Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Status
Not open for further replies.
Another reason to keep the Mac on Intel is that Mac hardware can be used to run Windows and Linux.

Linux already runs on ARM. Runs great. I've got an ARM server running Ubuntu in the closet right now. It probably uses less power than just the fans on the AMD server. Windows 8 will eventually run on ARM as well, so there will be no problem for a future ARM-based Mac hardware to run (contemporary) Windows and Linux.

Someone please slowly explain how an ARM-based processor could come even close to beating, an, um, I'm going to go with G5.

Given the same weight battery with a long enough battery life, an ARM based processor will already handily beat both PPC and x86 chips in performance. The G5 and i5/i7 chips are only faster if you feed them a lot more power and cooling, something which isn't unlimited in a super-thin-and-light laptop or other mobile computer. Even so, 2 chip generations from now, ARM will close most of that gap.
 
Intel could do the same thing — replace the current RISC core (I think it's based on the HP PA-RISC) with an ARM core.

Post a link on the HP PA-RISC idea, or it never happened. HP and Intel worked together on IA-64 (Itanium), but the P6 architecture predates that effort by a long ways.


Even so, 2 chip generations from now, ARM will close most of that gap.

Which will put ARM only two generations behind what Intel will have then.
 
Given the same weight battery with a long enough battery life, an ARM based processor will already handily beat both PPC and x86 chips in performance. The G5 and i5/i7 chips are only faster if you feed them a lot more power and cooling, something which isn't unlimited in a super-thin-and-light laptop or other mobile computer. Even so, 2 chip generations from now, ARM will close most of that gap.

Is this an Apple way of interpreting the specs? Because in English you are saying the following:

* MBP with SB lasts 8 hours on battery charge
* MBP with ARM will be two orders of magnitude slower but it'll eventually do the job. It'll last 20 hours (do not forget that the main power consumer nowadays is LCD panel, not the CPU)

But it'll work just fine in the closet. :D
 
The source of this rumor probably overheard something about Apple and Intel and ARM and switching and decided that Apple is going to switch from x86 to ARM. He got it wrong. Apple is going to switch from ARM to x86. Usually architecture updates bring 15...40% performance boost (just check how iPhone progressed in four years). With this type of progress in 2013 ARM chips will be as fast as Atom chips are today. Does anybody think that Apple will switch in 2013 from Sandy Bridge performance to Atom performance (in today terms)? Obviously not. The reverse switch however does make sense. All indications are that in 2012 Atom will be on par with ARM in terms of power consumption (performance never was an issue) and in 2013 Atom will outperforms ARM in both CPU performance and power consumption terms.

So yes, Apple is switching: from ARM to x86 for iOS (bad news for some iPad owners but not as bad as when switch occurs on desktop platform).

I think that makes the most sense of all. :)
 
If we will be connected to icloud all the time and have all are files, documents, pictures, music, video and games uploaded... then whats the point of having powerful hardware when the data centers can take care of the cpu and gpu intense operations faster and more efficiently then a machine at home. I'd rather have a cheap input device with long battery life and pay for a monthly fee for services such as video and image editing or video game streaming. The hardware you would buy would need a simple cheap processor, Yes???
 
Big Mistake

PC user all my life, but mac won me over 4Q10. My home systems are now all mac and I enjoy the platform. I need to hear more about the compatibility of ARM, but unless it is absolutely, I mean absolutely seamless with intel architecture, then I think the move is a big mistake. Even then, if the benefits of ARM are not not stupidly huge--I still think the move a big no no. Such a move, bringing yet another operating system into the mix would cause me, and I think a lot of others, to pause and question my shift to OSX.
 
Guys, for everyone who missed it, there is no move to the ARM architecture for Macs. The actual rumor is that Intel will build the next ARM cpus used in the iToys. This just got messed up in the media and now we have this overblown "rumor" that is about as serious as a clown on opening night.

That other rumor of x86 iToys is also a misunderstanding : Intel can very much manufacture ARM processors, they have done so before with the StrongARM line.
 
Lol

My Lord, this is funny.

I'm going to have to call this one a no go. ARM lags far behind Intel when it comes to chip technologies that help mitigate the limits presented by the physical media of the chip itself.
 
Why do people keep saying that ?

