You think Apple will want to set the impression that x86 is “pro” and their own chips are therefore inferior? I think they will avoid that.
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.
Kuo: Apple to Launch Several Macs With Arm-Based Processors in 2021, USB4 Support Coming to Macs in 2022
- Thread starter MacRumors
- Start date
- Sort by reaction score
The question will actually be:
How many OSX customers are moving to an ARM based OSX?
How many are moving to W10?
My mission critical software got a whole lot better (and faster) by my moving from OSX to Windows 10. I truly did not realize how far behind I was sticking with OSX. The software just works. Hell of a thing to say, but there it is. It is a TCO win, any way you want to look at it.
When I made the jump from OSX to W10, I just reinstalled all but 1 of my applications and was good to go. Zbrush is the last product I need to move. $800 price tag, so that isn't moving until Xmas.
My legacy apps continue to function - I only need to upgrade programs when something is added that I could actually use. Couldn't do that with OSX.
When Apple abandoned Aperture I learned to NEVER again make Apple product "busines critical" and always have a plan for moving away from Apple. It is really dumb to be dependent on any single supplier.
Most Mac users use their Mac for web serfing, email, and Youtube. The ARM will be fine for that. I'm in the minority and need to use a virtual machine to run other OSes.
I guess I could simply drop the Mac and use real rather than virtual PC hardware.
Most Mac users use their Mac for web serfing, email, and Youtube. The ARM will be fine for that. I'm in the minority and need to use a virtual machine to run other OSes.
I guess I could simply drop the Mac and use real rather than virtual PC hardware.
I doubt that ARM would have a problem selling a license to whomever wants it. They get paid whether Apple, Intel, or AMD starts making ARMs, so I think Intel very well could make ARMs. Whether they’d do it is another thing altogether.
We’ll see once they ship. Most of the benchmark comparisons are AMD to Intel and and they show AMD beating Intel, but Apple’s been benchmark beating Intel for years before AMD got to this point. I think you’re right in saying that “Nothing ARM designs offers will touch Zen 3/RDNA 2.0”, but what Apple designs already is within shooting distance of AMD’s benchmarks and that’s with processors that shipped last year (and, in some cases, 2018).
Why don’t folks here actually read the context of what they’re commenting about or are they so desperate to make a point of any type for the sake of it. One more time.
- I have no issue with Office 365, Office per Se when I can run the full-feature product. Office is a brilliant package on X86 for example.
- I can afford Office in any guise...I choose 365 for my business
- I could not execute my job or run my business without it, or at least with any modicum of success
As much as I enjoyed Logic, I am dabbling in Ableton Live just in case. I also chose Resolve over FCPX for the same reasons as you.
But why people are so excited about an ARM based Mac ?
There is 0 benefit for the end user (even LTE connectivity is possible in x86)
An ARM based Mac will NOT be more performant to an x86 one, people who think otherwise don’t understand what ARM and x86 instruction set are, in fact in real world usage it will be hugely underpowered ! Benchmark are not the real world !
An avx-512 capable intel chip will crush every ARM chipset without much effort !
There is 0 benefit for the end user (even LTE connectivity is possible in x86)
An ARM based Mac will NOT be more performant to an x86 one, people who think otherwise don’t understand what ARM and x86 instruction set are, in fact in real world usage it will be hugely underpowered ! Benchmark are not the real world !
An avx-512 capable intel chip will crush every ARM chipset without much effort !
I really can’t wait. Even just to stop silly comparisons based on specs with Windows computers.
I'd be surprised if Intel aren't at least mulling this over behind the scenes, and testing and developing their own line of Arm based chips. What better guarantee of the company's future than to be ready for an if you can't beat 'em join 'em scenario! I would add even if Windows on Arm doesn't eat properly into the ultrabook market, Microsoft seem to be set on it as their answer to Chromebook, so it might become a sizeable market in its own right.I doubt that ARM would have a problem selling a license to whomever wants it. They get paid whether Apple, Intel, or AMD starts making ARMs, so I think Intel very well could make ARMs. Whether they’d do it is another thing altogether.
