Wow, I hadn't realized Apple had bought into the 'Sun' model that to be the best you had to have your own CPU. Well I knew they were doing their own silicon, I think that is actually kind of nominal for folks who want to innovate in the small device space since you can't get to the components otherwise, but to change the CPU architecture in the process is a pretty bold step. Sun discovered that you can get a jump on the competition like that, the first SparcStations really blew away the competition, but ultimately you can't outspend even three other competitors at the same time.
Of course this was also why Samsung has been a much bigger threat to Apple than the other Android handset makers, they too make their own SoC silicon.
So we get to see three companies pursuing products in the same space with three different strategies:
Apple - Custom OS (iOS), Customized Silicon
Samsung - Third party OS (Android), Customized Silicon
Asus/HTC - Third party OS (Android), Third party silicon (nVidia)
These kinds of situations don't come around that often so when they do they make for great learning experiences.
Apple acquired PA Semi, which wasn't just an ASIC group, but arguably the best group of processor designers for high-performance, low-power computing. The group previously built StrongArm (a famously high MIPS-per-watt processor in the 90s). It's a much more specialized skillset than normal ASIC dev.
The reason it should work out better for Apple than it did for Sun is
(1) Apple sticks with ARM architecture so they don't have to custom-build the whole toolchain (eg compilers)
(2) Apple ships in insanely high volume so the cost of chip development is pretty low amortized over their volume
(3) Performance-per-watt is really important in mobile, and they're doing it with an amazingly accomplished team in that area.
I wish people would stop obsessing about PA. That acquisition was a fiasco. All the talent bailed after a tiny lock in period, went to a "startup", and were promptly "acquired" by the googs. The whole thing was a mess that should have ended with a vp or director "pursuing other opportunities".
The Intrinsity talent buy was way better. Those guys actually put out product and have a functional team.
I have over 100 people in my LinkedIn network who were ever at PA Semi, and 53% of them are still at Apple. Where's the evidence that "all the talent bailed" ? Some of the key architects and leaders moved on, but I think that's not unusual.
Yup. Apple has been putting together a CPU team for a long long time now, including high-perf simulators, libraries, etc. etc.
However, the point about this catching up to them and biting them is a real possibility. Now, they have a lot of discipline, it seems, and that is critical. The other part is having a very tightly closed platform with control of the entire toolchain.
I'm not sure that Sun is the best comparison here, as they did both the ISA and the microarchitecture on their chips. If the claims in this article are to be believed, Apple is doing their own uarch, but keeping the armv7 ISA (and gluing some other ARM building blocks together).
When Sun's architecture started to fall behind, there weren't really second sources for compatible chips. Combine that with a greater reliance on ISA optimized assembly than today, and you can see why their customers started to leave.
If Apple did roll their own ARM this time around, they almost certainly built prototypes of the iPhone 5 with a "standard" ARM9 design for contingency reasons. Getting the software to work on either hardware should have only been a matter of setting some optimization flags and swapping out a few header files. Easily worth the risk, in my opinion, as they can always revert back to a uarch designed by ARM if their own designs start to lack performance or become too expensive in the future.
I agree it certainly isn't a 'clean' comparison, but at the time Sun didn't have the option of licensing the 68K architecture from Motorola and extending it in their own silicon. I still remember a talk with both Bill Joy and Andy Bechtolsteim where Bill was over the top enthusiastic about finally having the ability to change the CPU to work better with the software rather than having the software work around what the CPU thought was best. I don't doubt for a minute that he at least would have been just as enthusiastic if the 68K could be augmented in a Sun specified way. There is a huge efficiency in customizing the CPU to do what you want.
This I found interesting:
"If Apple did roll their own ARM this time around, they almost certainly built prototypes of the iPhone 5 with a 'standard' ARM9 design for contingency reasons."
How likely do you suppose it is that they did exactly that and called it the '4S' ? Having worked in places doing new CPUs (Intel in the '80's, Sun in the '90s) you always make a product based on the old stuff in the event the new stuff doesn't hit the schedule. The Sun-3/60 was exactly that machine, it was the 'in case the SparcStation 1 slips' machine.
It seems very unlikely that the A5 used in the 4s was a fallback plan.
First of all, they were already getting a huge speed bump by going from single core A8 to dual-core A9. So existing iPhone users would have been happy. There was also no foreseeable external competitive pressure at the time, everyone knew that there weren't going to be any S4 phones for 6-9 months, and that Tegra 3 would be a joke.
They were also still on a 45nm process, and the A5 with dual-core A9s was already huge (120 mm^2). They would not have wanted to increase that at all, even if they had a working design of their own. What would have been the options? Seems like it would have been either reducing GPU resources, or dropping a core. Hard to see either of those being worth it.
