This post originally appeared at MattRichman.net and was re-posted at Tech.pinions with Matt’s permission.
When Steve Jobs announced the transition from PowerPC to Intel processors in 2005, he revealed something that, in hindsight, seemed obvious to everyone who didn’t anticipate the switch:
There are two major challenges in this transition. The first one is making Mac OS X sing on Intel processors. Now, I have something to tell you today: Mac OS X has been leading a secret double life for the past five years.
[…]
We’ve had teams doing the “just in case” scenario. And our rules have been that our designs for OS X must be processor independent, and that every project must be built for both the PowerPC and Intel processors. And so today, for the first time, I can confirm the rumors that every release of Mac OS X has been compiled for both PowerPC and Intel. This has been going on for the last five years.
There’s not a doubt in my mind that if you substitute Intel for PowerPC and ARM for Intel, what Steve Jobs said then holds 100 percent true today, word for word. Mac OS X designs must be processor independent, every project must be built for both Intel and ARM processors, and each Mac OS X release in the last five years has been compiled for both Intel and ARM.
Somewhere on Apple’s campus, ARM-based Macs are already running OS X.
User Experience Would Improve
In his iPhone 5S review, Anand Shimpi compared the Apple-designed A7 processor with Intel’s fastest tablet chip at the time. He wrote:
For our cross-platform CPU performance tests we turn to the usual collection of Javascript and HTML5 based browser tests. Most of our comparison targets here are smartphones with two exceptions: Intel’s Bay Trail FFRD and Qualcomm’s MSM8974 Snapdragon 800 MDP/T. Both of those platforms are test tablets, leveraging higher TDP silicon in a tablet form factor. The gap between the TDP of Apple’s A7 and those two SoCs isn’t huge, but there is a gap. I only include those platforms as a reference point. As you’re about to see, the work that Apple has put into the A7 makes the iPhone 5S performance competitive with both. In many cases the A7 delivers better performance than one or both of them.
In September of 2013, the world’s preeminent independent processor expert compared Apple’s latest iPhone chip with Intel’s fastest tablet chip and concluded that the two perform similarly — even though the Intel chip draws more power, contains four cores versus the A7’s two, and is produced with a more advanced manufacturing technique. If Apple’s chip design team can create a phone processor that performs on par with Intel’s fastest tablet chip, the company’s “highest priority”, then there’s no reason to believe that the same team at Apple can’t design chips powerful enough for any Mac in the company’s lineup.
Apple has already released a line of A-series chips tailored specifically for iOS devices, and the company is most definitely working on a line of B-series chips tailored specifically for Macs. When that B-series chip — or set of B-series chips that runs in parallel — is ready, Apple will be able to switch to ARM-based Macs without sacrificing user experience. On the contrary, because the company is no doubt designing its line of B-series chips in tandem with Mac OS X, there would be iPhone-like hardware-software optimization, improving user experience.
Apple Would Make More Money Per Mac And Sell More Macs
Going from chip concept to manufactured product can be broken down into two separate and distinct steps. The first is chip design — figuring out what features the processor will have and how it will work. The second is manufacturing — turning a file that exists on a screen into a physical product you can hold in your hand.
Today, Intel designs the chips in Macs and manufactures them, profiting on both of those steps. But if Apple swapped out Intel’s chips for its own ARM-based designs, an external company would profit on only one step of the chip creation process, not both, leading to a decrease in the cost of building a Mac. By my conservative estimate, Apple would be able to drop the price of the base model 11- and 13″ MacBook Airs by $50 and still make more profit per unit on each than it currently does.
This cost savings would apply to the entire Mac lineup. Apple would be able to drop prices across the board and make more money per Mac than it does today — and with lower prices, the company would sell more of them, too.
Apple Would Be Able To Create Better Macs
When Apple announced the iPhone 5S, it explained that all of the fingerprint data associated with Touch ID “is encrypted and stored inside the secure enclave in our new A7 chip” where it’s “locked away from everything else”.
Apple wouldn’t have been able to create Touch ID if the iPhone were powered by an Intel chip instead of an Apple-designed one. There wouldn’t have been a “secure enclave” on the iPhone’s processor to store the fingerprint data, nor would there have been perfect hardware-software integration. Apple was able to implement Touch ID because it designed the A7 chip in tandem with the iPhone 5S’s software and the rest of its hardware.
I’d bet that there are features Apple envisions for the Mac that simply can’t be built while Intel designs the chips inside of them. To implement those ideas, Apple would need to switch the Mac to ARM-based processors, because only then would the company have the ability to design chips customized for specific features. If Apple moved the Mac to ARM-based chips, the company would literally be able to create better products than it can today.
This brings me to something else Steve Jobs said when he announced the transition from PowerPC to Intel. Ultimately, he explained, Apple switched for one simple reason: “We can envision some amazing products we want to build for you, and we don’t know how to build them with the future PowerPC roadmap.”
The same logic applies today. It’s not a stretch to imagine Tim Cook walking out on stage and saying, “We can envision some amazing products we want to build for you, and we don’t know how to build them with Intel’s chips.”
As I first said more than three years ago: ARM-based Macs are definitely coming.
For the most part I agree with you. There are a few things I am not clear about.
1. Intel gives you support chips that provide eSATA, ethernet networking, USB 3 and Thunderbolt. The A series processors have never had to provide those before. While eSATA, ethernet and USB 3 support should be pose no problems to implement in Mac A series variant, Thunderbolt may be a different issue. Thunderbolt is mostly an Intel invention that Apple has participation in. Apple may not have a license to implement Thunderbolt on a non-Intel processor.
2. Would Apple have to give up the ability to run Windows on a Mac? Apple I am sure has a better idea than I how many potential sales could be lost here.
3. Who would build the Mac variant processor? TSMC? Samsung? Intel???
Hi Gary,
Good news on Apple/Thunderbolt:
Apple created the original IP w/ patents on TB and shared these inventions with Intel. So, yeah, Apple is completely licensed and royalty free here.
“3. Who would build the Mac variant processor? TSMC? Samsung? Intel???”
The same manufacturers that are building the processors for the iPad and iPhone.
I think that Apple having to use Samsung as a CPU manufacturer who violates their patents is at least an uncomfortable situation. I read reports that TSMC did not want take all of the buisness Apple could give them because they did not a single customer that vital to their bottom line.
Intel is the really interesting one. They still want the mobile world to switch to Atom which is not going to happen. If Intel ever got past their ego there are billions of dollars that they could wrest from Samsung, that is Apple’s business. With the PC business declining, it seems to me that becoming Apple’s CPU foundry is there for the taking as long as you can execute. It would also be difficult for Intel to accept Apple ditching x86 for A series and then be asked to make the things.
Apple having to outsource CPU manufacturing limits their options. Apple cannot simply get into the silicon foundry business. Talk about specialized knowledge and huge capital expenditure, wow!
Intel already makes ARM chips for third parties so doing it for Apple would be down to opportunity cost. If Apple says “We will sub you for a founders and give you the orders to fill it”, then that’s a done deal.
Apple is an ARM licensee, but designs their own versions.
It really bothers me that Apple has to get their stuffed fabricated by Samsung and Intel watching these Billion$$$ go to Korea when they could pick them up.
Intel is acting very much like Microsoft, selling their customers products that serve Intel’s agenda first and hopefully the customers will go along with that.
I argued that Intel should fab Apple’s A-series chips back in April, which is what led to my internship at Intel:
http://mattrichman.net/post/82689013158/apple-samsung-and-intel
It’s great that it got you an internship at Intel! However, you did not seem to persuade them.