Because it's what Apple is aiming for in the long run. If the iPhone/iPad weren't the hottest devices in the solar system, maybe this wouldn't be happening. But let's face it, while Macs may be more profitable, iOS devices are outselling them by huge margins. What signal does this send Mr. Jobs? It tells him that portable/thin is even more where it's at than maybe even he imagined and they don't come much thinner! ;)

iOS and OS X already share the kernel and Foundation, not to mention Quartz. That's about the only parts you can converge. The UI itself (Appkit vs UIKit) use completely different paradigms, so there's no real possible merge there, nor would you want or wish to.

And you haven't noticed a signal flare lately (cough...Lion...cough) that Apple intends to slowly change that Paradigm? iPad is getting more desktop features (e.g. print, dual-cores, more ram, larger more desktop-like programs, etc. and eventually more storage I'm sure). Meanwhile, Lion not only has the mid-Snow Leopard "App Store" that is awfully similar to the iOS App store (which hints at possible closed software 'some day', especially if most sales shift to the App store; that would signal Apple that people don't care about being able to buy open software and if it's not going to hurt their business, it sure as heck IS going to help it with a 30% take of software sales.

Meanwhile, Lion is taking even more steps to merge the GUI elements. Launchpad is just a start to get users used to a new way of loading/controlling their "App Store" software. And gee doesn't it look a bit like iOS methods? It will peacefully co-exist with the Dock (for now). Now look at the patents Apple has been filing for new monitors that tilt down so they can be used more like a tablet when needed and the multiple monitor output capability of Thunderbolt and it's not hard to see hybridization of the traditional desktop and the iOS touch interface that people seem to enjoy. Look at the trackpad changes Apple has been making over the past several years with gestures, multi-touch, etc. Notice how it's now a free option to get the magic trackpad instead of a mouse. I think you will eventually (when it's cheap enough) see a "iPod Touch" type trackpad that acts as both a magic trackpad and additional display device and dedicated launch pad device. This will combine iOS display + external monitor into one hybrid package for the new notebook and perhaps a larger 'iPad" sized display for "desktop" integration. In fact, you will probably be able to plug them in and use them as such on (then) older Macs to get the same functionality. Everyone has one anyway, so why not use them on the desktop?

The point is that while they are separate now, they are heading for integration. It will take time. ARM on the desktop isn't happening this year or next year. Maybe in 2013. They probably won't abandon Intel entirely until at least 2014. But you can bet if they can even get in the same relative 'solar system' as Intel, they will dump it. It will give them some separation from Windows again (it's just too easy to switch back right now; you only need to reboot to dump OSX) and it may make them more money in the long run and money, image and vision is what matters to Steve, not functionality or we'd have USB3, OpenGL4.x, SLI support, updated graphics drivers, etc. by now. They are clearly low on Apple's priority list.

ARM does nothing to converge things.

Nothing? I wouldn't go that far. Future apps will likely require only one compile to generate both desktop and mobile versions of the app. Once they merge the interfaces into their compiling tools, it'll likely be standard. On the plus side, it means you'll likely be able to play all those iOS games on your iMac and MBP (seeing the lack of games on OSX, this could help in that area). Given ARM will suck for full scale computer gaming, that's probably all you're going to get in the future anyway. Game developers avoid OSX, but are now supporting iOS so again, I give the Wii Versus Xbox360 analogy. Many games for Xbox360 aren't available for Wii (not enough power), but there are other games for Wii that are not available for Xbox360 either. The Mac hasn't seen a lot of that sort of thing in awhile. The market is too small to ignore a PC release, but you do see it with iOS due to the popularity of it.

I do agree that ARM is way too slow to replace Intel right now, but if Intel gets into the ARM game themselves, things could change quickly. They already have the technology to move quickly and their new 3D transistor technology is just itching for a new experiment in faster/lower power. They are already teaming up with Apple as it is, so I wouldn't write anything off. That 2005 PPC/Intel thread linked here is hilarious in how much is sounds exactly like the "it can't be", "never happen", "it'll kill the Mac" talk in this thread. The only major difference is that ARM is slower whereas x86 was faster at the time. But we're probably talking about a 4-6 year time frame here. I'm sure Apple will keep their options open, however. I do think they are planning to merge the GUIs of iOS/OSX over time in a hybrid-like format as described based on their current plans through Lion.
 
