The iPhone 5 Review
by Anand Lal Shimpi, Brian Klug & Vivek Gowri on October 16, 2012 11:33 AM EST- Posted in
- Smartphones
- Apple
- Mobile
- iPhone 5
Apple's Swift: Visualized
Section by Anand Shimpi
Based on my findings on the previous pages, as well as some additional off-the-record data, this is what I believe Swift looks like at a high level:
Note that most of those blocks are just place holders as I don't know how they've changed from Cortex A9 to Swift, but the general design of the machine is likely what you see above. Swift moves from a 2-wide to a 3-wide machine at the front end. It remains a relatively small out-of-order core, but increases the number of execution ports from 3 in Cortex A9 to 5. Note the dedicated load/store port, which would help explain the tremendous gains in high bandwidth FP performance.
I asked Qualcomm for some additional details on Krait unfortunately they are being quite tight lipped about their architecture. Krait is somewhat similar to Swift in that it has a 3-wide front end, however it only has 4 ports to its 7 execution units. Qualcomm wouldn't give me specifics on what those 7 units were or how they were shared by those 4 ports. It's a shame that Intel will tell me just how big Haswell's integer and FP register files are 9 months before launch, but its competitors in the mobile SoC space are worried about sharing high level details of architectures that have been shipping for half a year.
Apple's Swift core is a wider machine than the Cortex A9, and seemingly on-par with Qualcomm's Krait. How does ARM's Cortex A15 compare? While the front end remans 3-wide, ARM claims a doubling of fetch bandwidth compared to Cortex A9. The A15 is also able to execute more types of instructions out of order, although admittedly we don't know Swift's capabilities in this regard. There's also a loop cache at the front end, something that both AMD and Intel have in their modern architectures (again, it's unclear whether or not Swift features something similar). ARM moves to three dedicated issue pools feeding 8 independent pipelines on the execution side. There are dedicated load and store pipelines, two integer ALU pipes, two FP/NEON pipes, one pipe for branches and one for all multiplies/divides. The Cortex A15 is simply a beast, and it should be more power hungry as a result. It remains to be seen how the first Cortex A15 based smartphone SoCs will compare to Swift/Krait in terms of power. ARM's big.LITTLE configuration was clearly designed to help mitigate the issues that the Cortex A15 architecture could pose from a power consumption standpoint. I suspect we haven't seen the end of NVIDIA's companion core either.
At a high level, it would appear that ARM's Cortex A15 is still a bigger machine than Swift. Swift instead feels like Apple's answer to Krait. The release cadence Apple is on right now almost guarantees that it will be a CPU generation behind in the first half of next year if everyone moves to Cortex A15 based designs.
Facebook
Twitter
RSS
276 Comments
View All Comments
youwonder - Wednesday, October 17, 2012 - link
I find it kind of ...odd that the S3 has a much larger battery than the one X and the same SoC yet posts significantly worse LTE browsing numbers, and is the only phone using LTE to get worse results with it than using 3G(granted that is the international vers, doesn't look like they had time to do testing on the AT&T or verizon variant running 3G). Does the samoled screen really draw THAT much more power than an LCD? also there's this which makes me wonder more:http://blogs.which.co.uk/technology/smartphones/be...
Of course, I don't respect these guys as much as anandtech when it comes to accurate results, and they did things much differently (broadcasting their own 3g signal and putting all phones on max brightness), but still the odd results here make me wonder if a small mistake wasn't made.
Zink - Wednesday, October 17, 2012 - link
Max brightness gives the gs3 an advantage because its screen is so dim. The other phones are using LED lighting as well but they go much brighter and have to shine through the LCD panel.youwonder - Wednesday, October 17, 2012 - link
Good point, I guess it's mostly just me wondering why the GS3 LTE variant posts such horrible numbers even compared to it's 3G version when anand specs a good amount of time explaining why the opposite is true.phillyry - Sunday, October 21, 2012 - link
Don't know why but it does tank on LTE.rarson - Wednesday, October 17, 2012 - link
I'm getting so sick and tired of seeing the word "literally" injected into all sorts of sentences that it doesn't belong in. This word only needs to be used when describing something literal. It's not a synonym for "really" (not yet, anyway).andykins - Wednesday, October 17, 2012 - link
Alright, language purist. :Pjoos2000 - Thursday, October 18, 2012 - link
http://theoatmeal.com/comics/literallyphillyry - Sunday, October 21, 2012 - link
Great link. That's too funny - literally!dfonseca - Wednesday, October 17, 2012 - link
On the last page, section "Final Words" / "iPhone 5 Device Conclusions", it's written:> At a high level, the iPhone 5’s cameras appeared to be some of the least unchanged elements of the new device however in practice the improvements are significant.
"Least unchanged" means "most changed." It should probably say "most unchanged," or "least changed."
Nice review, kudos to all authors.
mattlach - Wednesday, October 17, 2012 - link
I had the original iPhone, followed by the iPhone 3G and then the iPhone 4, and just switched to a Samsung Galaxy S3 in July.When the original iPhone came out, while it was the first to do what it did - and that's why I bought it at its steep no-contract introductory price - it wasn't exactly revolutionary, everything in the market was moving in this direction, but it was pretty well executed and nothing else did it at the time.
I upgraded to the 3G on launch, as I thought the edge speeds were dreadful, but was disappointed, as the phone wasn't fast enough to take advantage of 3G, and AT&T's 3G was pretty mediocre anyway. It didn't get important features its competitors had, like copy and paste until very late in the game, and I started to think that I should have gotten an Android phone instead.
By the time I got the iPhone 4, I was tired of my slow 3G experience and just wanted an upgrade to something faster. The iPhone 4 was a good upgrade, but I really only got it because I didn't like AT&T's Android offerings at the time. I had been thinking about going to Verizon and getting an Android for some time. The 3G should have been my last iPhone, it was a mistake to buy the 4.
Having realized my mistake, I waited 2 long years with the 4 until I could finally get out of my AT&T contract and go to Verizon and get a GS3, and it felt great.
The additional freedom of what I run on my phone, not being controlled by Apple and their agenda as to what makes it into the App store, and the fact that I finally no longer had to have iTunes installed on my computer were fantastic.
My computer has been iTunes free for 3 months now, and it feels great!
I was concerned for a while that once the iPhone 5 was released, they would come out with something that would make me regret my choice of the GS3, but it turns out they didn't.
I'll likely never buy anything Apple again. It feels like a huge relief to say that.