I’ll bet that was a great experience though.
Ha!
I tried to change the internal estimate of when Apple would be technologically capable of using ARM-based designs in Macs, but I failed at that, too. There’s only so much an intern can do 🙂
Makes me remember my first engineering internship.
Apple is undoubtedly running something in the labs. Question is whether the benefits of ARM Macs are such that they are worth the trouble (e.g. loss of Windows compatibility and annual refresh burden). In other words, there would have to be some very significant jump in what an ARM Mac can do.
What is that amazing feature that makes it all worthwhile? Saving $50 per Mac is nice, but not nearly enough. Double the battery life, possibly. Much better GPU, possible. Better instruction set, who knows. Secure TouchID enclave, could be part. I think they would need a compelling set of benefits.
The one thing that could accelerate all of this is if Apple were to lose confidence in Intel’s roadmap. In that respect, Intel has not been doing themselves any favours.
The biggest ARM gives Apple is customization. It allows Apple to build in custom features into their in-house designed processors that it can’t do with buying off-the-shelf x86 processors (ie: TouchID). These are features that help give Apple’s product a competitive advantage. It isn’t just about performance, although that is important.
One of the things that would be essential for this is a new version of Rosetta, the PPC->x86 translation capability. Many apps could be built as native ARM but that would take time. What’s more, apps that aren’t rebuilt mean that Apple loses some control over what happens. As we well know, they hate that.
Problem is – Apple doesn’t own that technology. Maybe they’ve been building it it – who knows? But the PPC->x86 one was created by a company called “Transitive”. And it seems that the reason Rosetta is no longer part of OS X os that Transitive was acquired, and the new owners didn’t want to license it.
But wait – the new owner is IBM. And didn’t Apple just do a deal with IBM for iPads and enterprise and the like? Maybe there’s something more too??
I believe that it is in Intel’s strategic interest to tell Apple that they CAN’T HAVE anything built on 14nm 🙂
Within 3 years Apple would be back in the same position as they were with PowerPC in 2006, lacking performance and with a frustrated user base.
And if Apple want a Mac chip with A7 or A8 like performance they can save themselves $50 using “Bay Trail” or its successor for about $15 each if they negotiate nicely. Why would they put their users through another round of “compatibility issues” to yiled a MAC with “same or worse” performance??
We’ll see.
I’m not sure that it is in Intel’s strategic interest to turn away Apple’s business. If Apple cannot get what they need for the products they want to build from Intel, they are still swallowing hard and doing business with Samsung.
In the early 90’s Andy Grove showed John Sculley (then Apple’s CEO) Intel’s projected production numbers for x86. He scared Sculley into starting an effort to move Classic MacOS to x86. Apple actually got it working but backed away at that time from switching from the 68K.
Intel ought to look carefully at the installed base of iOS users and projections of Apple iOS device sales. The volume is in winning Apple’s business. Withholding 14nm only makes sense if Intel can wrest business in mobile away from ARM. With about 2 billion people using ARM code I don’t think this is likely. Abusing customers like Apple is just bad business.
“Withholding 14nm only makes sense if Intel can wrest business in mobile away from ARM. With about 2 billion people using ARM code I don’t think this is likely. ”
That exactly encapsulates my thinking 🙂 Except that I think it is VERY likely if, as I suspect, TSMC and Samsung’s FinFET efforts are late. Intel needs another “Intel Inside” marketing success and they’re not going to achieve that fighting for “brand equity” with the likes of Apple, Samsung, HTC etc. etc.
Those 2B people need to WANT an X86 processor like they wanted a Pentium(tm) in 1994 and they’re going to need a reason to want one. A nice fast CoreM processor with excellent peripherals and software is that reason. Every time Tim Cook stands up and claims technological superiority: that damages Intel’s equity and it needs to be addressed.
As Craig Barrett said … “If you’re going to compete with us you’d better be quick because we’re tearing up the road as we travel.” ARM’s current “incumbency ” in mobile is an aberration which we intend to fix.
They have had two decades to fail at fixing that.
Intel abandoned the low end (controllers and embedded) because they could make more money in PCs thanks to their symbiosis with Microsoft. It was a strategic mistake but it was tactically correct at the time (I was part of those discussions).
They lost sight of their mission which is encapsulated in their name “INTegrated ELectronics”.
“Those 2B people need to WANT an X86 processor like they wanted a Pentium(tm) in 1994”
The new consumer market for devices doesn’t care about specs anymore. The Age of the Nerd is over. Nobody cares about X86, Pentium, Ghz, whatever, those days are over.
You’re right, times have changed but it wasn’t all nerds that bought those 2B PCs delivered before 2007. It was consumers and businesses that needed new capabilities.
The same drivers are in place plus, of course, Fashion, Social Media and “things”.
It needs the right technology and the right marketing.
The market for computing devices today is truly consumer-facing, for the first time. That does change the game quite a bit. Outside of the kind of folks that comment on threads like this (a tiny minority), people aren’t even aware of the technical aspects of the devices they use.
Exactly!! and that’s the marketing challenge.
Consumers need to be persuaded that their new device is worth $50 exra with an Intel processor and having that processor be faster, more customisable, lower powered, with better graphics and proper integration of “peripherals” can only help.
As we have seen with Blackberry … the market is full of pitfalls and opportunities.
That’s why marketing people earn more than process engineers and get sacked more regularly 🙂
Consumers care about jobs-to-be-done. Does device X do Y? Yes? Okay. I don’t believe normal consumers can be made to care about specs. As an example, off the top of your head, please tell me the horsepower and the torque of the car you drive. Also please tell me the cubic foot capacity of your refrigerator. Then please tell me the grade of lumber used in the framing of your house. Most people in the first world own all these things, and yet almost no one would know these ‘specs’. You know that your car is fast enough, your fridge is large enough, and that your house isn’t falling down.
I’m not convinced any amount of marketing will make consumers give a crap about specs. The best that can be done is to show jobs-to-be-done, powerful things that can be done with a device, along the lines of Apple’s ‘powerful’ iPhone campaign. But that doesn’t make anyone care about the processor, it makes them care about the iPhone as a whole.
That’s more or less what I said.
Seemed to me you were talking about specific specs, “Consumers need to be persuaded that their new device is worth $50 exra with an Intel processor and having that processor be faster, more customisable, lower powered…”
The consumer market does not care about the guts of the device, they care about the device as a whole experience. If Apple can deliver a great experience using different ‘guts’, no one will bat an eyelash. Only nerds in comment threads will argue the merits of the specs. Nobody else cares.
Of course that drives the nerds crazy. Imagine talking to the carpenter that framed your house, who is so proud of the quality work and the fine lumber he or she used, and then you tell that person you don’t really care as long as your house doesn’t fall down. That would drive that person batty. How can you not care? Don’t you know about load distribution and wood density!!?? I used the finest screws, and I glued the corners!!!!! And you say “Whatever.” That’s the consumer market.
Its the job of Intel marketing to translate those attributes …..
“faster, more customisable, lower powered, with better graphics and proper integration of “peripherals” ”
Into … “Sweeter form-factor” “Does the job better” “Does things nothing else does” “safer” “more reliable” “More future-proof” etc. etc.
Now – if Apple wants to use their undoubted marketing prowess to share the benefits of Intel’s cutting edge process, that’s fine. But if all they want to be is overlords of a sea of competing widget manufacturers … Intel will undermine them just like they undermined the PPC Mac in 2006.
