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.

Custom Code to Understand a Custom Core Apple's Swift: Pipeline Depth & Memory Latency
Comments Locked

276 Comments

View All Comments

  • avner - Tuesday, October 16, 2012 - link

    Are you sure the HTC One S has a removable battery? Last time I checked, it was internally sealed because of the phone's unibody design....
  • tipoo - Tuesday, October 16, 2012 - link

    I'll be curious to know if the paint holds up well after a few months, but I have no idea how long you get to keep this for testing. The easily scratchable back is already rustling some jimmies on the internet, but I don't know if that's something that's overstated again or if it is a real problem.

    The softness of aluminum they can't get around, I know that much (the front of the iPod Classic and any Macbook Pro are quite dent prone), but the paint chipping is a separate issue.
  • dagamer34 - Tuesday, October 16, 2012 - link

    Not paint, anodization. You should watch the video on page 1.
  • Alucard291 - Tuesday, October 16, 2012 - link

    Yeah and in this instance its holding even worse than paint. So your point is?
  • doobydoo - Friday, October 19, 2012 - link

    Depends how thick the paint is, and what the paint consists of. And his point was to improve the accuracy of the original post, not necessarily to disagree with its conclusion.
  • tipoo - Saturday, October 20, 2012 - link

    Does a name make it more durable? :P

    I'd still like a follow up in a few months if that is possible.
  • extide - Thursday, December 6, 2012 - link

    It ISN'T paint. That's the point. Anodized aluminum is actually quite a bit harder than regular aluminum so the anodization process makes it more durable. All they really need to do is make the anodization layer a bit thicker.
  • KPOM - Tuesday, October 16, 2012 - link

    I was wondering what was taking so long. Now I know. Thanks for all the great information.
  • dyc4ha - Tuesday, October 16, 2012 - link

    Hi,

    I can't help but wonder why the price aspects are omitted in the review. As per the famous AT saying: "there are no bad products, only bad prices". So what about the iPhone 5 value proposition for $650+tax? Contract pricing? i.e. is it worth signing on for a contract?

    I'm asking because, well it is a widely accepted 'fact' that Apple overprice its products, but I would love to hear perhaps justifications from reputable sources such as yourselves. Most of my friends already jumped the ship from 4/4s to a GS3. I paid for a number of apps over the years, not sure if I want to repurchase all of them if I end up switching.

    Cheers
  • darwinosx - Tuesday, October 16, 2012 - link

    iPhone 5 starts at $599 not $650. Go check retail on a Galaxy S3.
    it is not widely accepted that Apple "overprices" Compare iPad and iPhone prices to current Android. Prices are same and sometimes cheaper.
    Compare Mac prices to equivalent build level PC's.
    People sho say Apple overprices don't know Apples or their competitors prices. They just want to whine about Apple.

Log in

Don't have an account? Sign up now