Yesterday Apple unveiled its third generation iPad, simply called the new iPad, at an event in San Francisco. The form factor remains mostly unchanged with a 9.7-inch display, however the new device is thicker at 9.4mm vs. 8.8mm for its predecessor. The added thickness was necessary to support the iPad's new 2048 x 1536 Retina Display.

Driving the new display is Apple's A5X SoC. Apple hasn't been too specific about what's inside the A5X other than to say it features "quad-core graphics". Upon further prodding Apple did confirm that there are two CPU cores inside the SoC. It's safe to assume that there are still a pair of Cortex A9s in the A5X but now paired with a PowerVR SGX543MP4 instead of the 543MP2 used in the iPad 2. The chart below gives us an indication of the performance Apple expects to see from the A5X's GPU vs what's in the A5:

Apple ran the PowerVR SGX 543MP2 in its A5 SoC at around 250MHz, which puts it at 16 GFLOPS of peak theoretical compute horsepower. NVIDIA claims the GPU in Tegra 3 is clocked higher than Tegra 2, which was around 300MHz. In practice, Tegra 3 GPU clocks range from 333MHz on the low end for smartphones and reach as high as 500MHz on the high end for tablets. If we assume a 333MHz GPU clock in Tegra 3, that puts NVIDIA at roughly 8 GFLOPS, which rationalizes the 2x advantage Apple claims in the chart above. The real world performance gap isn't anywhere near that large of course - particularly if you run on a device with a ~500MHz GPU clock (12 GFLOPS):

GLBenchmark 2.1.1's Egypt offscreen test pegs the PowerVR SGX 543MP2 advantage at just over 30%, at least at 1280 x 720. Based on the raw FP numbers for a 500MHz Tegra 3 GPU vs. a 250MHz PowerVR SGX 543MP2, around a 30% performance advantage is what you'd expect from a mostly compute limited workload. It's possible that the gap could grow at higher resolutions or with a different workload. For example, look at the older GLBenchmark PRO results and you will see a 2x gap in graphics performance:

For most real world gaming workloads I do believe that the A5 is faster than Tegra 3, but the advantage is unlikely to be 2x at non-retinadisplay resolutions. The same applies to the A5X vs. Tegra 3 comparison. I fully expect there to be a significant performance gap at the same resolution, but I doubt it is 4x in a game.

The A5X doubles GPU execution resources compared to the A5. Imagination Technologies' PowerVR SGX 543 is modular - you can expand by simply increasing "core" count. Apple tells us all we need to know about clock speed in the chart above: with 2x the execution resources and 2x the performance of the A5, Apple hasn't changed the GPU clock of the A5X.

Assuming perfect scaling, I'd expect around a 2x performance gain over Tegra 3 in GLBenchmark (Egypt) at 720p. Again, not 4x but at the same time, hardly insignificant. It can take multiple generations of GPUs to deliver that sort of a performance advantage at a similar price point. Granted Apple has no problems eating the cost of a larger, more expensive die, but that doesn't change the fact that the GPU advantage Apple will hold thanks to the A5X is generational.

I'd also point out that the theoretical GPU performance of the A5X is identical to what Intel is promising with its Atom Z2580 SoC. Apple arrives there with four SGX 543 cores, while Intel gets there with two SGX 544 cores running at ~2x the frequency (533MHz vs. 250MHz).

With the new iPad's Retina Display delivering 4x the pixels of the iPad 2, a 2x increase in GPU horsepower isn't enough to maintain performance. If you remember back to our iPad 2 review however, the PowerVR SGX 543MP2 used in it was largely overkill for the 1024 x 768 display. It's likely that a 4x increase in GPU horsepower wasn't necessary to deliver a similar experience on games. Also keep in mind that memory bandwidth limitations will keep many titles from running at the new iPad's native resolution. Remember that we need huge GPUs with 100s of GB/s of memory bandwidth to deliver a high frame rate on 3 - 4MP PC displays. I'd expect many games to render at lower resolutions and possibly scale up to fit the panel.

What About the Display?

Performance specs aside, the iPad's Retina Display does look amazing. The 1024 x 768 panel in the older models was simply getting long in the tooth and the Retina Display ensures Apple won't need to increase screen resolution for a very long time. Apple also increased color gamut by 44% with the panel, but the increase in resolution alone is worth the upgrade for anyone who spends a lot of time reading on their iPad. The photos below give you an idea of just how sharp text and graphics are on the new display compared to its predecessor (iPad 2, left vs. new iPad, right):

The improvement is dramatic in these macro shots but I do believe that it's just as significant in normal use. 