Last edited:
Because it's what Apple is aiming for in the long run.

You missed the point entirely : iOS and OS X already are converged. They always were. The same way Meego and Ubuntu are "convergeances". Both using the same underlying software and architecture.

The mouse/keyboard paradigm will always be different from the Touch screen paradigm. Touch screens will never work on laptops and desktops, tactile feedback and a mouse are superior input for those.

ARM doesn't change this. ARM doesn't make "one compile for a universal app possible", that is already possible with current Universal binaries that can house an ARM iPhone app, an ARM iPad app and both a x86 and x86_64 Mac app.

People claiming "convergence" is coming are blind to the fact is already is here. That was my point, you missed it by about 300 miles but your rant was an entertaining "end of the world" type rant so I'll give you points for that.
 
Rumor: will Apple move from Intel x64 CPUs to ARM CPUs in future laptops?

Perhaps, but it is still a silly rumor. iOS is CPU agnostic so it could just as well be the reverse, if Intel decides to give ARM some competition (it is a fast growing market, I think it is only reasonable that Intel isn't just going to sit on its laurels)

It is safe to say that today and for the foreseeable future (next year) then the answer is no. Apple will continue to use x64 (aka x86-64) on all their Macs for the next year.

The performance of currently available ARM processors is (on a good day) similar to the PowerBook G4. Just to put things in perspective. And the G4 had a way better FPU regardless (though the GPU performance of the A5 is way better than what was offered on the PowerBooks).

Point is, moving to ARM would be a major setback in performance on the Mac Books. So does anyone think that the next line of Mac Books or the one after that (or the one after that) is going to offer ARM processors?

Well, if you think having the equivalent of a decent G4 PowerBook with a 24 hour battery live sounds good, then perhaps you would. I for one, don't.
 
Windows 8 being available on ARM platforms would make this move, albeit a bold one, pretty viable.

I strongly doubt Apple cares about Microsoft's plans. Furthermore, MS will probably not changes the architecture support for the high-end versions of the OS. ARM is for lightweight low-power applications. Sure you can throw in more cores, but unless the developers are writing code to take advantage you really haven't bought anything.
 
Agree with the above but I'll add more. Where will we be in ten years? In 20? I think maybe we might be talking about "kilo-cores" in 20 years.

I've hear so many people say "I don't need more than four." Yes you don't need more than four to run the software you run today but what software will you be using in 20 years. I suspect you wil be able to

(1) Point with fingers, like you do with a multi touch pad but only in the air in 3D space in front of your screen. Stereo web-cab will track all 10 fingers and your eyes. as you move a 3d object that floats in front of the screen.

(2) voice will be used to some extent. Most likely to do a google-like search of voice messages or other audio/video recordings. (A student might ask his tablet to find the place where his teacher talked about some subject.)

(3) the "Find my car keys" app lets you wave the camera around your house and it will find some object for you.

(4) Searching a text book by "idea" rather than text string. "Find where it talks about that civil war battle, the one that had Gen McClellan and that bridge". (Yes I purposely messed up the question, software should figure it out.)

(5) Automated assistant app for phone: you say to the phone: "call fred, tell him I can't make the meeting this afternoon." The phone figures out which of ten Freds and places a voice mesage in his phone for you.

OK you get the idea, NONE of the above is going to happen on an 8-core Intel CPU. Those are kilo-core problems and all those 1,000 cores need to run on battery power.

Couple of things

1. No way you are going to be using your computer 20 years from now. I probably don't need more than 4 cores now but if I do I will buy a new system (a computer is essentially useless after 5 years).
2. Most of what you describe is software not hardware.
3. (bolded text) SERIOUSLY!?!? You think the best that intel is going to have in 20 years are 8 core CPU's! I mean, look at how far they have gone in the last 20 years.

Not to mention that now what uses the most power when the CPU is idle is the screen.

I think that the best architectural design change would be to design the processor to actively shift activity. Instead of separate cores design a big block of circuitry that can process so many calculations a second and have it dynamically alter itself to shift from a virtual so many core mode to a different core mode. Of course the laws of physics may be incompatible with this.
 