Indeed. If the main idea behind running ARM is easier development to Mac and iOS so one should just skip the Mac part all together. The iPad pro is as capable a MacBook air for 95% of the stuff and can do 130% of what an ARM Air would do.
If the goal is to reduce cost, one should start with throwing out the touchbar from the MacBook pro.
Battery life has not been an issue in Macbooks for a while, especially when you get 16000mAH Powerbank for $50 nowadays.
[automerge]1585389064[/automerge]
Over at the AMD Mac thread people were pointing out thatMacOS can't use more than 64 threads. This was part of the argument why Apple won't go over to AMD.
Wrong. The idea of using ARM-Based on some of the macs is to provide better performance without the need for active cooling to prevent overheating. Apple does not necessarily need to stop selling x86-Macs when that happens.
Windows10 on ARM has been around for roughly 3 years, and in that time, Microsoft hasn't bothered to ship an ARM-native version, of what was certainly once, but I'd imagine still, considered one of it's "cash cows": Office.
If Microsoft can go 3 years after release (and let's not forget they'd had Windows RT which was ARM only, for 6 years already at that point) without shipping a native version of Office, why would you ever expect them to?
So if the platform developer isn't going to buy into native Apps, why would any 3rd party developer do it? And if no one buys into making native Apps, and they all have to run on an emulated x86 processor anyway, what the **** is the point?
I doubt it. The iBook and PowerBook became the MacBook and MacBook Pro, because previously the "Power" referred to the architecture of the CPU - PowerPC.
Right, because when Apple made it clear that Cocoa was the future of the platform, all development efforts shifted to making versions of software using Cocoa.. And then when Apple made it clear that 64bit was the future of the platform, all development efforts shifted to making versions of software that are 64bit.... Oh wait.
Apps written using Xcode using the platform frameworks Apple provides, are likely pretty simple to re-compile either as Arm only, or providing two binaries, or fat binaries as you mention. Thats certainly true.
But Microsoft Office isn't Arm native, nearly a decade after their first ARM based Windows release. Adobe has a horrendous track record of keeping it's Apps ahead of platform changes.
That only works if the existing App was built for iPad. It isn't a two-way street.
How? This (or versions of it) are oft mentioned but never explained. What "integration" benefit does an Arm based macOS provide, over the status quo?
For reference:
- iPadOS and iOS apps literally compile and run on x86 right now, inside the Simulator. So any code written for iOS/iPadOS devices, due to the OS' limited nature of third party Apps, is inherently safe to run on an x86 CPU.
- Catalyst allows developers of iPadOS (on ARM) apps to produce a macOS (on x86) app from the same codebase. It's not zero effort, it's not magic, but the problems are of polish in the UI/Interaction, not compiling code for x86 or for Arm.
- Existing iPadOS/iOS Apps built for Arm (and with no Catalyst involvement) will not work on macOS, regardless of the CPU used. If it were purely a CPU arch issue, you could build an iOS app for x86 (as happens whenever using the iPhone/iPad simulator), extract the app package from the simulator and just run it. Hint: you can't.
So, I ask again. What benefit does an Arm based macOS offer for "integration"? I of course expect zero answer because no one ever has an answer when pushed, because apparently the majority of people posting "theories" here have zero clue about the difference between a CPU architecture and a software framework.
You mean, the exact thing that Apple has said they have zero intention of doing, for ****ing years?
I don't know. Disinfromation implies mal-intent. Mostly I see people posting opinions full of technical terms that have zero basis in reality. I don't think it's deliberate, but it shows a clear lack of understanding of the subject matter.
No, we do know. The logic is not hard. If Apple drops Intel/compatible processors for ARM, Bootcamp as you know it, will be affected very heavily, if it even continues to exist.