Apple continues to invest heavily in the aspects of its devices that users interact with the most frequently. Spending a significant amount of money on the display makes a lot of sense. Kudos to Apple for pushing the industry forward here. The only downside is supply of these greater-than-HD panels is apparently very limited as a result of Apple buying up most of the production from as many as three different panel vendors. It will be a while before we see Android tablets with comparable resolutions, although we will see 1920 x 1200 Android tablets shipping in this half.

The CPU

TI was one of the earliest partners with ARM on the Cortex A15 and silicon just came back from the fab at the beginning of this year. Even if Apple were similarly instrumental in the definition of the Cortex A15 architecture, it would be Q3 at the earliest before it could have working silicon available in volume. With no A15 design ready and presumably no desire to jump into the custom-designed ARM CPU market quite yet, Apple once again turned to the Cortex A9 for the A5X.

Apple confirmed that there are only two Cortex A9 cores on the A5X and it neglected to mention operating frequency. I suspect the lack of talk about CPU clocks indicates that they perhaps haven't changed. We could still be looking at a 1GHz max operating frequency.

Although we've speculated that Apple moved to a 32nm design with the A5X, it is entirely possible that we're still dealing with mature 45nm silicon here. It would explain the relatively conservative GPU clocks, although the additional GPU cores would balloon die size to 150 - 160mm^2 (roughly twice the size of Tegra 3). If A5X is 32nm, assuming a relatively conservative 80% scaling factor Apple would be able to maintain a die size of around 125mm^2, similar to the previous generation A5.

A quad-core CPU design does make some sense on a tablet, but only one that is either running heavily threaded workloads or is subjected to pretty intense multitasking. As we found in our iPhone 4S review, many iOS apps are still not very well threaded and have a difficult time utilizing two cores, much less four. On the multitasking front, Apple has enabled task switching but there's still no way to run two applications side-by-side. The most CPU intensive workloads on iOS still require that the app is active in the foreground for user interaction. Apps can work in the background but it's not all that constant/common, and again, they aren't pegging multiple cores. Apple built a very efficient, low overhead platform with iOS - it had to thanks to the hardware specs of the original iPhone. A result of iOS' low-overhead, very efficient design is expectedly low CPU utilization for most tasks. This is not to say that CPU performance isn't important under iOS, just that it's hard to find apps that regularly require more than a single core and definitely hard to find those that can benefit from more than two cores.

I will say though, Apple could easily add more cores if it wanted to spend the die area without a significant impact on power consumption. Remember that idle cores can be fully power gated, effectively reducing their power consumption while idle to zero. Apple could also assume a fairly conservative CPU governor and only wake up the third and fourth cores when absolutely necessary (similar to what we see happening with Tegra 3 on Android).

What about the Next iPhone?

Apple has traditionally used the iPad SoC in the subsequent iPhone release that followed later in the same year. It would make sense to assume that we'll see a smartphone version of the A5X SoC (at lower clocks) later this year. The A6? That'll probably debut next year with the 4th generation iPad.

Memory Capacity

Apple wouldn't let us run any third party applications on the new iPad so we couldn't confirm the actual memory capacity of the new model. On stage at the event, Epic mentioned that the new iPad has more memory and a higher output resolution than the Xbox 360 or PlayStation 3. The Xbox 360 has 512MB of memory, and Apple's A5/A5X has a dual-channel LPDDR2 memory controller. Each channel needs to be populated evenly in order to maintain peak bandwidth, which greatly narrows the options for memory capacity on the new iPad. 768MB would imply 512MB on one channel and 256MB on the other, delivering peak performance for apps and data in the first 512MB but lower performance for the upper 256MB. Given the low cost of DRAM these days, I think it's safe to assume that Apple simply went with two 512MB DRAM devices in a PoP configuration on the A5X for a total of 1GB of LPDDR2 memory in the new iPad.