“Into … “Sweeter form-factor” “Does the job better” “Does things nothing else does” “safer” “more reliable” “More future-proof” etc. etc.”
I would agree. The problem of course being that as long as different guts are good enough to offer the same end user experience, there’s no opportunity for differentiation. That’s the point really, if ARM guts can deliver the same user experience as Intel guts, there’s no marketing wedge to be applied, there’s nothing to undermine.
Of course the question is whether or not ARM guts can be good enough (keeping in mind that slight technical differences simply will not matter to consumers).
Well the “ARM guts” aren’t good enough in desktop or notebook PCs
They aren’t good enough in servers.
They are about equal in chromebooks and tablets now that K1 and Denver are here – but that’s temporary.
They’re “incumbent” in smartphones because Qualcomm has good integration and the other fabless outfits offer “almost good enough” for pennies. But note that ALL of the foundries are operating with some form of state direction or funding so the playng field is not level at the moment ; the price of cutting-edge processed silicon will rise soon.
Its the job of Intel marketing to promote software that makes last years tablet NOT “good enough”. Microsoft (who I have viewed as “the enemy” since about 1998) used to do that job for them.
The ARM guts aren’t good enough today, I’ll buy that. The point is they could be, perhaps sooner than we think. As for promoting software, iOS has everyone beat on that front, by about one hundred miles.
I agree it would be a good idea for Intel to show off capabilities, but once again we’re back to the specs not mattering anymore, it’s the whole device experience that matters as viewed through the lens of software. Consumers would have to see Intel guts doing things that ARM guts can’t do, important things that matter to consumers. And if ARM guts ever do get good enough, ‘poof!’ there goes Intel’s differentiation. Intel can’t just say “Hey, this one is 2.5 and that one is only 2.1”, nobody cares about small technical differences.
Keep in mind that any technical differences have to be very large before normal people even notice. I’m using an iPad 2 purchased in 2011, and it does what I need. A new iPad Air is vastly superior from a ‘guts’ perspective, but in everyday use I don’t notice the difference. We got my parents an iPad Air recently so I’ve used the Air quite a lot, helping them with it.
Another example, we’ve got two iMacs in our house, different years, I couldn’t even begin to tell you from daily use which one has the more powerful processor, I have no idea. I’d have to look it up. But when I get a new iMac, typically after five or six years, then I notice the new machine is snappier, but it takes that large a difference in the guts before I even notice.
Agreed on IOS. That’s the genius of Apple 🙂
“The point is they could be, perhaps sooner than we think.”
ARM have discovered “Pollacks rule” … You need 4X the complexity to get 2X the performance, Meanwhile the power goes up 4X unless your process gets better.
http://en.wikipedia.org/wiki/Pollack's_Rule
Amusingly, ARMs highest performing star (Nvidia K1 and Denver) used the lowest clocked and most crippled Atom and Haswell parts in their launch presentation…
http://www.extremetech.com/wp-content/uploads/2014/08/denver-64-bit-nvidia-core-performance-relative-640×320.jpg
And that on the SAME DAY Intel detailed their next gen process and declared it production-worthy.
Sonner than we think … huh??
I’m not qualified to assess the merits of ARM vs Intel. You may be right that ARM can never get close to Intel’s performance, but of course the processor is just one aspect of an entire system. I tend to think never say never is a wise approach. I think we agree that Intel will have to show apps, jobs-to-be-done that are unique and only possible because of Intel’s performance. Simply marketing a spec sheet doesn’t work today in the consumer market. Early Android device marketing was often spec heavy, and I don’t see that as much anymore. They’re obviously learning.
“Well the “ARM guts” aren’t good enough … in servers.”
You might need to re-examine that statement.
https://www.facebook.com/careers/department?dept=engineering&req=a0IA0000006cPTMMA2
http://semiaccurate.com/2010/08/23/facebook-first-jump-arm-servers/
I might…. but I won’t 🙂
1 job req and a 4 year olf article from Charlie …. I bet that has Intel quaking in their boots. Come back when ARM has a bigger server share than X86 has in mobile.
But first read this :
http://www.eetimes.com/document.asp?doc_id=1319378
My favourite quote is “The road to ARM server is longer and more contorted than previously imagined” – Rick Merritt
What is working against Intel now in mobile is consumers care little about the processor or its branding. Does it run the software I want to run and already own? This is why Intel ruled the PC world but, the bulk of software out there for mobile runs on the ARM architecture.
People want x86 in their PC. Intel will keep that business as long as they tend it well. (Delays in Broadwell make me wonder about that!) It is a shrinking market and if Apple can convert successfully the market for x86 shrinks further.
As far as volumes go in the mobile space the thousands of Android device makers and Apple iOS devices almost equal 100% of the mobile CPU market combined. Apple is the single largest user of ARM architecture CPU’s.
Now if Apple does not switch to Intel and finally finds a suitable foundary for their own designs, where does this leave Intel?
These discussions usually end with “We’ll see, won’t we?”
You seem to think that Intel is alone in struggling to make 14nm work… We’ll see, won’t we ? 🙂 I know where my money is !
Is X86 a shrinking market ?? PC shipments were flat in Q2, AMD grew their X86 console shipments and X86 tablets grew from 0 to 15M in 1H 14. That is not a shrinking market. X86 servers have 90%+ share and are eating IBM and Oracle’s RISCy lunch. Sure, low end PCs have been cannibalised by tablet growth but Intel are addressing that and will have 20% share by year end, aiming at 40% in 2015.
If Apple finds a fab where does that leave Intel??? With 3X better fabs pumping out 14nm and 10nm Silicon to Apple’s competitors plus 3X fully depreciated 22nm fabs ensuring that the foundries can’t make any money from their 28nm and 20nm lines.,All that with a bunch of tame manufacturers delivering better, faster,cheaper end products that take food off Apple’s table.
We’ll see, won’t we ??
I never said that Intel is alone in having trouble with 14nm. It just seems like Broadwell is going to be very late. Part of my speculation is that perhaps the Desktop x86 is way down on the Intel priority list. I am fully aware that Intel is the best in the semiconductor industry at fabrication technology.
Given that, Broadwell being so late could be an indicator of Intel putting all its bets on the mobile x86.
Your speculation that Intel could leverage its fabrication technology to flood the market to overcome its present position with speeds and feeds. Consider this, in the early 1980’s the x86 was far from the best microprocessor technology available. It was limited to 1 Megabyte of addressable memory, and had an ugly segmented memory architecture. (In fact IBM chose that architecture partly because it was not very powerful! The last thing IBM wanted to do was to disrupt their mini computer sales at the time. The IBM PC was intended to kill the Apple 2.) The first IBM PC used the 8088 which was an 8 bit data bus version of a 16 bit processor. The Motorola 68000 at the same time could address a linear 16 Megabyte memory space, had 32 bit internal architecture. If speeds and feeds count so much why didn’t the 68000 become a dominant technology? Simple, the 68000 did not run the software that the IBM PC customer wanted.
What I am saying is that this situation is now Intel’s problem, apps are written for the ARM platform not x86!
What good is 3X manufacturing technology if Intel is unable to sell many processors??? If that were the case then Intel might be forced to shut down excess capacity and may not have as much to re-invest in basic technology.
Sure either scenario is possible but, if you are Intel what is plan B?
BTW cheaper competitors have not taken food off Apple’s table so far….