Guys, for everyone who missed it, there is no move to the ARM architecture for Macs. The actual rumor is that Intel will build the next ARM cpus used in the iToys. This just got messed up in the media and now we have this overblown "rumor" that is about as serious as a clown on opening night.

That other rumor of x86 iToys is also a misunderstanding : Intel can very much manufacture ARM processors, they have done so before with the StrongARM line.


Can you supply some links so I can catch up on this?
 
You missed the point entirely : iOS and OS X already are converged. They always were.

That depends on what you mean by "converged", though. iOS Apps do not run on OSX (they could with a recompile and a few interface changes, but they do not right out the door. Furthermore, OSX software will NOT run on iOS devices either. Underyling the same or not, there's a distinct gulf between them at this point in time. Converging hardware and/or interfaces would be necessary to make them completely compatible. I've said for years that OSX should have an iOS mode to run straight iOS software (e.g. for gaming for reasons alluded to above). The developer software certainly does run it, for example so it wouldn't be hard. But to make the iOS STYLE interface work on normal OSX, some changes have to be made. Apple is clearly making some of those changes (e.g. Launch Pad).

The same way Meego and Ubuntu are "convergeances". Both using the same underlying software and architecture.

So does my PPC PowerMac running Leopard (more so than iOS) and yet I cannot run Firefox 4 on it. I cannot run various Intel-based games on it. I cannot do a lot of things on it with newer software and there's no simple recompile to make it happen either. Convergence means more than just sharing the same underlying architecture or anyone with Darwin could just run OSX apps.

The mouse/keyboard paradigm will always be different from the Touch screen paradigm. Touch screens will never work on laptops and desktops, tactile feedback and a mouse are superior input for those.

Never say never. Apple has already submitted patent designs for adjustable touch-screen displays (probably for future iMacs). Trackpads could easily be integrated iPod Touch displays (clear trackpads with output). Predicting the future reliably isn't easy.

People claiming "convergence" is coming are blind to the fact is already is here. That was my point, you missed it by about 300 miles but your rant was an entertaining "end of the world" type rant so I'll give you points for that.

I know your points about the underlying stuff, but you seem to think the overlying parts cannot be merged and that's nonsense. An iPad is just a keyboard-less Netbook for all intensive purposes. It could easily run OSX proper if Apple wanted it to (and most Apps would then just need to be compiled for ARM; or they could have used an Atom like the Hackintosh Netbook I'm typing on right now).

You could have a future notebook that's like an iPad but has a slide-out keyboard tray (or flips over like some older notebook/pads did). The idea that the notebook/mobile lines couldn't be combined is absurd. They're easy to combine. It's the desktop line that's in question, but it can certainly take on many of the same interfaces with trackpad instead and/or a hybrid combination of touch-screen + keyboard/mouse. Again, the patent suggests Apple already has this in mind. Apple clearly believes the "trucks" need to be "sports cars" in the future. I don't have to like it, but I'm saying that's where they're going. Feel free to disagree. It won't change my opinion about the signs I'm seeing from Apple (regardless of what future CPUs they may or may not use).
 
iOS 1.0 = Panther
iOS 2.0 = Tiger
iOS 3.0 = Leopard
iOS 4.0 = Snow Leopard
iOS 5.0 = Lion

Based on the version of Darwin, and the API changes coincident with Mac OS X, too. Until iOS 4, the iOS release schedule was behind Mac OS X. If Apple releases iOS 5 alongside Lion, things got normalized.

Your definition of fork is kinda wrong. A fork is a copy, whose changes don't affect the parent. iOS 4 and Snow Leopard is actually the same thing, only thing is that AppKit has been replaced with UIKit, and Aqua has been replaced with the touch GUI. You might consider iOS a fork off Mac OS X.
And for PPC, there is no difference at all between Leopard PPC and Leopard Intel, there is just Leopard. The Intel compatibility exists alongside PPC, hence it's not a fork. You just didn't know it's there because there were no Intel Macs, thus Apple did hide it from us gossipy dudes. Look at Linux and on how much architectures it officially runs - none of it is a fork. But if there are no Alpha Macs, why should Apple A) bother implementing Alpha compatibility and B) tell us that technically, it runs on Alpha, but you can't because there is no consumer hardware.

