called “Retina”. The reason for this name was revealed by Steve Jobs during the presentation of the iPhone 4: "due to the high density of pixels in the "Retina" display, the human eye is unable to distinguish them". This statement is almost correct, as I described in the previous post, since under normal viewing conditions most people are not able to see the pixels details. Although under certain conditions there are a number of people with very good visual acuity that could actually distinguish them.

However, many scientists and science aficionados, who are otherwise usually very enthusiastic to any Apple's action, accepting them as a kind of dogma, do not like the choice of this name. And, I include myself in this category, although I follow quickly most Apple products, I can still maintain a critical view! The problem is that the use of the “Retina” name seems to indicate that the characteristics of the display are similar to those in our retina. Considering the Apple global influence, the name is certainly unfortunate because the differences are simply enormous. And this is not a trivial issue, if you consider the central role of the display in the marketing of the Apple products (i.e.; the new iPad is called "Resolutionary"...)

"Retina" versus retina.

Let me discuss some of the features of our retina compared with the display of the new iPad. This has a resolution of 2048x1536 pixels, more than 3 million in total. Each pixel has a size of approximately 96 micrometers and contains three sub-elements (red, blue and green) to reproduce colors. Our retina has two types of photoreceptors, cones, used in daylight and color vision, and rods, mainly used in night vision. We have about 7 million cones, with a size of around 2.5 micrometers, and over 120 million rods. You can have more detailed information on the retinal topography in the excellent paper by Curcio et al. (Science, 236, 579-582, 1987). The pixels of the display of the new iPhone are about 40 times larger than the cones. The figure below shows schematically that within each pixel of the screen we could place about 1500 cones! Just for comparison, keep in mind that one pixel of the iPad2, with poorer resolution, would contain 4 of the new display.

Figure 1
image figure 1
Figure 2
image figure 2

And the next figure shows, approximately to scale, four pixels on the ”Retina”  display compared to the mosaic of cones in the central retina (the fovea, the area with the highest resolution).

But resolution is NOT everything in a display.

As many readers correctly pointed out, while important… resolution is not the only quality indicator for a display. So, let’s consider other features. For example, one iPad placed at reading distance covers a field of view of about 30 degrees, a small fraction of our field of view that is approximately 200 degrees. And the contrast ratio of the "Retina" display is around 1:1000, a modest value compared to the spectacular range of sensitivity of our eyes, 1:1000 million. Yes, this is a huge number... but, you can think in the enormous range of luminance operation of our visual system: from a bright sunny day (100000 cd/m2 luminance) to a dark night (0.0001 cd/m2).

And what about color reproduction?

Figure 3
image figure 03

Arthur Ho, a principal scientist at the BrienHolden Vision Institute in Sydney, and a world expert in applied vision, cleverly noted in a comment to the previous entry the following: “Resolution is one thing... but I'm still waiting for Apple (or indeed, any other manufacturer) to make a display that can render our entire visual color space. For that parameter, expanse is as if not more important than resolution”. I fully agree with Arthur. A weak point in current displays, including the "Retina", is the relatively poor color reproduction. Comparing the range of produced colors with those that the human eye can distinguish, the difference is very large. Figure 3 shows the chromaticity diagram of the "International Commission on Illumination" (called CIE 1931), which represents the range of all colors. The marked triangle covers the area reproduced by the “Retina” display. A relatively small fraction… (although larger than some than its predecessors).

“Retina” displays: an inappropriate (and unfortunate) name?

If you consider only these data, it is clear that displays are still technologically far from our wonderful retina. While recognizing that the current displays are getting better, the reason given by Apple to justify the name "Retina" is not very convincing: "a resolution that makes the pixels indistinguishable." An old VGA resolution monitor could be then considered as "Retina" as well in the sense of having pixels indistinguishable if the observer viewing distance is large enough... In this case Apple was not very kind with many of us admiring the complexity, sophistication and extraordinary capabilities of the human retina. In my opinion, it is inappropriate, even for Apple, to call "Retina" to a good screen, but so far from our retina.

Is the display resolution of the new iPad adequate for the eye?

Dear Prof. Artal.
I read that the resolution of the new iPad screen is so good that the pixels cannot be seen with the naked eye. Is this correct based in what we know about the eye? I tried myself to figure out this before deciding if I should change my Ipad2, but unsuccessfully. Could you please help me? Anna. Santa Fe. New Mexico, USA.

Dear Anna, yes… this question really comes at a perfect time, because I just bought today the new Ipad at the Apple store in Bondi Junction in Sydney (by the way, it was easy and quick with no queues), and of course... I was personally curious about this issue. As many readers may know, one of the main improvements with the new iPad is the higher screen resolution as compared with the previous Ipad2. To be honest, I always had the feeling that the resolution in the Ipad2 was good enough, although of course you can see the individual pixels easily. But, what happens with the new IPad? Is the resolution good enough? Can you tell the difference between the two devices? These are perfect questions for any “Visual Optics” aficionado. Although this could be a good exam question for my students… I will answer here for all interested readers.

New Ipad versus Ipad2 display resolutions. The new iPad has a high-resolution "retina" display with 2048x1536 pixels. The Ipad2 has a 1024x768 pixel resolution. Since both screens have the same size, this means basically that the new Ipad has approximately 10.4 pixels/mm in comparison to 5.2 pixels/mm in the Ipad2, so the linear resolution is doubled (although as a reader pointed out in two-dimensions, each "old" pixel contains four "new" pixels). Tablets are used for near vision and although the reading distance varies from individuals we can assume a standard distance of 33 cm. This means that a pixel in the new Ipad subtends a visual angle of around 1 minute of arc (1 arcmin = 1/60 degrees), as compared to the pixel in the Ipad2 subtending 2 minutes of arc.