Ah, but X86 has a very good Android ARM emulator and Intel has a well-funded program to ensure that the vast majority of Android apps “just work” on X86.
http://www.bit-tech.net/news/hardware/2014/08/21/intel-unity-partnership/1
The only other notable mass-market OS (IOS) could be ported to X86 in months (In fact I would be SHOCKED if its not already running in an Intel lab somewhere).
I find it hugely amusing that the ‘plucky upstart’ ARM is now reduced to claiming “incumbency” and spreading FUD in order to protect itself from defections 🙂
http://www.theregister.co.uk/2014/05/02/arm_test_results_attack_intel/
I think you are missing a big point. The overwhelming dominant vendor of Android Smartphones is Samsung. Do you really think that Samsung will switch to Intel processors??? Samsung makes their own why would they want the x86?
This is no slam dunk for Intel. As you say, we will see.
I’m not missing the point at all … having better silicon and better marketing makes the transition inevitable.
Indeed, Samsung will be a tough nut to crack but picture this scenario.
– Samsung is 2 years late delivering a 14nm FinFET process.
-TSMC is 3 years late (since they’re a year behind Samsung)
– All other tier1 vendors transtion to 10nm Intel silicon because its the only way to produce phones/tablets that compete with Intel’s tame “white-box” outfits.
– Samsung products gain a reputation for being slow and “laggy”.
What do you think Samsung would do ???
P,
Doesn’t the CPU vs SOC dimension count significantly against Intel for the next 3 years at least? Today’s ARM chips are complete SOCs with CPU, and several other functionalities. Intel’s CPUs have only the CPU function and perhaps one or two others. This makes a difference under a mobile TDP in delivered power, efficiency (battery life), and cool running temperature.
Also, Apple //may// have the //best// large scale engineering culture on the planet. If they win this race with Intel, they will have at least proven that they do, indeed, have the wicked smartest team that ships*.
*(See 64bit in 1300M A7s with no 64 bit ARM or 64bit x86 shipping in a phone a year later yet.)
I think “better integration” counts for about 1 more year.
“Bay Trail” was an SOC designed for Netbooks so it lacked a lot of tablet and phone functionality. So yes Qualcomm is kicking I’s butt because they have better integration… especially integrated LTE.
Intel has improved with “Moorefield” which has the same GPU as Apple’s A7 plus an “Integrated Sensor Hub” (whatever that is) but still lacks on-board LTE BUT on board LTE isn’t a dealbreaker in hi end phones, apparently.
And next year Intel has its fully integrated “Broxton” offering on 14nm ..
As for who ships what …. we’ll see, won’t we??
Comparing Moorefield with A7 is incredibly misleading, and doing so simply highlights how much better at chip design Apple is than Intel. A7 was announced in 2013 and has been shipping in volume for nearly a year now. Moorefield was “announced” but it’s not actually shipping yet.
http://ark.intel.com/products/81194/Intel-Atom-Processor-Z3560-2M-Cache-up-to-1_83-GHz
http://ark.intel.com/products/81195/Intel-Atom-Processor-Z3580-2M-Cache-up-to-2_33-GHz
http://ark.intel.com/products/75121/Intel-Core-i7-4765T-Processor-8M-Cache-up-to-3_00-GHz
I’d sure hope that Intel’s soon-to-be-launched chip can perform on par with what Apple launched last year…
A: My comment wasn’t aimed at you and
B:Did I compare Moorefield with Apple’s A7 ? I simply stated that it had the same GPU .
The direct competitor to Apple’s A7 was “Bay Trail” which is approximately equivalent on CPU performance but worse on GPU and has about the same power characteristics.
http://www.anandtech.com/show/7335/the-iphone-5s-review/5
And Apple’s A7 has a “glass jaw” in that its is optimised for its application … it reverts to being no better than a 2007 Atom when asked to do sequential memory access. It is NOT “desktop class processor” – it is a “specialist processor” that’s very good at its task, which is running IOS.
http://www.futuremark.com/pressreleases/understanding-3dmark-results-from-the-apple-iphone-5s-and-ipad-air
“Did I compare Moorefield with Apple’s A7 ?” — In my opinion, yes.
“The direct competitor to Apple’s A7 was “Bay Trail” which is approximately equivalent on CPU performance but worse on GPU and has about the same power characteristics.” — And Bay Trail was made at 22nm vs. A7’s 28nm. If Apple’s 28nm design performs on par with and in some cases better than Intel’s 22nm design, Apple designs better chips.
“And Apple’s A7 has a ‘glass jaw’ in that its is optimised for its application … it reverts to being no better than a 2007 Atom when asked to do non-sequential memory access. It is NOT ‘desktop class processor’ – it is a ‘specialist processor’ that’s very good at its task, which is running IOS.” — Like how a B-series chip would be very good at its task, which is running OS X.
“It is NOT ‘desktop class processor’” — Here is what Anand Shimpi said: “With six decoders and nine ports to execution units, Cyclone is big. As I mentioned before, it’s bigger than anything else that goes in a phone. Apple didn’t build a Krait/Silvermont competitor, it built something much closer to Intel’s big cores. At the launch of the iPhone 5s, Apple referred to the A7 as being ‘desktop class’ – it turns out that wasn’t an exaggeration.”
http://www.anandtech.com/show/7910/apples-cyclone-microarchitecture-detailed
You expand the discussion into new areas 😉
On “Desktop Class”:
Anand’s reference is simply to size … Note his use of the word “BIG”.
What he SAYS it that it is “BIG” in that it used more functional blocks than Krait and Silvermont to build those IOS optimised “Six decoders and nine ports to execution units” that only work for sequential memory access. Intel’s “Larabee” was BIG and optimised to a task (twice as BIG as Apples A7) – but it wasn’t a “desktop class processor” in operational terms.
On Architecture:
They could indeed make a “B” series chip that would be optimised to run OSX.
How would that be any better than an Intel Core processor that is also optimised to run multitasking operating systems?? … Extra magic pixie-dust??
Here’s how the last 2 highly experienced RISC design teams ended up pitching their wares against X86
– IBM “Power” are out of the PC business and have less than 2% of the server market by volume.
– SPARC are clinging on in servers burning Larry Ellison’s billions.
DEC Alpha, AM2900,MIPS etc. all died along the way.
On expanding ARM:
ARM and Apple are up against many problems as they attempt to claw their way into the higher regions of CPU performance .
1 is Pollack’s rule …. which applies to everybody
2 is the fact that many of the new features they will need to adopt are patented as Apple have recently discovered after they were sued by WARF over the design of the A7
http://www.patentlyapple.com/patently-apple/2014/02/apples-a7-processor-targeted-in-new-patent-lawsuit.html
On Process:
Bay trail was indeed built on 22nm vs 28nm … but the comparison is misleading because Intel’s 22nm process was optimised for CPUs and speed and the foundry”s process was optimised for SOCs and power. Its like driving a racing car on grass and claiming your tractor is faster.
The first time you will get to make a real apples-apples process comparison will be with the launch of Intel’s X86 “SOFIA” phone chip next year which will be built on a nice area optimised low power TSMC SOC process.
And “enquiring minds” should ask who OWNS and PAYS FOR these foundry fabs that generate the cheap wafers that Apple uses… who paid for their process development … huh??? Why did all the “publicly held” fabs go bust??
“They could indeed make a “B” series chip that would be optimised to run OSX. How would that be any better than an Intel Core processor that is also optimised to run multitasking operating systems?? … Extra magic pixie-dust??”
The B-series chip would be better optimized for OS X, Apple would pay less for it, and Apple’s control of the processor would open the door to better Macs.
