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
Freakie - Sunday, September 16, 2012 - link
I agree that there is probably more here than even Anand is acknowledging could be (also Anand seems to be trying to throw as much good light as he can at Apple for this). Apple already has used ARM ISA designs in the past, and even went further and tweaked the Samsung's to make it just a bit more like how they want (or rather, PA Semi did that I believe... highly doubt there are any engineers actually at Apple that can do that stuff when they already own companies that do it). I doubt that this situation is any different than that previous situation, where they have just taken an already customized package and customized it some more. Of course, I think we wont know that for-sure until we get die-shots of the thing, like how we were able to tell that their previous one was almost exactly a Samsung processor.Penti - Sunday, September 16, 2012 - link
Well the A4 was for example called the Apple A4 even when it was an Samsung and Intrinsity project for their then pretty small Apple customer, later they bought Intrinsity and came up with A5 and A5X and die-shrinks all on Samsungs fab with their help, they rebadged it before they bought Intrinsity so of course the first product was largely a Samsung project with Intrinsity as a subcontractor.It would be fun if it's a custom chip here, but I'm kind of not leaning towards that because it largely didn't bring in any new features or enhancements they bragged about. It's certainly possible to upgrade a A9 with the FPU from the A5/A7 or whatever that is already on the market. The problem with the A5/A7 configuration of the FPU would be that it's a reduced registry D16-variant. But they probably have the FPU done already for the A15 proper, we have to wait and see how they went about creating this chip. Odd timing if it's a new custom architecture chip.
I'm pretty sure the PA-Semi and Intrinsity teams have no external work anymore as they were merged with Apple totally and don't exist as separate units any more. To bad that PA-Semi's processor went to do largely nothing. Intrinsity also helped other firms to design PPC and MIPS processors, and did ARM-work for others than Samsung in Texas were they were based, too. Basically they did specialist design tools for the business and collaborated and co-design stuff. I'm sure that other firms in the Texas area has snatched up people that doesn't work for Apple any more, any way we won't see their work in anything more then in Apple chips. It's sadly a downside of mergers if they don't want to keep and develop the product line. Apple is largely a retailer (as most people work in Apple stores and related) and a software company but they should have a few hundred semi design people from their buyouts alone, it's not like Dangers buyout by Microsoft or Microsoft's Nokia partnership here at least as they do deliver products. I just noted that they didn't say "totally new design, new powerful GPU etc etc". It just sounded like here, here's a new chip buy it :) Intrinsity team is the ARM guys. Several of the key PA-Semi guys left Apple after the acquisition or even months before but they should have plenty of skilled engineers still. Not sure they need such a large design team if they are only going to cook ARM-SoC's together with other IP-suppliers (physical IP) and manufacturers though, not sure how many is left or has been hired, we just have to see what they have come up with. Old Intrinsity team is the one who are best skilled at optimizing external design.
Fx1 - Saturday, September 15, 2012 - link
I knew ANAND shouldnt have said it was a A15. So cock sure of himself on this one and boy was he wrong.I think an apology is in order.
DigitalFreak - Saturday, September 15, 2012 - link
No one cares.jenjohnson88 - Saturday, September 15, 2012 - link
Many people said it couldn't be A15, and Anand/Brian arrogantly said it was.you'd better believe we care that something was claimed as FACT when it was merely a guess.
JarredWalton - Sunday, September 16, 2012 - link
So, maybe Anand should do an apology article where he says something like, "I was wrong, it's not A15. Here's what I know." Oh, wait.. that's exactly what this article is.tipoo - Saturday, September 15, 2012 - link
Lol. An apology? They try to inform us the best they can, sometimes they make mistakes. Get over it.jenjohnson88 - Saturday, September 15, 2012 - link
you don't get it do youthere's a difference between saying it's an educated guess, and what Brian did.
On twitter he said he had insider information confirming it was A15.
Kristian Vättö - Sunday, September 16, 2012 - link
And I bet he had insider information, but that ended up being incorrect. There is also a difference between Twitter and Anandtech.com. If you look at our articles about iPhone 5, we have not said it's A15 for sure:"Based on the performance gains, Apple's history of SoC naming and some other stuff we've heard recently, it looks like Apple has integrated two ARM Cortex A15 on Samsung's 32nm LP HK+MG process."
http://www.anandtech.com/show/6280/apple-iphone-5-...
"Apple reserves major Ax SoC number iterations for architecture changes, combine that with the performance claims as well as some other stuff we've heard offline and there's one conclusion: the iPhone 5 uses ARM Cortex A15 cores inside."
http://www.anandtech.com/show/6279/apples-iphone-5...
Fx1 - Sunday, September 16, 2012 - link
I read the stories and there was one titles A15 inside A6 CPU. Thats pretty damn definitively written and highly misleading.There is just complete Apple Fanboy style reporting on Anandtech now. I own a MBP and had 5 iPhones before my S3 so i know how to spot them.
Time to take off the rose tinted glasses Anand. The IPhone 5 is a rip off at £529 with £105 worth of components. MAYBE YOU SHOULD REPORT ON THE COMPLETE OVERPRICING OF APPLE PRODUCTS.