The iPhone 5's A6 SoC: Not A15 or A9, a Custom Apple Core Instead
by Anand Lal Shimpi on September 15, 2012 2:15 PM EST- Posted in
- Smartphones
- Apple
- Mobile
- SoCs
- iPhone 5
The A6 GPU: PowerVR SGX 543MP3?
Apple made a similar "up to 2x" claim for GPU performance. It didn't share any benchmarks, but there are four options here:
1) PowerVR SGX 543MP2 (same as in A5) at 2x the clock speed
2) PowerVR SGX 543MP4 at the same clock as the MP2 in the A5
3) Marginally higher clocked PowerVR SGX 543MP3
4) Next-gen PowerVR Rogue GPU
It's too early for #4. The first option makes sense but you run into the same issues as on the CPU side with higher voltages used to ramp clocks up (also possible that you drop voltages in the move to the new process technology).
The second option trades voltage for die area, which based on the A5X Apple is clearly willing to spend where necessary.
The third is sort of the best of both worlds. You don't take a huge die area penalty and at the same time don't run at a significantly higher frequency, and you can get to that same 2x value.
The third option is the most elegant and likely what Apple chose here. Remember that overall die size is dictated by the amount of IO you have around the chip. The A5X had four 32-bit LPDDR2 memory controllers, which gave Apple a huge die area to work with. The move to a smaller manufacturing process cuts down the total die area, which means Apple would either have to add a ton of compute (to fill empty space, no sense in shipping a big chip with a bunch of unused area) or reduce the memory interface to compensate. Pair that knowledge with the fact that Apple doesn't have the same memory bandwidth requirements on the iPhone 5 (0.7MP vs. 3.1MP display) and it makes sense that Apple would go for a narrower memory interface with the A6 compared to the A5X.
How much narrower? Phil Schiller mentioned the A6 was 22% smaller than the A5. We can assume this is compared to the 45nm A5 and not the 32nm A5r2, which would mean that we don't have any more memory channels compared to the A5. In other words, it's quite likely the A6 has a 2x32-bit LPDDR2 memory interface once again.
Final Words
There's not much more to add for now. We'll have a device in a week and I suspect the first reviews will be out a day or two before then. Then the real work begins on finding out exactly what Apple has done inside the A6. If anyone has been dying to put together some good low level iOS benchmarks, now is the time to start.
This is a huge deal for Apple. It puts the company in another league when it comes to vertical integration. The risks are higher (ARM's own designs are tested and proven across tons of different devices/platforms) but the payoff is potentially much greater. As Qualcomm discovered, it's far easier to differentiate (and dominate?) if you're shipping IP that's truly unique from what everyone else has.
Now we get to see just how good Apple's CPU team really is.
Facebook
Twitter
RSS
163 Comments
View All Comments
madmilk - Sunday, September 16, 2012 - link
Sunspider is not multithreaded (actually it's a limitation of Javascript in browsers). Sometimes the engine can use multiple cores, but there's no way the Tegra is being fully utilized here. That said, the additional two cores are hardly used elsewhere either.lilmoe - Sunday, September 16, 2012 - link
Sadly, but I also agree. First, on the CPU side, they tried to gain more popularity by providing false "exclusive" info, and now they're saying that it's "pleasant" and making it seam as if it's revolutionary (on the vertical integration side) when Qualcomm has been doing this for years now (and doing a better job on the CPU side at that).Second, about the GPU, they're still trying to speculate something TOTALLY off. Anand and crew know pretty well that Apple's claims of "2X" and "4X" are utter bull in real world performance (only in a sub-benchmarks do we get anything close to their claims). Refer to the comparison of the Apple A5X and nVidia Tegra3 for more info.
Sorry, but by the sounds of this chip, it would probably be comparable to the Exynos 4 Quad, but not nearly as powerful as the Snapdragon S4 Pro. Exynos 5 Dual will wipe the floor with all of the above...
It's really disappointing. I thought we had a reliable source on low level tech. They can be fanboys of whatever they want, but I didn't expect it to affect the way they reported tech news.
doobydoo - Sunday, September 16, 2012 - link
You question the GPU claims Apple made, yet it's common knowledge that the iPad has the fastest GPU of any tablet and the iPhone 4S, despite being a year old - still dominates the GPU benchmarks in America - beating the Samsung Galaxy S3, for example.It is virtually guaranteed that the iPhone 5 will have the fastest GPU in any smartphone when it's reviewed.
doobydoo - Sunday, September 16, 2012 - link
Oh and your comments on the CPU are nothing short of totally unfounded guesswork.Death666Angel - Sunday, September 16, 2012 - link
Considering the size of the SoC in the iPad, it better beat everything else. :PAnd the SGS3 international version is faster than the iPhone4s, so I guess it sucks to be stuck with the Krait version.
lilmoe - Sunday, September 16, 2012 - link
Fanboy much? Go back to the review that this site has provided for the Galaxy S3, the Mali400 GPU in the Exynos 4 Quad is 25% faster than PowerVR in the iPhone 4S.Benchmarks are cool, they give us a roughly good idea of what to expect out of a platform, but real world performance can differ slightly to significantly.
doobydoo - Sunday, September 16, 2012 - link
The benchmarks I referred to are from this site, and they are facts - nothing opinionated and your fanboy statement is therefore a bit illogical.I specifically stated the GPU benchmarks in 'America' - the GPU in the iPhone 4s outperforms the American version of the SG3.
The Mali 400 is only found in the international version of the SG3 - and it's only marginally faster.
Given Apples claim of 2x faster GPU you would expect the iPhone 5 to easily take the GPU crown once again.
Flunk - Sunday, September 16, 2012 - link
It doesn't really matter, it just needs to be faster than the last iPhone because it's a closed system. People don't buy the iPhone for relative performance vs Android.bill4 - Saturday, September 15, 2012 - link
Apple has to spend a fraction of it's 500 zillion in the bank on something. And it certainly isn't modernizing that piece of crap iOS...EnzoFX - Saturday, September 15, 2012 - link
They obviously don't think they need to. I'm inclined to agree. It's still simple enough for 99% of people. It's an OS. You have icons and you have functionality. The only thing it can do is evolve. People are easy to dismiss the new features and updates because the home screen still looks the same. Do you need live widgets on there? Well I'd argue most people don't. At least not in any way that Android does it. Perhaps something simpler the way WP8 or whatever does this. But that's risky with little payoff, Apple iOS is already well established. There are a ton of features added, mostly through apps, and that's an argument most will understand. That it's all about the apps.Do you yell at Windows 7 and wish it was more evolved? I'd say most people here are content with the basic start menu/icons that it offers lol. So again, it would have to be a whole new redesign, and again don't think that's warranted whatsoever. It's all about the apps that will round out the OS. Like the reason people STICK with Windows OS', because it supports all their favorite and wide range of apps. Apple is bringing the apps. Look at their focus in the last couple years, they're pushing their iLife stuff and buffing up specs, and empower devs to develop more and more. When the Smartphone needs to evolve, then iOS will prob then too. Until then, a smartphone is just a smartphone, there's not much reason to take the next big leap.