I agree, mostly. Given the recent uptick in developer/pro-focused changes (decidedly more dev-friendly Mac mini, iMac Pro, new Mac Pro, new MacBook
Pro getting thicker!) I don't see any way Apple is going to give all those developers/pros the finger and drop x86 compatibility, but if they have *some* Macs with an Arm processor in it, I'd imagine they all will - and the Pro lines will have both. There's zero chance Apple is going to want a "Pro" machine to not be able to run an app that can run on the "consumer" Mac.
(a) That has literally nothing to do with the use of Arm or Intel. At all.
(b) Plenty of existing Mac software relies on a server-side component and the local app is merely a "thin client".
(c) The concept of thin-client software has been in use for decades. One of the most well-known examples, is called the "World Wide Web". Maybe you've heard of it. You're using it right now.
You're asking questions that are far too logical for this discussion.
I mostly agree with this - specifically the part I bolded. My work has 0% to do with Apple or Macs specifically, but I (currently) choose to use Macs because I find it the most productive with the least issues. Obviously others will have different experiences, that part is highly subjective.
But as you say, that doesn't mean I'm going to make my business reliant on Macs, or Windows PCs, or anything that's provided by a single company really. In terms of software, Open Source is a huge blessing for business - and in recent years there's usually several reasonable options for "compatible" hardware from vendors if that's the route you end up taking.
Forever? Nothing is forever. I do expect Apple to support existing users when there's little apparent advantage to dropping compatibility.
Literally the only "positive" I've seen anyone mention in this thread that bares any semblance of reality, is potentially lower power usage. Heck, half the "pro ARM" posts here are chanting "it will make Macs compatible with iOS" with zero ****ing idea what they're talking about.
.... and did I suggest that Apple should keep every API and Framework in existence around for use? No, I did not. I specifically said I agree with breaking compatibility when it has a purpose - i.e. removing a **** load of 32bit frameworks and libraries.
That doesn't make any sense. No one would "lose their mind" as you put it, because of continued Intel compatibility.
Given that this is about a custom Apple CPU, nothing is guaranteed. Virtualisation on ARM CPUs, is an extension like on any other CPU. It's not some fundamental key part of the design. That means if Apple think it isn't important, they can not implement those extensions. I'd imagine the A-series chips in iPads and iPhones already skip these extensions.
... So because people use a feature that has been available for 12 years, and they express discontent at rumours of a change that would render such a feature would be unavailable, "its your own ****ing problem"? What a very selfish view you hold.
You might want to tell Apple that, they've apparently spent over a decade supporting a feature that @Zdigital2015 says "you don't buy a Mac for this".
Then don't boot Windows. I don't either. I ****ing hate windows. Doesn't mean I can't recognise that some people find it useful. Also doesn't mean I use x86 VMs on a daily basis for work.
Great translation. I'm sure that's what everyone really means.
Plenty in this thread have implied literally that.
There is nothing stopping them from doing that now if they wished - they supported Universal Binaries (aka fat binaries) during the PPC to Intel transition.
Have you read anything I've written? Nothing about an Arm or Intel CPU precludes Apple from writing Xcode for iPadOS. You can imagine it all you ****ing want, but the CPU is 0.01% of the issue here.
You already can use an iPad with a monitor.
Wow how on earth did they manage that, Macs and iPads use different CPUs. This literally proves my point that CPU architecture is not a key issue in any of the "integration" points you or others keep suggesting an ARM based Mac would somehow "solve".
.... Where did I say any of that?
How much goodwill does Apple start with, for CPUs? We're so far at one "new in name only" CPU release from Apple. How many more like this before they have no goodwill left either?
When did I say it's the be all and end all? I said it's an important factor. I also said I'm generally in favour of breaking compatibility when there are clear advantages. So far I've yet to see many actual advantages suggested. For the record: claiming that it will solve an issue that doesn't exist, is not an advantage.
Personally I've never really found "clock speed" to be that indicative of performance alone.
