What resolution would a “Retina display” 27-inch screen be?

Earlier, via @viticci, I saw a tweet from @marcedwards that said:

Retina 27” Thunderbolt display: 5120×2880 = 14,745,600 px

4K film: 4096×2160 = 8,847,360 px

Retina iPad 3: 2048×1536 = 3,145,728 px

This is based on the widely-held but fallacious belief that a “Retina display” is twice the resolution of a current-day display. There’s another widely-held but equally incorrect belief that to qualify as “retina” a screen must have more than 300 pixels-per-inch. I have previously debunked both of these for TUAW. In fact, Apple defines the term like so: “[a] Retina display’s pixel density is so high, your eye is unable to distinguish individual pixels”. This means the truth about what is and isn’t a Retina Display is more subtle, and demands consideration of viewing distance as well as screen size.

Let’s use the methodology I outlined in my TUAW post to consider the 27” Thunderbolt Display.

A Thunderbolt Display has 2560x1440 pixels over a 27” screen — that works out to 109 pixels-per-inch. If I had 20/20 (i.e. average) vision, I’d have to be further than 32” back from the screen to longer be able to resolve individual pixels.

I’m typing this on my 27” iMac right now and I’ve just measured the distance from the screen to my eyes as about 28”. This is probably a typical sitting position, At this distance, the finest detail I am able to discern is a mere 123 pixels-per-inch. In other words, my 27” iMac only has to go up to 2896x1629 (or some slightly higher and more palatable numbers) resolution to be a “retina” display at my current sitting distance. This is far lower than the resolution of a 4K film.

Footnote: I’ve glossed over some interesting details with regards to different types of visual acuity for different types of image, which you might like to read about.

Thunderbolt and iOS

MG Seigler, for Techcrunch:

My hunch is that we will hear something from Apple later this year about Thunderbolt use with iPads/iPhones. Perhaps during the iPhone 5 unveiling in the fall. Now that the technology is out there on two of their most popular devices (MacBook Pros and iMacs) and probably pretty soon on another one (MacBook Airs), Apple will have to address this. And the reality is that Thunderbolt seems to be a more viable and smart near-term solution for device sync rather than a full-on cloud sync.

…

One big problem that users have with the iTunes-based syncing mechanism for iPhones/iPads today is that it’s slow. Apple has done a lot of work to speed this up over the years (remember when backups would take something like 30 minutes to an hour!?), but it’s still not ideal. Thunderbolt could do a lot to alleviate this pain point.

Right now, iPhone/iPad syncing is done via USB 2.0. These ports theoretically transfer data at 480 Mbps (but it’s usually less in practice). Thunderbolt will up that speed to 10 Gbps spread over two channels. In other words, a 10-minute sync could be cut to a matter of seconds.

Two problems.

One: Thunderbolt is currently only on handful of new Macs. Suppose, for a second, that a new iOS device wouldn’t work with six-month-old Macs, or even brand-new Mac Minis, MacBook Airs, and so forth. I think it’s quite clear that’d go down very badly with users even before you factor in the hordes of people using iTunes on Windows who have no access to Thunderbolt at all.

Two: the flash RAM inside iOS devices is cheap stuff, miles and miles away from exotic (and much more expensive, and physically much more bulky) memory used to make SSDs from. It’s not capable of sustained write speeds fast enough utilise the faster interface. Moving to Thunderbolt on iOS for device syncing would be like upgrading your car’s transmission without changing the engine — it doesn’t get any faster overall when you change just one part of the drivetrain.