“The B-series chip would be better optimized for OS X, Apple would pay less for it, and Apple’s control of the processor would open the door to better Macs”.
That’s what you wrote in the article without any indication of how those “optimisations” would be implemented to allow better performance than an X86 chip that already runs Windows, plus 17 flavours of Linux plus Solaris plus Android and has a dedicated compiler team that has a proven track record of being very very good.
Its the same logic used by medieval monks to predict the 2nd coming:
“That John the Baptist was good! Jesus Christ was great! … Just imagine how AWESOME the next one will be??”
I would add that the only example of an “optimisation” on A7 that you give is their “Secure Enclave” and that’s implemented in software using ARM’s standard “trustzone” extension. X86 has similar extensions for the same purpose : “Secure Key”, “OS Guard”, “Enhanced Identity Protection” and “Trusted Execution” etc.
Apple simply used spin to claim it as their own unique feature. You believed the spin and generated this article based on the above logic. You might have asked the question “So if my fingerprint data is in this new “Secure Enclave”… where did they used to store my equally important password data??”.
Apple sells 20M Macs a year and may indeed generate their own high end specialist “Core-like” processors. If they do it makes sense for it to be an Intel custom foundry deal using an X86 core to maintain compatibility. They could even use that to emulate IOS in the same way that X86 emulates Android and switch their phones to X86 to take advantage of the faster lower powered Goldmont X86 cores on 14nm and 10nm processes.
Apple produces better designs than Intel does. The only reason Intel is able to stay competitive with Apple’s designs is because of its manufacturing advantage.
Intel is not really in any phones yet, but it is making huge inroads in tablets. Samsung, Asus, Acer, HP… all have Intel-based Android tablets. And Intel-based Windows tablets are being introduced by the truckload at IFA… that doesn’t mean they’ll sell.
Android OEM can be CPU-agnostic: take advantage of Intel’s price-dumping today, go back to ARM tomorrow, or even MIPS (there’s the anecdotal MIPS-based Android tablet).
Apple doesn’t have that flexibility.
“Part of my speculation is that perhaps the Desktop x86 is way down on the Intel priority list.”
This is correct. Intel’s CEO said as much:
“I’ve made it Intel’s highest priority to create the best products for the fast growing ultra-mobile market segment,” said new CEO Brian Krzanich in a statement.
http://bigstory.ap.org/article/intel-cuts-sales-outlook-pc-sales-slump
Nope.
It might be attractive on the surface, but examined in any depth this doesn’t really make sense.
——– Why it won’t happen
You would essentially be building a netbook, or even worse, Apples version of Windows RT.
I’ll predict right now, that even an A8 won’t have the performance of the lowest end Intel machine Apples Mac line. So at best you would only be able to build a low end machine BELOW the lowest end machine in the line.
That means you aren’t replacing x86 Macs, but co-existing with them(as second class offshoot). Since you are co-existing indefinitely, you create confusion between x86/ARM Macs.
You are creating the oddball situation where you have x86 macs and new low end ARM Macs with almost no third party software and little incentive to develop for it, because most Macs are x86, even most of the new models will remain x86 for foreseeable future. This is nearly the exact same problem as WinRT. I called it stupid when Microsoft did it, and I would call it stupid if Apple did it. But Apple isn’t run by Steve Ballmer so I have little worry they would bungle something this badly.
——- Better alternative #1
If you want to create a lower power laptop with longer battery life, use a Intel Atom. Same benefits without the grief above.
——- Better alternative #2
Instead of diminishing the Mac, grow the iPad. Build a bigger iPad with some kind of convertible that offers a real decent keyboard.
Well yes if they did what you suggested it would be foolish and so they wouldn’t do it. So if they move to ARM it won’t be to produce a second rate Mac. The limit to the A7 is that it has to run in an iPhone, put an A* in a Mac and the bar for what could be done gets raised a lot.
I used an A8 as still not changing anything. When do you think what I wrote won’t be true? A9? A12? A20?
It’s not changing in the foreseeable future. So doing it remains foolish for the foreseeable future.
Perhaps I wasn’t clear enough in the article, so let me try again now:
Intel designs Atom chips for smartphones and tablets. Apple designs A-series chips for smartphones and tablets. Apple’s A-series chips can hold their own with Atom, even when Atom is produced at 22nm and A7 at 28nm.
Intel designs Core chips for desktops and laptops. Apple is working on B-series chips for desktops and laptops. There’s no reason to believe that the B-series line won’t be powerful enough to run a Mac.
There are lots of caveats and controversy in benchmarking. For example, Apple is excellent in tweaking the software to optimize the benchmark results. If depending on one sources, it sound like cherry picking.
Even if A7 is slightly better than Atom, that does not necessary mean that Apple can make an ARM chip as powerful as the Intel Core series.So far no ARM chip vendor can do so. Why do you think Apple can be an exception?
Apple is the only ARM licensee with a big incentive to create an ARM chip to match Intel’s Core series. If Qualcomm or NVidia created a Core rival, in what product would that chip even be used? A Windows PC? Windows doesn’t run on ARM.
Windows does run on ARM – its just not very popular for some reason 🙂
http://www.microsoft.com/surface/en-gb/products/surface-2
And Nvidia are pushing K1 into Chromebooks and Denver (alledgedly) into boutique tablets
http://www.pcworld.com/article/2458620/acers-chromebook-13-runs-nvidias-tegra-k1-processor-prices-start-at-280.html
They’re “Desktop Class”…. haven’t you heard ???
http://www.extremetech.com/wp-content/uploads/2014/08/denver-64-bit-nvidia-core-performance-relative-640×320.jpg
1) The Surface 2 runs Windows RT, not Windows. Windows does not run on ARM. If Qualcomm developed an ARM chip that performed on par with Core, that Qualcomm chip, due to compatibility issues, could not be used in, say, an HP Envy laptop like this one: http://shopping.hp.com/en_US/home-office/-/products/Laptops/HP-ENVY/J3G66AV?HP-ENVY-15t-Select-Laptop
2) Your comment assumes that Surface 2 didn’t sell well because it’s powered by an ARM chip. Do you think that it would have sold well if it were powered by an Intel chip instead?
“The Surface 2 runs Windows RT, not Windows”
Windows RT is the full version of Windows for the ARM architecture.
The full version of OSX would, would essentially be Apples version of Windows RT.
Dumb idea for Microsoft, and a would be dumb idea for Apple, but of course there are no real indications that Tim Cook is that dumb.
I apologize. The words that I typed did not accurately convey the thoughts swirling around in my head. So let me try again 🙂
“Windows RT is the full version of Windows for the ARM architecture.” Yes. No disagreement there.
What I meant to say, but did not at all articulate, is that the traditional Windows ecosystem is not fully compatible with ARM. You may be able to run Windows itself on an ARM-based chip, but a lot of the other stuff won’t work.
“the traditional Windows ecosystem is not fully compatible with ARM”
As would be the case with OSX version compiled for ARM.
I don’t think it would.
Windows RT (ARM) did not replace Windows (x86). OS X for ARM would replace OS X for Intel. Replacing vs. supplementing creates different incentives for developers.
The comparable event isn’t Windows on ARM. It’s Apple switching from PowerPC to Intel.
In 2006-2007, when Apple switched to Intel, the new OS X replaced the old OS X. It’d be the same with a switch to ARM-based Macs. OS X for ARM would replace OS X for Intel. And developers, like they did then, would port their apps to the new architecture.
