An Update on Apple’s A7: It's Better Than I Thought

When I reviewed the iPhone 5s I didn’t have much time to go in and do the sort of in-depth investigation into Cyclone (Apple’s 64-bit custom ARMv8 core) as I did with Swift (Apple’s custom ARMv7 core from A6) the year before. I had heard rumors that Cyclone was substantially wider than its predecessor but I didn’t really have any proof other than hearsay so I left it out of the article. Instead I surmised in the 5s review that the A7 was likely an evolved Swift core rather than a brand new design, after all - what sense would it make to design a new CPU core and then do it all over again for the next one? It turns out I was quite wrong.

Armed with a bit of custom code and a bunch of low level tests I think I have a far better idea of what Apple’s A7 and Cyclone cores look like now than I did a month ago. I’m still toying with the idea of doing a much deeper investigation into A7, but I wanted to share some of my findings here.

The first task is to understand the width of the machine. With Swift I got lucky in that Apple had left a bunch of public LLVM documentation uncensored, referring to Swift’s 3-wide design. It turns out that although the design might be capable of decoding, issuing and retiring up to three instructions per clock, in most cases it behaved like a 2-wide machine. Mix FP and integer code and you’re looking at a machine that’s more like 1.5 instructions wide. Obviously Swift did very well in the market and its competitors at the time, including Qualcomm’s Krait 300, were similarly capable.

With Cyclone Apple is in a completely different league. As far as I can tell, peak issue width of Cyclone is 6 instructions. That’s at least 2x the width of Swift and Krait, and at best more than 3x the width depending on instruction mix. Limitations on co-issuing FP and integer math have also been lifted as you can run up to four integer adds and two FP adds in parallel. You can also perform up to two loads or stores per clock.

I don’t yet have a good understanding of the number of execution ports and how they’re mapped, but Cyclone appears to be the widest ARM architecture we’ve ever seen at this point. I’m talking wider than Qualcomm’s Krait 400 and even ARM’s Cortex A15.

I did have some low level analysis in the 5s review, where I pointed out the significantly reduced memory latency and increased bandwidth to the A7. It turns out that I was missing a big part of the story back then as well…

A Large System Wide Cache

In our iPhone 5s review I pointed out that the A7 now featured more computational GPU power than the 4th generation iPad. For a device running at 1/8 the resolution of the iPad, the A7’s GPU either meant that Apple had an application that needed tons of GPU performance or it planned on using the A7 in other, higher resolution devices. I speculated it would be the latter, and it turns out that’s indeed the case. For the first time since the iPad 2, Apple once again shares common silicon between the iPhone 5s, iPad Air and iPad mini with Retina Display.

As Brian found out in his investigation after the iPad event last week all three devices use the exact same silicon with the exact same internal model number: S5L8960X. There are no extra cores, no change in GPU configuration and the biggest one: no increase in memory bandwidth.

Previously both the A5X and A6X featured a 128-bit wide memory interface, with half of it seemingly reserved for GPU use exclusively. The non-X parts by comparison only had a 64-bit wide memory interface. The assumption was that a move to such a high resolution display demanded a substantial increase in memory bandwidth. With the A7, Apple takes a step back in memory interface width - so is it enough to hamper the performance of the iPad Air with its 2048 x 1536 display?

The numbers alone tell us the answer is no. In all available graphics benchmarks the iPad Air delivers better performance at its native resolution than the outgoing 4th generation iPad (as you'll soon see). Now many of these benchmarks are bound more by GPU compute rather than memory bandwidth, a side effect of the relative lack of memory bandwidth on modern day mobile platforms. Across the board though I couldn’t find a situation where anything was smoother on the iPad 4 than the iPad Air.

There’s another part of this story. Something I missed in my original A7 analysis. When Chipworks posted a shot of the A7 die many of you correctly identified what appeared to be a 4MB SRAM on the die itself. It's highlighted on the right in the floorplan diagram below:


A7 Floorplan, Courtesy Chipworks

While I originally assumed that this SRAM might be reserved for use by the ISP, it turns out that it can do a lot more than that. If we look at memory latency (from the perspective of a single CPU core) vs. transfer size on A7 we notice a very interesting phenomenon between 1MB and 4MB:

That SRAM is indeed some sort of a cache before you get to main memory. It’s not the fastest thing in the world, but it’s appreciably quicker than going all the way out to main memory. Available bandwidth is also pretty good:

We’re only looking at bandwidth seen by a single CPU core, but even then we’re talking about 10GB/s. Lookups in this third level cache don’t happen in parallel with main memory requests, so the impact on worst case memory latency is additive unfortunately (a tradeoff of speed vs. power).

I don’t yet have the tools needed to measure the impact of this on-die memory on GPU accesses, but in the worst case scenario it’ll help free up more of the memory interface for use by the GPU. It’s more likely that some graphics requests are cached here as well, with intelligent allocation of bandwidth depending on what type of application you’re running.

That’s the other aspect of what makes A7 so very interesting. This is the first Apple SoC that’s able to deliver good amounts of memory bandwidth to all consumers. A single CPU core can use up 8GB/s of bandwidth. I’m still vetting other SoCs, but so far I haven’t come across anyone in the ARM camp that can compete with what Apple has built here. Only Intel is competitive.

