The New iPad’s out and being gobbled up by eager fans. The main feature is a new high-definition “retina display” that’s supposed to make your old tablet feel antiquated. You might be wondering, as I was, whether this is the be-all and end-all of displays. Then again, you might be wondering, what’s next? I asked the good folks at DisplaySearch to answer some fundamental questions that have been bugging me for a bit.
First some background on the display. It measures 9.7 inches diagonally, with 2048 x 1536-pixel resolution at 264 ppi. For comparison, HDTV requires a minimum of 1920 x 1080. Apple calls it a retina display, because “at normal viewing distances” the human eye can’t distinguish individual pixels. People seem to really like it.
First question: If a human can’t distinguish individual pixels in this display, is there a point to going to even higher resolutions? Paul Semenza, a senior vice president at DisplaySearch says yes. Yes there is.
“What the eye can resolve is determined by distance as well as how good your eyesight is,” he points out. So the pixels in smartphones and other devices that you hold close to your face would need to be tinier for them to go unnoticed. And indeed they do. The iPhone 4, the first device to wear Apple’s “retina display” mantle, has a higher resolution at 326 ppi.
“The tablet is an interesting case because it can be viewed at different distances,” he says. Apple probably did some careful figuring about how close people were when surfing the web or reading a book and probably even factored in the average distance for playing Angry Birds.
If users start holding tablets closer, a higher pixel density will come in handy.
There are, of course, other reasons to improve display resolution. For one, it might simplify devices, by allowing them to display images in their native formats rather than in compressed or interpolated ones. A device maker might also want to offer crazy things like true HD in 3-D. Most of today’s 3-D depends on splitting the display in two so half the pixels are for the left eye and the other half are for the right. Doing that in full HD would require even better resolution than what the New iPad offers.
So yes. There is a reason to keep shrinking pixels.
Second question: The New iPad’s display is really pushing the limits of its technology, right? Can we go further?
Probably. “There’s a limit out there,” says Semenza. “But it keeps getting pushed out.”
The New iPad and other tablets use LCDs made up of pixels, each of which blocks or lets pass its little patch of light. The light comes from one or more LEDs, and whether it’s blocked or passed depends on an array of thin-film-transistors. Today these are made of amorphous silicon—silicon with no crystal orientation—and that’s one of the fundamental limits.
Electrons move pretty slowly in amorphous silicon, so you need a certain size transistor to produce enough current to properly control a pixel. Keep in mind that the TFT itself blocks some of the light. You can shrink the pixel quite a bit, but not the transistor. And as the pixel and transistor get closer in size, more and more of the light is blocked. You can already see it at work in the New iPad, which required twice as many LEDs as the iPad 2. “Going from 2 to New, there was a hit, in that the [area for light to escape] went down,” says Semenza.
Replacing the amorphous silicon with something faster means you can shrink the TFT. So for high-resolution, phone-size displays such as the iPhone 4, manufacturers use low-temperature polycrystalline silicon. As it’s name implies, the material has some crystallinity to it, and so it moves charge faster. However, it’s made by starting with amorphous silicon and then using a laser to anneal the stuff into its polycrystalline state. And that’s expensive—too expensive, so far, for a tablet-size screen.
Polysilicon isn’t the only alternative, either. Sharp, which is contracted to supply the New iPad’s display but has so far not delivered, has developed a process for making TFTs out of a class of semiconductor called amorphous oxides—specifically amorphous indigum-gallium-zinc oxide (a-IGZO). In the May 2011 issue of IEEE Spectrum, John F. Wager and Randy Hoffman told us all about those new semiconductors. Amorphous oxides have electron mobilities that are some 20 to 40 times faster than amorphous silicon, and they don’t need that costly laser-annealing step. So you could have your tiny transistors for cheap. But Sharp’s process apparently wasn’t ready for the New iPad’s release, so we’ll have to wait and see if it lives up the promise.
Frankly, Semenza would have been surprised if Apple went with an amorphous oxide screen even if it had been ready. The company isn’t typically the first to try out radically new technology, so don’t expect crazy things like OLED displays in Apple products until they’re pretty common elsewhere. Says Semenza: “If I had a dime for every time somebody asked when Apple was going to do OLED…”
Samuel K. Moore is the senior editor at IEEE Spectrum in charge of semiconductors coverage. An IEEE member, he has a bachelor's degree in biomedical engineering from Brown University and a master's degree in journalism from New York University.