You are full of erroneous assumptions. To switch completely from Intel to ARM, ARM chips would have to at least match Intel’s top end desktop chips.
That is a pipe dream.
AMD is a company fully devoted to competing with Intel on desktop and they can’t do it. But you think one small division of Apple will emerge from secrecy with fully formed competitive desktop chips rivaling Intel?
You are seriously out of your depth here. You are making a series of far fetched pronouncement with no real understanding of just how far off the mark you are.
“But you think one small division of Apple will emerge from secrecy with fully formed competitive desktop chips rivaling Intel?”
I’m not qualified to speak to the technical issues here, but this quote does remind me of the Palm CEO pre-iPhone saying “PC guys are not going to just figure this out. They’re not going to just walk in.”
Betting against Apple being able to do something usually isn’t a smart bet, that’s all I’m saying. Now, you could be 100 percent correct, as I said, I don’t know enough about the technical end of things, but your quote did remind me immediately of other times when people said Apple couldn’t possibly do something.
I’m with you, SG. While I am not up on the tech involved and I am sure such a task is monumental. Apple has a track record of taking action when they feel they are being held back by a supplier, either in their ability to be timely with supply or the capabilities of their tech.
Joe
I was going to post that exact same quote…
Guys. You are comparing Apples and Orangutans.
The first iphone was largely a brilliant composition of technology parts, most of them not actually built by Apple. This is Apples forte. design, user experience, sum of parts, that exceed the whole.
But catching the Intel high end CPUs is akin to the worlds fastest marathon runner beating Usain Bolt in the 100 yard dash. Being the fastest marathon runner doesn’t translate into the 100 yard dash.
Even AMD that mainly exists to challenge Intel on the high end, can’t match them, after several generations of trying.
Competitive CPUs don’t emerge from nowhere. They take generations of working in the exact problems space, to create competetive solutions.
This is part of the real reason Intel is behind in Mobile that the most dabblers don’t get. Intel isn’t behind in mobile because there is a problem with x86 architecture, it is because Intel let it’s experience lapse/stagnate in this area, and they will take a couple of generations to get up to competetive levels. This is from a company that has more resources devoted to CPUs than any other on earth. Similarly look to NVidia, the leading GPU company on earth, taking about 4 product generation to build a competetive mobile GPU in SoC part.
Apple’s expertise is user experience (and even growing experience in Mobile SoC) won’t make an Intel competetive desktop CPU, and there is ZERO evidence Apple is even trying to get into this space.
What I think makes such a move improbable is because ARM is not likely anytime soon to offer parity on higher end machines, like the MacBook or Mac Pros. And I don’t see Apple splitting up their desktop/laptop lines into ARM and Intel OS X machines. And I don’t think they are likely to offer an iOS desktop/laptop like device. It might be cleaner to offer a third hardware device, but Apple has shown no indication that they feel there is a need to complicate their offerings into yet another line of device.
But I would not doubt that, in house, they already have OS X running on their ARM processors. That could be either or both, OS X on a mobile device or ARM in a PC device.
Joe
As noted below … Win RT IS windows and it runs on ARM to the limit of Microsoft’s capability (or desire) to port the OS.
1) If ANYONE made an ARM chip that performed on a par with present Intel Core processors the FIRST place they would try and sell it is in servers because that’s where the software overhead is lowest and the ASP is quite high. So far this year at least 3 ARM server initiatives have folded and the best ARM server chip available (AMD’s A1100 “Seattle”) delivers 70% of the performance and uses 125% of the power of Intels competing Atom based server chip (The Avoton C2750).
2) If Surface 2 running Win RT had used an Intel ATOM processor instead of a Tegra 4 it would have been equally pointless but it might have permitted the software overhead to allow proper emulation… like X86 Android does on Atom.
“Apple is working on B-series chips for desktops and laptops.”
You seem to be involved in some kind of circular reasoning here.
Assuming Apple is building ARM Macs, based on your other assumption that Apple is building ARM chips targeted at Macs (because they need them for ARM Macs???).
Where is there any evidence that Apple is working on ARM chips specifically for Macs?
Two assumptions don’t equal one fact.
My starting assumption is that Apple wants to switch to ARM-based Macs.
So how would the company do that?
It wouldn’t put A-series chips in Macs, because A-series chips are designed for smartphones and tablets. Apple doesn’t use Intel’s smartphone and tablet chips in its Macs, and it wouldn’t use its own smartphone and tablet chips in its Macs either. Instead, it would design a line of chips specifically for the Mac.
So if you assume that Apple wants to switch the Mac to ARM-based chips, as I do, then you have to assume that it’s working on a line of B-series chips, too.
So your argument for Apple switching, is you prior assumption that they want to. Circular reasoning at it’s best.
Do you disagree that Apple wants to switch the Mac to ARM-based chips?
If not, how else would the company do it?
I think it should be rather obvious that I disagree with your assumption.
Following Microsoft down the Windows RT road is fraught with substantial downsides and almost no upside.
Only a shallow read of the situation, sees merit in it.
Beyond that it is utterly ridiculous to use your assumption that they want to do this, as “evidence” to support your case that they will do it.
This article has less in common with the typical well thought Techpinions piece, and more in common with the shallow commentators they regularly skewer.
See my comment below. Transitioning to ARM-based Macs would not be “following Microsoft down the Windows RT road”.
I agree on the larger iPad, I’m pretty sure a 13 inch iPad is coming. As for keyboards, there’s lots of great keyboard cases for the iPad. I’m sure that would also be true for a larger iPad.
I think it would be better with an official apple one with keyboard spacing that matched their desktops/laptops.
You might be right, Apple could sherlock the third party keyboards (or perhaps only one kind of keyboard and stuff like the Clam Case Pro will still have a use case). I’m a fan of tablet + keyboard, it’s a great device. But it has to be a tablet first. Then add whatever hardware accessories you need to expand capabilities. I think of the iPad as a Many-in-1 device.
What if each ARM Mac had multiple Arm chips instead of one? If Apple wrote the OS cleverly, Apps could use an idle chip, not a core on a chip mind, but a whole chip for some processor intensive job while the rest of the OS chugged along on the other chip. This would also give power/battery life gains as a chip could be idled when not needed much as individual cores are now.
Apple has already laid the foundation for this with Grand Central Dispatch:
http://en.wikipedia.org/wiki/Grand_Central_Dispatch
Thanks for that link, Matt. Interesting. Mostly over my head but it looks like Apple would be able to push the queue for jobs to be processed up one level to the available CPU or have the OS look at all the cores individually regardless of chip without the developer having to understand the problem directly.
I think this lends more credence to your premise that Apple could make the switch to ARM. Multiple ARM chips are still cheaper than one good (above celleron level) Intel chip and the power/battery performance is better.
“Mostly over my head” — that makes two of us 🙂
That is just a system for queuing/dispatching threads. It doesn’t solve the real issue with parallel processing:
Most tasks aren’t highly parallel in nature. There are only a few tasks that benefit from significant ramp up in cores (mainly rendering activities done by GPU anyway).
This is big reason that, while many Apple competitors went to 4 cores, Apple stayed with 2 cores and worked instead on increasing IPC of the cores. Fewer better cores leads to better performance than increasing numbers of weaker cores.
Not everything is amenable to being programmed to take advantage of multiple cores. Lots of modern software still relies on single threads for most of their functionality. That’s why Apple’s IOS devices are still only dual core while android devices have all gone to quad-core — because the benefits of multithreading are limited for most everyday apps. Just as in the real world, you can’t always split a job in half, give it to two people, and expect it to get done in half the time, there are many fundamental programming tasks that cannot be broken into multiple threads.