Eye’s normal resolution. Normal young subjects with good vision without the need of glasses, or well corrected, have a resolution for distance objects that ranges from 1 arcmin (this is equivalent to visual acuity 20/20, or 1 in decimal notation) to in some exceptional cases 0.5 arcmin (20/10 or decimal 2). For different reasons that I will mention below, for objects placed near (as the Ipad) the eye’s resolution is usually lower, being normal around 1.25 arcmins (what is called in Ophthalmology J1), although in many young subjects with good vision could be around 0.75 arcmin (J1+++). If you are more interested in these visual acuity numbers and their underlying causes you can find more information in an article we published a few years ago (E.A. Villegas, E. Alcon, P. Artal, "Optical quality of the eye in subjects with normal and excellent visual acuity". Invest. Ophthalmol. Vis. Sci. 49, 4688-4696, 2008.

And for those readers with some additional interest in the eye properties, or in how the optics affects vision, you can have a look to this video and navigate through some of the previous blog's entries.

So, if you are young having very good vision, you will see the screen pixels in BOTH the iPad2 (2 acrmin > 0.75 arcmin) and the new iPad (1 acrmin > 0.75 arcmin).

If you are young with average, or slightly below average, normal vision you will see the pixels easily in the Ipad2 (2 arcmin > 1.25 arcmin) but not in the new Ipad (1 arcmin < 1.25 arcmin). If you are relatively old with a moderate near vision acuity (referred as J2) you may still see the pixels in the ipad2 (2 > 1.5).

I took a few pictures of the two screens (with a camera that can detect both pixel structures, so that would be equivalent to a good young eye). See below the differences between the two displays in a detail of an icon and some text from an ebook (click in the images below to better see the details).

Figure 4
image figure 04
Figure 5
image figure 05

Under optimum conditions, most people with normal vision are able to see the pixels in the iPad2, but not in the new Ipad. It is also clear from these pictures. Still in theory the new iPad would not be enough for young subjects with good visual acuity. An even better resolution is then required for those people? My answer is probably no because there are other considerations under real viewing conditions of use that probably would render invisible the pixels of the new Ipad for a majority of subjects. First, not every person of the fraction of subjects with a potential excellent visual acuity would be perfectly corrected. It is very likely that nearly all subjects could have defocus and astigmatism within at least 0.25 diopters, blurring the retinal image at the level of the pixel size. For those with nearly perfect refraction that are young (with good accommodation response) still a couple of another issues to consider:

Accommodation lag (added defocus)

Even when distance objects are at perfect focus, for near objects (such as the Ipad) there is typically an accommodation error of around 0.5 diopters. This means that the pixels would be somehow blurred. I am using as an example below a self-portrait of my daughter (Lucia Artal) she composed with letters. The small details in those letters can be compared as the pixel details in the Ipad. Note how a small amount of defocus blur the letters in the picture at right. The same would happen with the new Ipad’s pixels when subjects have some small acccommodation errors.

Figure 6
image figure 6
Figure 7
image figure 7

Low contrast pixel details

The values of the eye’s resolution (acuity) I used for the calculations above are typically corresponding to high contrast details (for example a black letter in a white background). However, in many cases the pixel details could be of lower contrast. The resolution of the eye would be lower for that type of target. You can check this yourself in the picture below. Not blur added, only the lower contrast renders the letter details of the picture more difficult to detect. 

What about presbyopic iPad users?

In the case of presbyopic subjects, those over 45 years old (the exact age depends on the particular refraction and individual conditions) that are not able to focus at near objects, the new Ipad pixels would be normally invisible. For the oldest persons, with a near visual acuity in some cases considered as compromised, but normal (J3) even the pixels in the Ipad2 would not be visible.

And myopic persons? (as myself)

In myopic subjects near objects are in good focus when not corrected. It is very common that the preferred reading distance will be shorter. In my own case, around 20 cm (and in the case of more severe myopes can be even shorter). This would render in principle the new Ipad pixels nearly visible. However, most myopes will not be able to see the pixels even placing the Ipad very close to the eyes. I can tell you that I am not able to see them myself at any distance!

In summary…

the new iPad display resolution is adequate for most eyes. Under normal viewing conditions only a few subjects would be able to see the pixel details. On the contrary, under normal viewing conditions a majority of subjects can see the pixels in the old iPad2. So, the difference is really evident. For this type of tablet device the resolution in the new Ipad is an excellent compromise and it is well matched to most eyes. The Apple’s vision researcher consultant did a good job in this case!

But, any problems with the new Ipad?

Well, it gets warmer than the iPad2, specially in some localized areas. Other issue is the extra dozens of grams heavier. And I believe it takes longer to recharge the batery. So far, these are not very significant problems, at least for me. And after all... although these are also physics problems (Thermodinamics, Mechanics and Electricity), they are not related to Optics...

Professor Pablo Artal

Pablo Artal was born in Zaragoza (Spain) in 1961. He received his M.Sc. degree in Physics from the University of Zaragoza, Spain, and the Ph.D. degree in Physics (Optics) from the University Complutense of Madrid in 1988. He was a post-doctoral fellow at the Institut d'Optique, Orsay, France in 1989-90 and a senior researcher at the Instituto de Optica (CSIC) in Madrid from 1990 to 1994. Since 1994, he is full Professor of Optics at the University of Murcia, Spain.

More Information about Pablo Artal

Pablo Artal
Pablo Artal