Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Time for ARM-based Macs?
- Thread starter osx11
- Start date
- Sort by reaction score
Well, not yet, because Skylake will be a huge improve in performance and GPU graphics (nothing like Ivy to Haswell I bet, that was hilarious) the next year. After that, if Apple A chips catches or at least are on par with Intel, then Apple might switch, not before.
But hey, seeing what Apple has done with Mac Mini, we know now that they don't doubt on downgrading a product.
Edit: I've seen the Ipad Air 2 benchmarks right now, and despite being a lil under the actual MBA (20-25% less performance) it's impressive, but again I think they'll wait to see what Skylake can offer, not before
But hey, seeing what Apple has done with Mac Mini, we know now that they don't doubt on downgrading a product.
Edit: I've seen the Ipad Air 2 benchmarks right now, and despite being a lil under the actual MBA (20-25% less performance) it's impressive, but again I think they'll wait to see what Skylake can offer, not before
Last edited:
Arm
Technically, they could probably switch right now. I bet they have been building OS X for ARM CPUs for a while to see how it runs.
I think they are probably trying to position the new device (Macbook with an ARM) in such a way that Intel doesn't feel threatened. The other thing to figure out is how to differentiate an iPad with a BT keyboard and this new device? In other words how do they justify spending $1000 or more on this new device that doesn't exist yet (at least not officially) if an iPad with a BT keyboard can do about the same for their target consumers?
Technically, they could probably switch right now. I bet they have been building OS X for ARM CPUs for a while to see how it runs.
I think they are probably trying to position the new device (Macbook with an ARM) in such a way that Intel doesn't feel threatened. The other thing to figure out is how to differentiate an iPad with a BT keyboard and this new device? In other words how do they justify spending $1000 or more on this new device that doesn't exist yet (at least not officially) if an iPad with a BT keyboard can do about the same for their target consumers?
You are right, they have been testing them.
Intel has already been threatened, they know about the increasing performance of Apple's integrated chips.
Macbook $899 for 128GB SSD, 11in screen, 4GB RAM. iPad $799 for 128GB SSD, 10in screen, 2GB RAM. Macbook includes >$250 Intel chip, iPad includes touchscreen sensor and <$50 Apple chip. Doesn't seem like there will be any pricing issue to me.
But does ARM have the power to drive a retina display in a laptop and process other functions as fast as current Intel microprocessors at the same time?
I would be all for Apple producing their own microprocessors if such can be functionally competitive.
I don't think that ARM processors will take over any time soon. I hate to admit it, as much as OS X is a superior OS in my opinion, we can't simply cut off a large number of Windows users. Most people need Windows in some form or fashion; whether it be for Office, etc.
A lot of people use iOS devices, but not necessarily a lot of people use OS X-powered devices. I mean, there are at least a million users (the amount of beta testers that signed up), but there are ALOT of Windows users. Including organizations, etc. powered by Microsoft Windows. It's unfortunate.
Anyways, I don't think Apple could pull that off. And, honestly, I don't think they should try to anyways. It's okay to test it, but Microsoft already tried to, and failed. Well, they deserved to fail, because they presented a crappy mess. They should have gotten the major apps and pulled all the strings first.
A lot of people use iOS devices, but not necessarily a lot of people use OS X-powered devices. I mean, there are at least a million users (the amount of beta testers that signed up), but there are ALOT of Windows users. Including organizations, etc. powered by Microsoft Windows. It's unfortunate.
Anyways, I don't think Apple could pull that off. And, honestly, I don't think they should try to anyways. It's okay to test it, but Microsoft already tried to, and failed. Well, they deserved to fail, because they presented a crappy mess. They should have gotten the major apps and pulled all the strings first.
What Microsoft did wrong was make PC users use a touchscreen interface, and make two versions of windows that were incompatible with each other (And one that couldn't run most Windows programs). Apple won't make any of these mistakes when switching to ARM.
Apple uses the Swift programming language on both iOS and OS X, and apps can easily be written for both iOS and OS X and ARM and x86, using the same code. When Apple makes an ARM-based OS X, it can just use all the Mac apps just the same, and easily port over OS X (I'm sure Apple already did, with their ARM Mac prototype testing. And it's not that hard, since Apple has already put 64 bit iOS [a branch of OS X] on ARM, I'm sure Apple can put all of OS X on ARM)
True, but as time passes, hopefully developers will write for OS X the programs you use on Windows.
Yes. Apple uses the GPU technology on the market for its integrated Ax chips, and this is steadily becoming powerful enough to drive a retina display, as consumers can experience firsthand with the graphics performance of the iPad Air 2.
Apple is producing their own microprocessors. And yes, they are functionally competitive.
The day apple goes to the ARM platform for their Mac line, is the day I stop buying Macs.
Just because it performs well on a benchmark doesn't mean it will be a good replacement to the intel cpus.
We're talking about a higher level of multitasking/multithreaded processing, and GPU performance. Aside from the hardware advantages of the Intel chipset, I also find the lack of software compatibility, particularly the ability to run windows as a show stopper for me with an ARM based Mac.