Mac OS X does not have to be ported to ARM!
iOS 4.0 is Snow Leopard, running on ARM. All you need to do is put AppKit and Aqua back in place. This involves basically no to very little change at all. Just like Wolfram Alpha was ported to Intel by one dude in a couple of hours, because as soon as Darwin and even layers further at the top are ported to ARM, which they are and which is actually the work intensive part, it's kiddie stuff to port the higher layers - in case they even need porting at all. Wolfram Alpha needed addition of Intel assembler alongside PPC assembler. Aqua and AppKit shouldn't come with that much assembler at all, and even if it does, Apples workforce is a little larger than a single dude who didn't even had a clue what he has to do before he got briefed by Apple.

Besides, we're not even allowed to look inside our Mac without voiding the warranty. What do we even care what's actually in the neat aluminum case, as long as it's as fast as possible and runs our apps? Apple could as well replace Mach-O binaries with byte code and change architectures with every hardware refresh, and we wouldn't even notice, nor would we really care as long as it's faster than the previous iteration.

If I need massive parallel power for video editing, I need a 1024- or 2048-core ARM machine, not a 24-core Xeon.



I don't really think iOS and OSX are on par already... May in a year or two...
 
When it comes to hardware/architecture...Apple can never make up it's mind.

Since Apple is the ONLY company that makes Apples/Macs, it should be pretty easy to standardize and come up with a great business relationship. It doesn't have to last 50 years but 5-10 years seems a bit short, especially considering all the kudos AND success of moving to Intel ~10 years ago.

But yet Apple loves to flip flop every 5-10 years...let's make clones! let's not. Let's buy Intel! Let's not. Let's use ____ vid cards...Let's not.

Us technology people can sometimes appreciate the Apple moves...but this is not the Wintel world where there are billions of choices of hardware. Apple has done EXTREMELY well with Intel and moving away could really hurt on many levels.

You are looking it the wrong way.
It's not about making their mind up... It's about giving the best to the consumer or at least trying to make the best deal out of it... Take for example the move from NVIDIA to AMD in the GPU's now...
 
I don't really think iOS and OSX are on par already... May in a year or two...

While iOS and OS X share the same base, Apple would have to neglect the Mac OS severely and update iOS aggressively at the same time so they would be anywhere near each other in two years....

...oh yeah... two years sounds about right. :p
 
I hope this is just another silly rumor.

Does Apple really think they can match Intel when it comes to processor development?

Does Apple really think it would be a good idea to subject users to a third processor architecture change?


As for the architecture change it's just about a matter of time.
And as for development ARM has improved more recently... Have you checked the dual core in the iPad2?
 
Almost laugh worthy considering that ARM has shown nothing to rival the likes of SB, much less the anticipated performance of IB. I'll believe it when i see it:rolleyes:

Like stated before several times something like is isn't due for the next 2-3 years... And a lot of things can happen in 3 years...
 
You are looking it the wrong way.
It's not about making their mind up... It's about giving the best to the consumer or at least trying to make the best deal out of it... Take for example the move from NVIDIA to AMD in the GPU's now...

That's being kinder to Apple than they deserve. It's about making money and appealing to as many as is possible, thus ensuring the largest possible market.

That involves finding the lowest common denominator, not giving the best to the consumer. I guess you could say in all fairness that Apple tries to make the best deal out of it, but giving the consumer the best? Not so much.

If that was the case, then Apple would have made the iMac slightly thicker, to be able to use a desktop GPU, supported Blu-ray already and offered a matte option to the screen. Instead Apple offers an average, but acceptable and appealing products. Shiny, thin and limited.

I don't mind it so much, but I wouldn't credit Apple with offering the best to the consumer either. Although not offering full BD support in the OS is nothing short of bewildering.
 
Like stated before several times something like is isn't due for the next 2-3 years... And a lot of things can happen in 3 years...

Exactly, which is why this rumor is stupid. It's not like Intel is going to ignore the rapidly growing handheld device market. In 3 years there may well be an Intel product that puts the ARM to shame.

They have the money, the know-how and the ability to do it. So a rumor predicting what will happen in 3 years is basically worthless (and assumes that ARM is the only thing being developed in the next 3 years)
 
Status
Not open for further replies.
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.