Finally, one could argue that they main plan was their own core@32nm, and the A9@45nm was the fallback if there were process troubles. But these days there seems to be a strong preference to alternate new designs with new process nodes, rather than doing both at the same time. And in fact this conservative option is exactly the one Apple picked next spring: the iPad refreshes were a mix of 32nm parts (simple shrink) and 45 ones (higher clock, more GPU). If Apple weren't willing to risk going to 32nm even for a design iteration as conservative as A5 -> A5x (and the process was clearly working for them with the A5 shrink), the idea that they would have tried for a radical new design + new node 6 months earlier doesn't seem credible.
Comparatively weak single-threaded performance, weak GPU performance, constrained memory interface, high power consumption when not running only the companion core.
Tegra 3 isn't terrible, but just like previous Tegra chips it was so hyped up before launch and again turned out to be among the weaker ARM chips of its generation. Exynos, Snapdragon S4, Apple A5 (and now A6) and OMAP 4 beat Tegra 3 on everything except synthetic multi-threaded benchmarks (because it has 4 cores, obviously).
I don't know about the GPU, my impression was that it was competitive (the single-channel memory interface probably doesn't help), but otherwise that's spot on. It really seemed to be a design whose main goal was to have the most cores for marketing purposes.
And of course they launched on a process that was just about to become obsolete, and haven't been able to do a shrink. Not sure why, though. Complications with the companion core? Just NVidia's general inability to get 28nm to work?
(Commercially it didn't help that they were unable to get access to a competitive LTE modem. That meant there was no shot at the US phone market. But that's not particularly a failing of NVidia, everyone but Qualcomm has that problem.)
Man, I would think that Chuck M. would know, but this time, apparently not. The Sun 3/80 shipped simultaneous with the SS1, the 3/60 pre-dated the SS1 by several years.
The degree of compatibility seems to be a major variable, so this puts the comparison into doubt.
> I still remember a talk with both Bill Joy and Andy Bechtolsteim where Bill was over the top enthusiastic about finally having the ability to change the CPU to work better with the software rather than having the software work around what the CPU thought was best.
They should have been able to make Java VM implementers' lives a lot easier on Sun platforms. I guess the question was also, did they necessarily want to?
That is true but in the early years Sun was not a 'good' licensor since they seems to compete with anyone who tried to build one. Fujitsu was a success story and arguably made better high end SPARC chips than Sun did. ARM Holdings on the other hand doesn't make any hardware at all so they are completely motivated to have their licensees be successful.
If Apple is really doing this, Intel can forget about ever putting Atom or any other of their chips inside the iPad. I know they've been hopelessly optimistic about it, but this is a clear sign it's just never going to happen.
Also, I still believe Apple will start making Macbook Air-like devices, maybe starting in 2014 with the Apple A8 (?) ARM-based chip, when ARM will have moved to the 64 bit ARMv8 architecture.
Well, you never know. If Intel continues to improve their mobile device x86 designs, they might surpass Apple and ARM's abilities some day - they do have a lot of architecture expertise and cutting-edge fab technology.
I still think it would be interesting to see Intel (or AMD) come out with a 64-bit only design for mobile / low-powered servers. I wonder how much space the 32-bit and older SIMD stuff takes on the chip?
Apple has been in the same position as Samsung for a while; both companies take other peoples' CPU and GPU core designs and integrate them into an SoC (A5 and Exynos, respectively). This would put them more in the same field as Qualcomm, which designs its own ARM cores using an ARM architecture license. It looks like this is what Apple is now doing too.
Kind of? There's been in house vlsi & soc at apple forever. Theyve had in house mobile CPU & graphics talent for a couple years at this point. Go check out the stuff Intrinsity was doing as a cash starved startup. Then extrapolate that same team being a well funded semi autonomous design group inside apple.
The Sun model ? Before the rise of the x86 based servers there pretty much evey major server vendor was doing this. IBM power - AIX, HP PA-RISC-AIX, Sun - you know already, Digital Alpha-Digital UNIX. Practically everybody had(and in the case of IBM still do) their own hardware including custom chips. The CPU world was way more diverse in the 90s and early 2000s. Sun was hardly the first or the only one.
Is Samsung making their own cores? The news in the article is that Apple is making their own cores now, they've been making their own silicon for the last few revisions, right?