Just because it performs well on a benchmark doesn't mean it will be a good replacement to the intel cpus.
We're talking about a higher level of multitasking/multithreaded processing, and GPU performance. Aside from the hardware advantages of the Intel chipset, I also find the lack of software compatibility, particularly the ability to run windows as a show stopper for me with an ARM based Mac.
You are right. People are really just guessing at the moment based on how fast they think an iPad is. We got a little taste of this with Windows RT on a Tegra chip and it was wicked fast but still slower than a Quad Core Atom... (in terms of CPU performance) Apple claims all this performance but if they ever think it belongs in a Macbook that is designed to get work done; it better not be a downgrade from the baseline Macbook Air or they would be making a mistake.
I continue to wonder why some people want this.
Do you really want to go through yet ANOTHER architecture transition, requiring apps to be recompiled, slow emulation to run legacy apps, all the bugs that WILL result from this, and so on?
Don't you remember the pain of the PowerPC -> Intel transition? That transition was required because PowerPC wasn't delivering the needed performance. However, Intel chips are providing excellent performance and are rapidly catching up with ARM when it comes to energy usage.
I don't want to deal with this crap again. The Mac platform has been through too many sweeping changes that break software compatibility and cause huge headaches. To compare, a modern Windows system can run software from the 90s. Try that with a modern Mac. HAH.
680x0 -> PowerPC -- Necessary, 680x0 architecture was rapidly falling behind.
MacOS Classic -> OS X -- Necessary, Classic architecture was very outdated and had no future.
PowerPC -> Intel -- Necessary, PowerPC did not deliver the performance needed.
Intel -> ARM -- NOT NECESSARY. Intel chips perform great and power consumption is getting better every day.
Let's leave it how it is. I don't want to have to replace all my software YET AGAIN.
Do you really want to go through yet ANOTHER architecture transition, requiring apps to be recompiled, slow emulation to run legacy apps, all the bugs that WILL result from this, and so on?
Don't you remember the pain of the PowerPC -> Intel transition? That transition was required because PowerPC wasn't delivering the needed performance. However, Intel chips are providing excellent performance and are rapidly catching up with ARM when it comes to energy usage.
I don't want to deal with this crap again. The Mac platform has been through too many sweeping changes that break software compatibility and cause huge headaches. To compare, a modern Windows system can run software from the 90s. Try that with a modern Mac. HAH.
680x0 -> PowerPC -- Necessary, 680x0 architecture was rapidly falling behind.
MacOS Classic -> OS X -- Necessary, Classic architecture was very outdated and had no future.
PowerPC -> Intel -- Necessary, PowerPC did not deliver the performance needed.
Intel -> ARM -- NOT NECESSARY. Intel chips perform great and power consumption is getting better every day.
Let's leave it how it is. I don't want to have to replace all my software YET AGAIN.
I want to start by saying I agree with the above post, but it's worth noting that if any group can handle this kind of switch, it's the veteran Mac user! And in all fairness, transitions shouldn't be ruled out. As noted above, Apple has switched many times in the past and despite growing pains, it's been to our advantage in the long run. Who knows how things will be in the future? We should stay flexible.
I still see an MBA convertible coming eventually. Think about it:
Design:
Take the rumored "iPad Pro" with a really powerful ARM CPU/iOS and use that as a detachable screen on an Intel-powered MBA base that has OS X.
How it could be pulled it off:
"Continuity" seems like a good first step towards doing this right where others have failed. You'd just want to automate the process a bit more on a convertible.
Use some kind of extreme near-field EHF wireless to get the video signals from the Intel base to the screen, a pad/pin type connector, or possibly an air-gapped optical thunderbolt port. Then add a few pad/pins for power/comms and some wireless for proximity detection to put the base in a "power nap" like state to get things ready for "docking".
What could stop it:
The "hinge". Doing that wrong (like letting Ive design it) would turn a ton of people off as I suspect such a device would be in the $2000-minimum range. I don't see how they could make it look like an Apple design without it trading off a lot of the durability that people expect from a laptop.
Design:
Take the rumored "iPad Pro" with a really powerful ARM CPU/iOS and use that as a detachable screen on an Intel-powered MBA base that has OS X.
How it could be pulled it off:
"Continuity" seems like a good first step towards doing this right where others have failed. You'd just want to automate the process a bit more on a convertible.
Use some kind of extreme near-field EHF wireless to get the video signals from the Intel base to the screen, a pad/pin type connector, or possibly an air-gapped optical thunderbolt port. Then add a few pad/pins for power/comms and some wireless for proximity detection to put the base in a "power nap" like state to get things ready for "docking".
What could stop it:
The "hinge". Doing that wrong (like letting Ive design it) would turn a ton of people off as I suspect such a device would be in the $2000-minimum range. I don't see how they could make it look like an Apple design without it trading off a lot of the durability that people expect from a laptop.