4G LTE Support

Brian did an excellent analysis on the LTE baseband in the new iPad here. Qualcomm's MDM9600, a 40nm design appears to be used by Apple instead of the 28nm MDM9615. In hindsight, speculating the use of a 28nm LTE baseband for the new iPad was likely short sighted. Apple had to be in the mass production phase for the new iPad somewhere in the January/February timeframe. Although 28nm silicon is shipping to customers today, that was likely too aggressive of a schedule to make it work for an early-March launch.

Apple offers carrier specific iPad 4G models on AT&T and Verizon, although both versions can roam on 3G networks around the world. Apparently the iPad 4G isn't SIM locked, so you'll be able to toss in a SIM from other carriers with compatible networks. LTE data plans are available from AT&T and Verizon with no long-term contract:

The Name

Apple surprised many by referring to the 3rd generation iPad simply as "the new iPad". The naming seems awkward today, but it's clearly a step towards what Apple does across many of its product lines. The MacBook Air, MacBook Pro and iPod all receive the same simple branding treatment; newer models are differentiated by a quietly spoken year or generation marker.

I still remember back several years ago when PC OEMs were intrigued by the idea of selling desktops based on model year and not on specs. Apple has effectively attained the holy grail here.

A Much Larger Battery

Apple claimed no decrease in battery life for the new iPad compared to last year's model and only a 1 hour drop over LTE. The problem is that the combination of A5X SoC under GPU load, the LTE baseband and driving/lighting all of those pixels in the Retina Display has a significant impact on power consumption.

Apple addressed the issue by increasing the new iPad's battery capacity by 70%. If the leaked PCB photos are accurate (they look to be), Apple increased battery volume by shrinking the motherboard size and increasing the thickness of the tablet.

The new 42.5Wh battery is downright huge. To put the new iPad's battery in perspective, this is nearly the same battery capacity as the what was shipping in the 2008 13-inch MacBook Pro. This is also a bigger battery than what's used in the 2011 11-inch MacBook Air:

Over the next two years you can expect to see the line between ultraportable and tablet blur considerably. Looking at where the new iPad falls in the chart above really begins to exemplify just how blurry that line is going to become.

With the display off, the new iPad looks and feels a lot like the iPad 2. The additional thickness is hard to see, but the additional weight is definitely noticeable. The new iPad isn't as heavy as the original model, but it's clearly heavier than the iPad 2. I don't believe the added weight is a deal breaker, but it is a step backwards. Maintaining battery life however obviously trumps added weight.

The math is pretty simple. If Apple is claiming 10 hours of battery life with a 42.5Wh battery, the new iPad with the iPad 2's battery would likely be good for just under 6 hours. Such a drop would be unacceptable and thus the new iPad gets a bigger battery and incurs additional weight from the new battery and display components.

Final Words

In many ways the new iPad was a known quantity. We knew to expect a faster SoC, a significantly higher resolution display and LTE support - Apple delivered on all fronts. The new iPad, much like another iPhone, is simply a tangibly improved version of its predecessor.

The iPad 2's display quickly became unacceptable from a resolution standpoint. The 3rd generation iPad's Retina Display completely addresses the issue and creates a new benchmark for other players in the tablet and ultraportable notebook space to live up to. It really is great to see Apple pushing display technology so aggressively and at reasonable price points. I do hope it's only a matter of time before we see a similar trend on the Mac side.

The finer details of yesterday's announcement were interesting - a much larger battery and 4x-nm LTE baseband. Arguably the most important information however is what Apple didn't talk about.

Today we have a first-world-problem with tablets, including the iPad - they are spectacular for certain usage models, but frustrating for others. Tablets aren't notebook replacements yet, but they can be more useful than a notebook depending on what you're doing. At the same time, tablets can be considerably worse than a notebook - again, depending on what you're doing. The solution to having the best of both worlds is to switch between or travel with two devices: a tablet and a Mac/PC. Ideally we'd like to see consolidation where you'd only need one.

Windows 8 proposes a solution to this problem: a single OS that, when paired with a convertible tablet (or dockable tablet like the Transformer Pad), can give you a tablet experience or a full blown desktop OS on a single device. Apple hasn't tipped its hand as to what the iOS UI strategy is going forward. I suspect we'll get some update at WWDC this year, but Apple is playing it very quiet at this point. Microsoft's strategy does bode very well for Windows users who also want a tablet, however it does alienate Windows users who want a more robust desktop experience. It's clear to me that Apple is trying to move the iPad closer to the MacBook Air in its product line, but it's unclear to me whether (or when) we'll see convergence there.