 

Introduction, Hardware & Cases CPU Changes, Performance & Power Consumption
POST A COMMENT

444 Comments

View All Comments

  • StigtriX - Tuesday, November 12, 2013 - link

    Actually, the new Kindle 8.9 has at least as good a display as the iPad Air, if not better:

    "Amazon’s always touted the displays on its tablets, and the screen it’s using on the HDX 8.9 is a doozy. At a ridiculous 2560 x 1600-pixel resolution and 339ppi, it’s either higher-resolution or more pixel-dense than virtually any other tablet on the market. It also has great viewing angles, excellent color reproduction, and high brightness levels. It’s really just a great screen, whether you’re gaming, watching video, or reading text — it’s right up there with, if not better than, the Retina display on the iPad Air."
    - The Verge

    "While the speakers feel like something of an afterthought, Amazon is clearly waging a battle on the display front. The company keeps upping its game, and indeed, the screen here dazzles, with 2,560 x 1,600 resolution and a pixel density of 339 pixels per inch. That's a big jump up from the HD 8.9's 1,920 x 1,200 display, not to mention the new HDX 7-inch tablet, which has a 1,920 x 1,200 screen. It's even enough to make the iPad Air's 2,048 x 1,536 resolution (264 ppi) seem modest. It's simply a gorgeous thing to behold, making movie watching a downright pleasure. Heck, it even managed to make Sharknado look pretty good, which is no small feat. All in all, images are sharp, the level of detail is impressive and the colors are vibrant."
    - Engadget

    But, I do agree with you that too many let their personal opinions colour their reading of reviews. I prefer Android and PC, but I have an iPad and have tried several Macs. Even though I prefer other units, there is no point in denying the excellent quality and style Apple products come with. They, like Nintendo, only release products "when they are done" - no rushing.
    Reply
  • MassiveTurboLag - Tuesday, October 29, 2013 - link

    Really? You clearly never looked over one of their Android phone reviews. They are stupidly detailed. Reply
  • dsumanik - Wednesday, October 30, 2013 - link

    Read this review with a grain of salt. Anand lai shimpi is heavenly vested in apple stock, doing everything he can to boost the dismal situation.

    Thinner bezels and light weight do not hide the fact that functionally, this iPad is the same as the previous 2 generations.

    Sent from my ipad3, which will be upgraded when apple actually updates the product line.

    Here's some basic ideas mr cook:

    Wireless charging
    Fingerprint scanner
    Thunderbolt sync or usb3
    Haptic feedback
    NFC
    Reply
  • Solon - Wednesday, October 30, 2013 - link

    Wireless charging - deeply inefficient charging; maybe 50% of energy is actually picked up by the device
    Fingerprint scanner - probably getting one next gen
    Thunderbolt sync or USB3 - maybe; and it is possible the hardware already can do this; but Thunderbolt (a PCI-E spec) will never happen
    Haptic feedback - dead concept; no one actually wants this; RIM failed at haptic
    NFS - in a tablet? and is NFC actually used? No, it isn't.
    Reply
  • dishayu - Wednesday, October 30, 2013 - link

    Agree with your wireless charging argument, but that limits you from charging your iDevice if you don't have your own cable. All android phones use microusb, so it's not hard to find one at all (speaking of the real world here). Apple chose not to afford themselves that luxury.

    Fingerprint scanner : Although I think it's a gimmick and useless for the masses (almost everyone except from corporate users), they could have done it now, since they already have it in 5s. Why wait till next gen?

    Haptic feedback : You probably would have said the same about touchscreens around 2005. I tried staying away from smartphones altogether until the Xperia Pro because there were no decent phone with a proper keyboard. After that I finally gave in an bought the HTC One and I still really miss my physical buttons. Haptic feedback could cure that.

    NFC : I use it as much as I use Bluetooth (which is to say not a whole lot but a very handy feature still and I'd like to have it)
    Reply
  • ws3 - Tuesday, October 29, 2013 - link

    Yeah. Anyone who wanted to be truly objective would stopping running all of those tests that the iPad Air does so well in. If you want to be truly balanced, you have to run a bunch of tests that the iPad sucks at too, like maybe an .apk loading test or a malware running test. Reply
  • zeagus - Wednesday, October 30, 2013 - link

    +lulz Reply
  • KoolAidMan1 - Wednesday, October 30, 2013 - link

    Perfect Reply
  • pdjblum - Thursday, October 31, 2013 - link

    Anand himself has stated where his loyalties lie: He said during a podcast that his advertisers were the motivating force behind the change in the look of the website. Obviously he is a slave to his advertisers who obviously want as much crApple content on the site as possible; of course they want it be overwhelmingly positive. This crApple model generates tons of hits for theverge and it does the same for anandtech. Reply
  • markthema3 - Monday, November 4, 2013 - link

    He changed the layout of the site to better place the advertisments on the page. That's not catering to an individual company. The conclusion that he is therefore a slave to the advertisers is unfounded. Reply

Log in

Don't have an account? Sign up now