Which means that you can’t just throw more cores at the problem, for a general purpose PC, you have to have a CPU where each core is capable enough to get single threaded tasks done in an acceptable amount of time.
“You would essentially be building a netbook, or even worse, Apples version of Windows RT”
Apple’s version of Windows RT is iPad / iOS. An ARM-based Mac would be Apple’s version of a Chromebook.
If iOS is a nephew of OSX why can’t it run on a Mac? Even if it was virtual, it seems it would be a software fork for OSX not a RT type but a family version. Im not knowledgeable to know what you can’t do, so maybe it is Hakem’s razor?
Why would you want to run iOS on a Mac for reasons other than development? iOS and OS X should remain separate. Look at what Microsoft did to themselves.
Sadly Microsoft’s biggest (recent) mistakes were to ignore that Android and iOS apps ran on both phones and tablets. This was their key to their success in the mobile market. Instead, Microsoft made their Phone OS and Tablet OS mutually incompatible. Then they took their Phone/Tablet UI and plastered it on their bread and butter OS and created the marginally successful Surface Pro and wound up with a touch UI on laptops (somewhat useful) and desktops (counterproductive since enterprise failed to love it). Win 9 better have a few miracles planned.
Now they have virtually no hardware mobile presence, a shame since they went through all the trouble of buying Nokia to get one, and a bifurcating main OS.
Apple has the advantage of seeing where Microsoft made its mistakes and won’t be melding iOS and OS X anytime soon, which means no Mac RT. ARM and x86 will stay distinct for the foreseeable future. iOS and OS X will continue to communicate with each other better over time and there will be an increase in filetype compatibility and who knows, iOS devices may become input devices for Macs, but they’ll remain distinct and compliment each other, not try to morph into a single system.
Two reasons why I don’t think this is going to happen anytime in the forseeable future (aside from the whole X86 compatibility issue).
First, Apple has spent a lot of time arm-twisting Intel into building the kinds of chips Apple wants. A big chunk of Intel’s roadmaps have been adopted at Apple’s behest. For instance, Apple boycotted Intel’s chipsets for a few years in the late oughts (sticking with Nvidia chipsets for the Core 2) because they were unhappy with the crappy GPUs that Intel was shipping. Result: intel’s integrated graphics are now (finally) more or less on par with AMD/nvidia mobile GPUs. Right now, IIRC, Apple is Intel’s single biggest customer. So, if Apple feels unhappy with the chips they are getting from Intel, all they have to do is demand a change and they’ll get it.
Second, ARM chips aren’t high performance enough, and they probably can’t be made high performance enough. Forget about A7’s parity with Atom. ARM based servers today are being sold for running highly parallel, low-demand jobs (running Apache, mainly), and they are being compared in benchmark reviews to Atom servers. If ARM could be scaled up today to be competitive with Intel’s fastest server CPUs, we’d be seeing such systems for sale, and we aren’t. Apple doesn’t ship Atom cpus in any of their macs for a reason. In fact, Apple doesn’t ship Celeron, Pentium, or even i3 chips in any of
their Macs. They’re only interested in selling high performance machines with high performance CPUs in them. And right now, it’s not possible to scale ARM cpus up to be competitive with i5 or i7.
Today, Intel 100% owns the high performance ground in CPUs, and they have every intention of keeping it. If Apple were able to create an ARM chip powerful enough to be worthy of putting in a Macbook Air, you can be damn sure that Intel would find out, and they would move heaven and earth to keep Apple’s business.
In order to get this to work Apple would have to string multiple A7/A8 chips together to provide enough horsepower, but such parallel systems are hard to program for
The ATP guys had a decision on this very topic, here’s the link http://atp.fm/episodes/77. FYI John Siracusa, pretty much explains why ARM based Mac’s wouldn’t be something that will happen anytime soon or even viable.
There are several facts and issues the article glosses over:
1- Mobile is “Intel’s top priority” *because* Intel is lagging there.They know they don’t yet have the right CPU, which is why they’re focusing on it and also subsidizing sales in that segment. Generalizing from Mobile to Laptops, Desktops and Servers feels like overreach.
2- Intel does have security extensions which make TouchID-level security possible. Don’t take Apple’s PR at face value.
3- Such a switch would have to be for all machines at once. ie, Apple would have to make a range of CPUs, for the smallest MacBook Air to the largest Mac Pro, via MacBooks, iMacs, Apple TV…. I’m guessing that’s about 1:20 range of perfomance, with 5-10 intermediate levels. That’s a lot of chips to coordinate.
4- going ARM would lock out the users that dual-boot or VM into Windows.
5- and require extensive work on drivers and apps,on top of the OS.
I’m indeed sure Apple has some Mac running on ARM deep in some lab. There’s a huge step going from the lab to the market, though.
Also, laptop and especially desktop processors have a lot more I/O. I’m guessing grafting that back onto a mobile platform will be as hard (and costly in terms of $$, power, space, heat…) as taking it out of a desktop/server platform has proven for Intel.
Thinking back on it, I think it would be easier for Apple to move to keyboard-based iPads (with split screen multitasking, mouse+trackpad support…). Why upset the whole Mac ecosystem, when you can simply build up the iOS one, in laptop and desktop modes.
Samsung, far from a wizard at software, have been doing just that for years on the Android side. Didn’t break compatibility, didn’t complicate the UI for people not using the features…
(sent from my $90 Minix X5 Android desktop)
Why Apple should do this:
1) to lower product costs & boost profits (compared to Intel processors) by $100+ per unit.
2) to gain independence from Intel’s update/release schedule.
3) to create more efficient processors, customized for MacOS.
4) to add a secure enclave to the Mac’s processor, which Intel doesn’t/can’t do.
5) to make it easier to port iOS apps to Mac, Mac apps to iOS.
6) to create the (theoretical) possibility of a true hybrid Mac tablet-laptop.
7) to distinguish MacBooks from copy-cats in the marketplace. Currently, Intel helps copy-cat manufacturers produce MacBook-like products. That will continue indefinitely as long as Intel makes processors for Mac.
Counterpoints to some objections:
1) Those with A8 Macs who want to run Windows can purchase a $300 Intel chip upgrade. Not every Mac owner should pay more for a Macbook so that a few have the option of running Windows.
2) It won’t take “years” to make the transition. I have read more than once that Apple tests OSX on MacBooks equipped with A7 processors — the same as it did prior to switching from PowerPC to Intel processors, and before that from 6800 to PowerPC. When it’s time, Apple will announce the switchover, and everything will just work. You’ve seen it before.
3) The A7 (and A8) are not clocked to run at top speed due to energy and cooling constraints on iPhone and iPad that don’t apply to Macs. Those who say they “already know” about an A7 processor’s performance probably have not tested it inside of Macs, in everyday use.
4) Nothing limits Apple engineers to one new processor a year. Some press accounts say that TSMC will begin manufacturing Apple’s A9 processor in volume in January (for the 13″ iPad or iWatch?). For all we know the company already has an A10 in the works.
5) Although it is easy to rule out parallel processing solutions in theory — i.e., using two or more A8 processors working on the same job as a substitute for one Intel processor — in practice it will come down to what Apple engineers and app developers can accomplish, versus what level of performance Mac users expect. The “known unknowns” in that equation mean that it’s unsolvable, even by the most experienced and insightful observers.
Add the new Swift programming language to the equation and the picture become even more clear… new Mac programs written in Swift will work on both Intel and ARM based Macs.