It isn't clear to me that Samsung has publicly enhanced the ARM architecture in the Exynos [1] series but they certainly could and exploit that in their own products. Like Apple they are full licensees to the ARM architecture (which means they can make derivative works of the ARM cpu) and of course if they did they would be able to exploit that in their version of Android if they so chose to. There were a few companies that did that with the 6502 'core' back in the day adding some custom instructions in the space that was reserved or undefined by MOS Semiconductor.
It's really hard to build a custom processor with decent performance these days. There are very few teams that can do it, and it takes way more investment than normal ASIC design. The people who can do it largely work for Intel, AMD, IBM, some at Apple, and probably a few other major shops.
So I suspect it would be hard for Samsung to build a custom ARM processor unless they teamed up or bought a team that already has this specialized skill. That's what Apple did...
Samsung is making them, but they don't design their own.. well, they design the chips but not the arm cores themselves-- AFAIK only qualcomm (and now apple) have done this in the smartphone space.
It is interesting to note that Apple now designes their own programming language, the compilers, the operating system, the processors and the products all that tech goes into. That's a pretty rare configuration of a company that hasen't existed since what? the 80s? [1]. That's a really unique opportunity to do really wild things, I hope they don't squander it. But I guess that requires a bold visionary, not of the Jobs type, more of an academic.
[1] I guess IBM still counts, but mainframes aren't that sexy anymore :)
It wasn't too long ago that Sun SPARC systems were a thing. Still I agree, it isn't that often someone can scale vertically like this. Hopefully they do something interesting with it.
I'd love to see them use what they learn in the mobile space to push the boundaries in their workstation market and experiment with something outside x86. Since they've already been through the PPC->x86 jump it would be as painful of a transition both on the technical and the sales sides.
I think you would find that in moving the apple workstations away from x86 would cause them a lot of pain as they would lose all those that use the apple hardware to run both OSX and Windows.
I would assume that Linux would be available for any future architecture but that can by no means guaranteed.
You might be able to group Google in that category too, depending on how far you take it.
In a way, I think it's exciting. It should allow them to do some cool things and Apple definitely has the resources to do such. Although, when I put my tin-foil hat on, I find it a little worrisome when one company controls everything. But ultimately I keep going back to my capitalism hat (don't ask me what it looks like) and I find it to be the best for us consumers for the competition it creates. :)
>It is interesting to note that Apple now designes their own programming language, the compilers, the operating system, the processors and the products all that tech goes into. That's a pretty rare configuration of a company that hasen't existed since what? the 80s?
Not to mention a company that doesn't "innovate" and hasn't "invented anything" as haters say...
One thing that makes Apple in a very special situation is that they have $100 billion cash to fund various forms of competitive advantage. From the supply chain to the hardware design to the software to the retail delivery experience, Apple can invest an optimize every point in the whole process. It will be fascinating to see how this plays out.
If this is true, it's strange that Apple didn't tout it more. I guess "And this Apple designed chip we told you is Apple designed already the last few times is even more Apple designed now!" is a hard message to sell, and from a consumer's perspective, those implementation details are completely irrelevant anyway.
I guess Apple will tout completely irrelevant tech specs when it sounds good and fits the message, but when it would just cloud the issue and be confusing they just leave it out, even if those tech specs would put them in a positive light.
Apple have touted it. Twice as fast as 4S, longer battery life. That is all that matters to the consumer and those are the benefits brought by the CPU.
That's the boring part. Sure, that's what's actually matters (or at least a good deal closer to what actually matters), but Apple hasn't been shy to share some technical details at pretty weird opportunities.
I think part of the problem is they already have introduced the cpu as "theirs" in a previous iteration (I don't remember which one) when they introduced for the first time the "Apple Ax". So I think it would now be difficult to explain that the cpu was not really completely made by them, that now this it is their design (built on the ARM IP) and put in silicon by Samsung... Too much hassle to explain I think for too few marketing gain.
They could have said "completely redesigned* - the closest they came to that was calling it "A6". They are right to sell the benefits, because that's what sells. Even Woz chastised them for saying "quad-core" graphics for the iPad 3. This time, they didn't even say that - just "2x faster graphics".
Maybe 5% of smartphone consumers would actually care about the nuts and bolts of it. I think Apple did a great job by showing it how it affects the user: battery life, speed for tasks, and demoing Real Racing 3.
I dunno. I totally geek out every time Ars does a CPU core review, but still I find the claim that the A6 is 2X faster in so many real-world situations to be more exciting than any geeky technical number they could claim.
I think it's weird that they'll gloss over CPU details, opting for "2x as fast", but then when the camera talk comes around, they go on and on about optics, filters, image processing, etc. I'm not sure why they go into so much detail on the camera details.
