For me, one of the most interesting things about the iPhone 5S launch is what lives within: the new A7 SoC. One of the first shipping 64-bit ARMv8 designs, and certainly for the consumer market, folks may wonder, why? Why did Apple feel the urge to go to a 64-bit architecture, one that doesn’t have much opportunity to stretch its legs on a 4″ phone? Was it simply to keep consistent with the next generation iPad? If you believe the A#/A#X pattern will continue with this generation, that must certainly contribute to it; however, the question remains: does the move to 64-bit on an iPad make sense enough, right now, to outweigh the cost and complexity?
There are plenty of arguments to be made on that point. Instead, I think it points to a new direction. Think about the iPads you see people using. How many of them are wrapped in cases? And, how many of those cases have keyboards attached to them? Even more, how many times have you seen folks typing furiously on those iPads+keyboards, just like a laptop? The answer is, a lot, especially for a something that was originally defined as a consumption device.
Combine that tendency with strong enterprise interest in the iPad and you have a new product opportunity, a sell-up option from the basic iPad, combining an attached (or detachable) keyboard with the touchscreen, all running iOS. Think of it as a touchscreen MacBook Air, but thinner, lighter and providing all-day battery life. The A7 (or A7X variant for a tablet-sized product) would really be able to stretch its legs, injecting a lot more performance headroom and scale in a device designed for productivity.
Here’s a quick comparison of the existing MBA and iPad: MacBook Air 11.6″
- Intel Core i5
- 4GB memory
- 128GB SSD
iPad (4th Gen)
- Apple A6X SoC
- 1GB memory
- 128GB eMMC
There are several ways this scenario could play out. Here’s one: an “iOSBook” 64GB SKU at $899 and a 128GB SKU at $999 while maintaining similar or better margins than the existing iPad (greater margins than the existing MBA).
Compared to Qualcomm’s SoC ASPs of $22, in-house production of the A# SoC is probably about the same, given the larger die, and much, much lower than the $120 ASP Intel is purportedly getting for their chips across the board (the i5 in the MBA is likely above this average). Meanwhile, an extra 1GB of memory, larger touchscreen and the keyboard attachment (or chassis) may offset the chipset cost reduction. Now, you have an ~12″ iPad, 2GB memory in a roughly clamshell form factor.
In terms of positioning, there’s some trade-off in performance relative to the MBA, but it contains a much higher resolution display and touchscreen with all the familiarity and apps of iOS (prevalent compared to OS X). For most people, this “iOSBook” would be more than sufficient, and if peak performance is paramount, then the MBA and MBP are still there for customers’ choosing.
There are rumours from the supply chain that Apple is testing a larger display for an iPad, slightly under 13″. While it could be a big tablet, perhaps it’s the panel for something a bit different.