Someday, all operating systems will run on a single-chip. However, this is already being done, but mainly with Intel chip makers and the i3/5/7 series in production. ARM is mainly Linux associated and it will not incorporate with Intel-based systems very stable-wise because they are governed by simple integrated logic through an ALU and program-stack that is only NT-based meaning Win32 based. In recent studies however, the UNIX compatible Intel iche-core series has proven to have support for all operating systems, because it supports win32 and X. These are programming language-support-agents that are the main constructs for the computer’s architecture.
I could see full-blown Windows 8 … compatible ARM processors coming out in 2018, since it may take a three-year leap from post-development standards in next year’s Gartner Series and InsideTech Reports.
After the three years though, however it will become a constellation prize for all the big players in the tech industry to re-integrate it with the existing Intel technologies and older OS platforms.
I wanted to thank you for this wonderful read!! I absolutely enjoyed every bit of it. I have you bookmarked to check out new stuff you post…
I really like what you guys are usually up too. This kind of clever work and exposure!Keep up the great works guys I’ve added you guys to myblogroll.
I really like and appreciate your post.Really thank you! Keep writing.
Really appreciate you sharing this blog.Much thanks again. Cool.
Thank you for great article. I look forward to the continuation.
Excellent article! We will be linking to this particularly great article on our website. Keep up the good writing.
Good respond in return of this difficulty with solid arguments and explaining all concerning that.
Thanks again for the article.Thanks Again. Cool.
Very informative article post.Really thank you! Want more.
I appreciate you sharing this blog post.Really looking forward to read more. Great.
Great blog post.Much thanks again. Keep writing.
Really informative post.Thanks Again.
I really liked your article.Thanks Again. Keep writing.
Thanks a lot for the blog post.Thanks Again. Really Great.
Here is a complete guide to watch Summer Olympics 2021 live online from anywhere without cable. Stream How to watch Tokyo Olympics 2021 live from home.
Very nice article and right to the point. I am not sure if this is really the best place to ask but do you people have any ideea where to hire some professional writers? Thanks in advance
This is one awesome blog.Really thank you! Keep writing.
I needed to thank you for this very good read!! I absolutely enjoyed every little bit of it. I have you book-marked to look at new things you postÖ
That is a really good tip especially to those fresh to the blogosphere. Simple but very accurate informationÖ Appreciate your sharing this one. A must read post!
Nice post. I was checking continuously this blog and I’mimpressed! Very helpful information particularlythe closing section I maintain such information a lot. I was seeking this particularinfo for a very lengthy time. Thank you and best of luck.
Good blog you have here.. Itís difficult to find quality writing like yours these days. I really appreciate people like you! Take care!!
It was founded in December 2002 and became a publicfirm on May well 11, 2011.
Very well said, your blog says it all about that particular topic.*,**,addiction rehab ppc
When I originally commented I clicked the “Notify me when new comments are added” checkbox and now each time acomment is added I get three emails with the same comment.Is there any way you can remove people from that service? Appreciate it!
Do you want to see what is DeviantArt ? Immediately click our name to see what DeviantArt is. 001
Excellent article! We will be linking to this particularly great article on our website. Keep up the good writing.
I am so grateful for your post. Fantastic.
Great, thanks for sharing this article. Keep writing.
Im grateful for the article.Much thanks again. Cool.
Very informative article. Cool.
Thanks a lot for the article post.
Great article. I am dealing with a few of theseissues as well..
Thanks for sharing such a fastidious idea, piece of writing is pleasant, thats why i have readit entirely
It’s arduous to search out educated folks on this topic, but you sound like you know what you’re speaking about! Thanks
Sweet internet site, super design, really clean and apply pleasant.
Major thankies for the article post.Thanks Again. Want more.
Great, thanks for sharing this article post. Fantastic.
fantastic points altogether, you simply gained a new reader. What could you recommend about your submit that you simply made a few days in the past? Any certain?
Really informative blog article.Thanks Again. Will read on…
A round of applause for your article post.Thanks Again. Will read on…
Very neat blog.Really looking forward to read more. Cool.
You’re so awesome! I don’t believe I have read a single thing like that before. So great to find someone with some original thoughts on this topic.
Thanks a lot for the blog.Really looking forward to read more. Really Great.
I really like and appreciate your post.Thanks Again. Fantastic.
Thanks again for the post.Thanks Again. Much obliged.
I needed to thank you for this excellent read!! I definitely loved every little bit of it. I’ve got you bookmarked to look at new things you postÖ
Appreciate this post. Let me try it out.Here is my blog build muscle fast
When I initially commented I clicked the “Notify me when new comments are added” checkbox and now each time a comment is added I get several emails with the same comment. Is there any way you can remove people from that service? Thanks!
Major thankies for the post.Thanks Again. Cool.
Very good article.Thanks Again. Awesome.
This is one awesome post.Really thank you! Fantastic.
I am so grateful for your post.Really thank you! Much obliged.
You are a very smart individual!
This blog is definitely rather handy since I’m at the moment creating an internet floral website – although I am only starting out therefore it’s really fairly small, nothing like this site. Can link to a few of the posts here as they are quite. Thanks much. Zoey Olsen
I gotta favorite this web site it seems very helpful handy
Hello there! This post couldn’t be written any better! Reading through this post reminds me of my old room mate! He always kept chatting about this. I will forward this post to him. Pretty sure he will have a good read. Thank you for sharing!
Wonderful goods from you, man. I have understand your stuff previous to and you are just extremely magnificent. I really like what you have acquired here, certainly like what you’re stating and the way in which you say it. You make it entertaining and you still take care of to keep it wise. I can not wait to read far more from you. This is really a terrific web site.
Super-Duper site! I am loving it!! Will be back later to read some more. I am bookmarking your feeds also
I’m truly enjoying the design and layout of your site. It’s a very easy on the eyes which makes it much more enjoyable for me to come here and visit more often. Did you hire out a designer to create your theme? Exceptional work!
I loved your blog article. Awesome.
Those are yours alright! . We at least need to get these people stealing images to start blogging! They probably just did a image search and grabbed them. They look good though!
I have been exploring for a little for any high quality articles or weblog posts on this kind of space . Exploring in Yahoo I finally stumbled upon this web site. Studying this information So i am happy to express that I’ve an incredibly excellent uncanny feeling I came upon just what I needed. I most certainly will make sure to don?¦t overlook this web site and provides it a look regularly.
You completed a number of fine points there. I did a search on the subject matter and found the majority of persons will have the same opinion with your blog.
I am so grateful for your article.Much thanks again. Much obliged.
An attention-grabbing discussion is worth comment. I believe that you should write extra on this topic, it won’t be a taboo topic however typically persons are not sufficient to speak on such topics. To the next. Cheers
F*ckin’ tremendous issues here. I am very happy to look your article. Thanks so much and i’m looking ahead to touch you. Will you kindly drop me a e-mail?
you could have an ideal weblog here! would you like to make some invite posts on my weblog?
I appreciate you sharing this blog article.Really thank you! Really Great.
Fantastic blog article.Much thanks again. Want more.
You are my inspiration , I own few blogs and occasionally run out from to brand.
Thank you for the good writeup. It if truth be told was a leisure account it. Glance complicated to more brought agreeable from you! By the way, how can we be in contact?
Thank you ever so for you blog post.Thanks Again. Really Great.
I have been absent for a while, but now I remember why I used to love this blog. Thank you, I¦ll try and check back more frequently. How frequently you update your web site?