In practice, modern phone SoCs are fast enough for most purposes. It's not like the 4S (or even the half-again-as-fast 4) are slow. A great improvement won't really make too much difference to most users. Smartphone cameras, on the other hand, are still somewhat lacking and an improvement will benefit most users substantially.
It's to counter the needless and destructive megapixel race in phones. Until they get their claws on something truly gamechanging like Pureview they need to train their customers to talk about camera quality, not megapixellage.
That said, at the manufacturing scales of these parts leakage in the transistor can be as big a power consumer as the switching.
Leakage current increasing as voltage increases at a rate that is of the order ~e^Vth
I wouldn't really say its strictly either exponential or quadratic, you'd have to know a lot more about the process and the implementation to know how significant leakage current is in the design.
Dynamic power is quadratic with voltage and is work-load dependent, i.e P = c * AC capacitance * Freq * V^2. Static (i.e. leakage) power has an exponential component (P = AVe^(k*V)), although k is typically pretty small. I'm not sure about the static vs. dynamic power breakdown of Apple's designs.
This is just a bunch of rumors and the author patting himself on the back for being well connected and talking to the right people. Well it may be true, but this article does not really prove it. This is one of the things that, if true, would come from official apple channels rather than un-sourced rumors.
> This is one of the things that, if true, would come from official apple channels
I don't think Apple have ever commented on their chip architecture, at least at this sort of level. It's going to be verified by somebody throwing the chip under an electron microscope, rather than Apple releasing documentation.
In all fairness, on this particular issue his track record isn't too good :-) Within the last a week he has been strongly of the opinion that it's going to be an upclocked A9, then that it's an A15 mysteriously launching months ahead of schedule, and now that it's a previously unheard of Apple design.
At this rate we're going to hear in two days that actually Apple licensed Krait from Qualcomm, and in a week that what really happened was a switch to Medfield.
And yet he was dead wrong about the new iPhone having Cortex A15 in the other post, that made everyone on the Internet think that Apple uses Cortex A15. So much for his reputation. I wonder what will remain of it if this proves wrong, too.
He speculated on the options Apple had for the iPhone 5 SoC, and speculated wrong. No big deal, it happens to every tech website or blog that tries to predict future products. Now he has some tangible evidence about the nature of the A6 core (the fact that XCode now has build profiles for ARMv7s + VFPv4) and supposedly got confirmation from someone in the know that the A6 is a custom design. Seems like a safe bet this is right to me.
It almost seem you want Anand to be wrong, you have some personal beef with him?
Just want to add that Anandtech has really stepped it up this year with great iPhone coverage. It's a nice contrast to countless other speculation/rumor articles with no valuable insight and "controversy = page hits!" articles that just fan the flames of the "Apple vs. Everyone Else" arguments that saturate tech blogs.
I totally agree. AnandTech is one of the very few remaining tech websites that are still really about the technology itself, instead of all the other aspects that ignite flamewars and hating everywhere else.
The Verge isn't too bad either if you ignore the comments section, but everything else is just like you said, speculation/rumor sites that value page hits over actually writing something informative.
This is an important development, as it marks Apple's first use of an internally-designed ARM core in any of its products. Apple's acquisition of processor design expertise is going to start paying off at a much more fundamental level than it has in the past.
I'm excited to see where this leads, as Apple possesses for the first time the ability to design the SoC for a device like the iPhone from the cores up.
Right, Apple bought P.A. Semi, and assuming that those guys stuck around after Apple switched from PowerPC to Intel, Apple has a proven CPU devel team.
Apple bought P.A. Semi in 2008, a bit over two years after the Intel transition. Apple also bought Intrisity (in 2010) which was made up of staff and IP from PowerPC licensee, Exponential Technology.
Apple have more history and clout with ARM than Samsung so they don't have to wait on Sansung for anything, they use them to fabricate the chips, they use others as well. Also with the volume they are using and the stabalising of life of some production setup's it wont be long until Apple build there own fabrication plant. Though I'm sure they will wait for a nice TAX break or chance to pick up somebody elses cheap. Though it is more than viable for them.
Apple may also be worried about clones and in that by having a more than customed CPU than most will alow them to lock there iOS more and more to there kit only. Whilst not an issue now, it may down the line if they get hit with a monoply commision legal run-in and in this way they would still make a larger cut than just liscencing out iOS.
The trend in every mobile generation has been add features, screen size/colour depth and then back to battery life. It will be after all the next major selling point amongst a collection of phones with high end features.
As for the A15, its more targeted at a level that is too much for what smartphones can handle batterywise and feature wise. That said they may of used parts of the A15 with part of the A9 core making a hybrid, not everybody has to use the cookie-cutter templates ARM has to offer and Apple are free to customise or basterdise them however they like.
