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
Death666Angel - Sunday, September 16, 2012 - link
That seems to be true. But I don't see how that influences his conclusion that Cortex A7 is not enough of a step up in performance.From where I'm standing, A15 is still a possibility, but highly unlikely given the time frame we are talking about. And I have no reason not to trust Anand's word that it isn't A15.
Honestly, I don't care that much either way right now. I'm not going to buy an iPhone and I will be eating up any hard facts about the SoC regardless of whether its a new design of Apple's take on an A15. It's interesting in either case. :D
MrSpadge - Sunday, September 16, 2012 - link
Not sure it has been said before.. but overall power consumption doesn't scale exponentially with voltage. Some leakage may scale exponentially, but that's only a small part. Basic dynamic switching consumption scales with V^2, overall it's about V^3 for an entire modern chip - if I remember something I read during the last 1 or 2 years correctly. I think it was on Anandtech ;)>I should probably give Apple's CPU team more credit in the future.
That's only fair - however, if they don't tell anyone about the new design, they can't expect to be given credit.
KPOM - Sunday, September 16, 2012 - link
It's a dual-core 1.0GHz model. It scored 1601 vs. 629 for the iPhone 4S and 1560 for the S3.http://forums.macrumors.com/showthread.php?t=14432...
It's pretty impressive.
shank2001 - Sunday, September 16, 2012 - link
And considering that geekbench utilizes all cores, and the S3 is a quad core device, the real world single threaded performance gap will be MASSIVE!Hardly anything running on a phone whether iOS or android is multithreaded.
akdj - Sunday, September 16, 2012 - link
Exactly! Half the clock speed...half the RAM, and half the cores, yet just as 'quick' with what I'm sure will be a significant savings in 'power' accomplishing the same efficiency, speed and pure--end user enjoymentThese speeds from Geekbench are very impressive. Comparing to the actual dual core SIII sold in the US, there's an even larger 'gap' in performance as the international version is the only one sold with 4 cores. This is an engineering marvel and a big surprise. Apple is now ready and executing their own chip architecture. Married with designing their own OS, even bigger gains in perceived performance will be noticable. I'm a developer on both platforms and own the 3GS, 4 and 4S...as well as the Galaxy Note, Nexus and GSIII. My daily drivers are the 4s and GNote (the latter strictly for business apps and functionality). Regardless of the apps I use that are ambidextrous (both Android and iOS), the 4s, to this day, are more fluid them their counter part 'Droid app. I'm blown away by the processing power in these mobile devices...but even more excited by what Apple continues to achieve using less power, less speed, and less memory.
My hope is Google will follow Apple's lead wrestling control of their OS back from the manufacturers and providers...not allowing the 'bloat-ware' and skins that are dragging out the update/upgrade process and bogging down system performance. ICS and Jelly Bean are excellent updates but the fluidity and efficiency of iOS and the SDK from Apple make development a much easier endeavor. As well, knowing when and what you're users/customers are using as far as the OS build is invaluable.
This is a phenomenal engineering improvement. A huge pat on the back to the chefs (engineers) at Apple. Thinner, lighter, faster...better screen and color gamut, twice the speed (of even the 'new' iPad), screaming graphics AND as good or better battery life? Holy. Shit! Accolades to Apple. That's true innovation.
Jeremy
name99 - Monday, September 17, 2012 - link
You don't have to guess, you can compare the single vs multi-core benchmarks line by line. They're a mixed bag --- A6 is usually ahead, but not always.BUT A6 is running at 1 GHz vs 1.4GHz for SGS III, which probably means A6 is burning half the power or so. It also suggests that if Apple ever get desperate enough that they need a quick speed bump to compete in the market, they can do so, boosting either the speed or adding an additional core or two, whereas Samsung has less of this sort of flexibility.
Finally Apple, just like with iPad3, are obsessed with memory bandwidth --- this device has about 1.5 to 2x what SGS III offers, depending on the benchmark. I wonder how much of this reflects company memory of the past. Back in the PPC days, this was our biggest flaw compared to Intel, the thing that made me, as an Apple SW engineer focussed on optimization and performance, weep in frustration --- we had better CPU cores, as long as you stayed in L1, cache, but our memory systems were so crippled that the core never had a chance to shine.
We seem to be living through the mirror image of those days --- now the Apple HW engineers aren't going to make that mistake again, while the competition hope that synthetic (ie core-only) benchmarks will hide their memory system real-world limitations.
RogerShepherd - Sunday, September 16, 2012 - link
If you look at the breakdown of the Geekbench scores, it's in line with a Cortex A15 being used http://bloggershepherd.blogspot.co.uk/2012/09/more...Death666Angel - Sunday, September 16, 2012 - link
Interesting. Also, disgusting forum posts over there. Yikes.tipoo - Sunday, September 16, 2012 - link
Geekbench also got the frequency, 1GHz. Even more impressive that it's faster than Krait at 1.5GHz. But then, Krait is what, a a year or so old.Didn't think I'd feel it, but the mobile SoC wars have gotten more interesting than the desktop CPU ones. At least there's leapfrogging going on here.
Brayan Schroeder - Sunday, September 16, 2012 - link
Ok, but the Apple A6 is better than ARM Cortex A15? Thanks.