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
cocoviper - Sunday, September 16, 2012 - link
I think you're forgetting there's a larger display and additional radio to drive. The fact that they held the line on power performance with those added and no real battery increase suggests tremendous power savings at the SoC level.lilmoe - Sunday, September 16, 2012 - link
They also have a larger battery to compensate for that larger screen and the new radios. I'd say this new SoC has the same power consumption as the previous A5 at best...Lucian Armasu - Sunday, September 16, 2012 - link
The battery life with LTE will be influeced only by how efficient the chip is. If they are using a 28nm radio, it should be pretty efficient, but probably not more than other companies.Death666Angel - Sunday, September 16, 2012 - link
As I understand it, smart software can also enhance LTE battery life. But mostly it's the chip and process node.Lepton87 - Saturday, September 15, 2012 - link
Was I the only one to notice that 2x the performance of A9 at 800MHz is actually pathetic? Anand always seems to drool over anything apple related, numbers be damned. Krait at 1.5GHz is going to mop the floor with that thing. Even Tegra 3 T33 will have comparable ST CPU performance with that thing. (A9 at 1.7GHz) This is already 2x faster than iPhone 4s CPU. Those numbers are actually PATHETIC and yet anand drools over that.darkcrayon - Saturday, September 15, 2012 - link
Well, if Apple's general claims are correct, that it has 2x the CPU and 2x the GPU of the iPhone 4S, the performance should be quite good. Remember, Apple didn't say twice the performance of "an A9". We're talking about the actual device here. The iPhone 4S is competitive with phones on the market with twice the clock speed at CPU tasks, and is faster than almost everything in GPU tasks.This new iPhone is clearly going to be no slouch in the performance department, regardless of your lack of interest because Apple produced it.
EnzoFX - Saturday, September 15, 2012 - link
Seriously. Biased much? So many here are getting lost in numbers. It remains to be seen if the hardware, combined with better optimized software (the #1 thing Apple has over any android offering) will mean pages are actually rendered faster, etc.In regards to graphics, I don't even really look at them these days. It seems Apple is simply uncontested in the graphics department. This can be bad, they need the competition to become more aggressive. Though this update I'm sure will be an iterative one, and will still be great, and still uncontested!
ltcommanderdata - Saturday, September 15, 2012 - link
Despite your complaints, I don't remember reading Anand claiming the A6 has superior performance to other current or upcoming high-end SoC. Instead the focus is on how to increase performance in the most power efficient way. While Apple has traditionally focused on having leading GPU performance in their SoC, they haven't done the same for the CPU. As such, it wouldn't be surprising if the A6 is merely competitive, but is not the fastest CPU. If the reason Apple moved to a custom design is to increase power efficiency, the question of how successful they were needs to be answered by comparing performance/watt between SoC designs rather than raw maximum performance. I'm not sure there are very good ways to benchmark performance/watt between various mobile devices though.cocoviper - Sunday, September 16, 2012 - link
Actually if you look at past benchmarks Apple's dual A9s at 800MHz pretty much tie with Tegra3 (due to more efficient software).http://images.anandtech.com/graphs/graph6098/49612...
2x that performance puts it squarely in the realm of Krait or slightly better.
lilmoe - Sunday, September 16, 2012 - link
Try again with Jelly Bean and a faster browser. Really, those Sunspider benchmarks don't show the whole picture when a webpage is loading and being rendered on the screen...But if you want to see a more valid comparison, I'd suggest you wait for IE10 on Windows Phone 8 (highly optimized for the Snapdragon S4, unlike Android) and compare the browser speeds and general OS performance on that with the iPhone 5... You'll have a better picture.