"Samsung would not give it to Apple before they used it themselves."
two mistakes in that phrase:
- A15 isn't Samsung's to give to Apple -- they license that from ARM. Yes, Samsung manufactures Apple's chips, but the chip is designed by Apple using IP licensed from ARM, Imagination Technologies and others, but as far as Samsung is concerned the chip is a black box.
- Samsung has given technology to Apple before they used it themselves several times in the past.
I also doubt the first part of your statement. It's early for an A15, but it's certainly within Apple's power to persuade ARM to help them get something they want to market early. Spending their billions to bend supplier's arms is one of the things that Apple does best.
Samsung being a conglomerate, isn't the chip division of Samsung operating independently of it's phone/electronics division? If so then I see no reason why Samsung Fab would deny a certain type of chip design to a customer like Apple.
1. Samsung are poised to be the first manufacturer to bring an A15 based SOC to market. There are no doubt many issues involved in the reality of manufacturing a new arm core, no one else is as close as Samsung. So i maintain that Apple would have needed Samsung to enable them to use an A15.
2. Yes, they have supplied apple with stuff that only they can make before using it themselves - eg. the iPad screen (but IIRC LG were meant to be supplying too), but i still don't think they would want to give apple a CPU platform that was undoubtedly better than their flagship phone so soon after releasing it.
3. ARM aren't really a supplier to apple, they just licence designs. ARM have no manufacturing expertise.
> "i still don't think they would want to give apple a CPU platform that was undoubtedly better than their flagship phone so soon after releasing it"
You need to bear in mind that the division of Samsung that sells the chips is totally separate from the division that sells the phones. The silicon division doesn't particularly care about the phone division - they care about their bottom line alone. And Apple turning up with a dump truck of cash, which is basically what they've been doing to secure component supplies, is not something they wil turn down lightly.
Apple is somewhat unique among consumer electronics companies in that it is very singly focused. Samsung, along with Sony, Microsoft, and others, are multi-conglomerates, with vast reaching business interests, and Samsung's chip fab division only really cares about the profits their division is making. Money from Apple, money from an internal cross-charge: profit is profit.
> "3. ARM aren't really a supplier to apple, they just licence designs. ARM have no manufacturing expertise."
I don't think this is what you mean. You mean ARM have no expertise manufacturing at scale. ARM have extensive manufacturing expertise, because they need to supply reference designs that work in production. ARM has manufacturing expertise because they need to produce designs that are easy to manufacture.
I would say ARM have expertise in 'designing for manufacture' which is distinct from the actual manufacturing, but lets say they have no expertise in 'production manufacturing'.
While Samsung does indeed contain a bunch of largely separate companies, both chips and phones are part of Samsung Electronics, there is one set of top management for both.
They have most likely done many calculations that clearly say "if we lose Apple on the supply side, we lose <insert ridiculous amount of money that eclipses smartphone revenue here>."
> So i maintain that Apple would have needed Samsung to enable them to use an A15.
I would guess if Apple is capable of designing their own Microarchitectures they would be perfectly capable of compiling a bunch of HDL code they can get straight from ARM. [1]
> but i still don't think they would want to give apple a CPU platform that was undoubtedly better than their flagship phone so soon after releasing it
Samsung also acts as a foundry, they manufacture designs of other people in the same sense a printing shop prints books written by other people. I don't think this involves Samsung "giving Apple a CPU platform", at least not any more.
[1] Yes, in reality thats quite a bit more complicated.
It's not necessary to support armv6 anymore. The only iPhones with armv6 were the original and the 3G, which was released more than four years ago. The 3GS and everything since has been armv7. Every iPad has been armv7.
The last version of iOS to support armv6 was 4.2.1, released almost two years ago.
The latest SDK only supports iOS 5 and up (if you want to target 4 you have to install an old one), and there are no ARMv6 devices which can run iOS 5, so it makes sense.
Just FYI: New Xcodes install into /Applications/Xcode.app (and things like instruments are now in /Applications/Xcode.app/bin/Developer/Instruments.app or something like that) and the old Xcode is in /Developer.
Of course this was also why Samsung has been a much bigger threat to Apple than the other Android handset makers, they too make their own SoC silicon.
So we get to see three companies pursuing products in the same space with three different strategies:
Apple - Custom OS (iOS), Customized Silicon
Samsung - Third party OS (Android), Customized Silicon
Asus/HTC - Third party OS (Android), Third party silicon (nVidia)
These kinds of situations don't come around that often so when they do they make for great learning experiences.