My 2018 MacBook Pro has literally the same clock speed (2.4Ghz) as my 2011 MacBook Pro. Guess what? One's a **** load faster than the other.
So if this rumour turns out to be just that, a rumour and nothing more, you won't be buying any more Intel based Macs I presume?
I don't have any insider knowledge of this, but my understanding is that the naming change was primarily because of how the Pro line was named. PPC models used "Power" in the name - PowerMac, PowerBook, and they wanted the new names to reflect the name Mac.
They just pretty much did. Just running on a different OS.
They did use the "Power" moniker (as in "PowerBook") long before PowerPC architecture (68000 PowerBooks).
In my time I've known literally one developer who was "happy" to use a MacBook Air for development, and he didn't even have to compile code.
Claiming that an iPad Pro is almost as fast as Apple's lowest performance Mac is hardly selling the case that an iPad Pro can replace developer's Macs.
This is not how it works !
state of the art ARM processors are not more powerful, benchmark are not representative, especially Antutu, which is not at all a scientific benchmark
Not really, it depends on how Apple able to optimize the ARM-Based Macs.
Golden Gate would be a fantastic macOS name for starting the transition to Arm.
Office for iOS has some feature limitations not because of inherent issues with the ARM architecture (though there might be a feature or two they simply haven't ported yet), but because of user interface paradigm differences between iOS and macOS. There is no reason to believe that a hypothetical Office for macOS on ARM would be particularly limited.
It's unbelievable, isn't it? I was pulling the 7W figure from a rough calculation in my mind, but I found a really detailed review where they actually measured it.
Let's review together the data they gathered measuring the actual power consumption (you can do this if you have a USB charger that tells you how much power is being drawn at any given time):
To accurately assess the power consumption of the Apple A12X we need to look at three data points.
The Standby consumption (0.23W) gives us the estimate of how much power is required to keep every component in the iPad running at their lowest power consumption while the screen is fully turned off. Most of this power is probably used by the A12X at idle.
The idle average consumption of the iPad Pro (9.35W) gives us a good estimate about the power consumed by the sum of all its components when the CPU is doing nothing. Most of it is used to power the big 12.9" screen and its backlight.
The load Maximum consumption (14.96W) gives us the maximum ever consumption recorded on the iPad Pro while running the CPU at full load.
Given these 3 data points, you can calculate the amount of power required by A12X on full load with the simple the formula:
standby + (load maximum - idle average) = 0.23W + (14.96W - 9.35W) = 5.84W !!
So, there you go. I was probably even overestimating a little bit with my 7W figure by doing math in my head.
For the people that still do not believe the math is right: you can compare the results of the Microsoft Surface Pro 6 in the same chart that uses a Core i5-8250U Intel CPU. That CPU has a TDP configurable up to 25W.
When using the same formula on the Surface Pro you get 36W - 8.7W = 27W, which is about the same 25W TDP we were looking for.
Last edited:
There is no denying that, at similar TDPs, Apple ARM chips significantly outperform Intel's Skylake architecture, and still slightly outperform Ice Lake.
The open question is how well that scales. If Apple were to do a 45W A13M for a 16-inch MacBook Pro, it would probably destroy Comet Lake-H (and Intel isn't doing an Ice Lake-derived H chip for another ~two years). But we don't really know.
[automerge]1585395801[/automerge]
I'm not sure what "amount of power required on full load" is supposed to mean here. The CPU almost certainly doesn't draw zero power on idle, so if anything, I would surmise from that data that the TDP is above 7W.
I'm interested to see what happens with iPad office software now there's the possibility of using a cursor, I would hope it will be updated to use a more traditional, full interface and feature set when a mouse and keyboard is connected!
No it’s false ! At the same TDP, ARM based chips outperformed Skylake based chips in a mobile OS with short bust specific task. Skylake runs on computer Os, with way more background services and with there large instruction sets, they are able to perform way more specific tasks
It’s like running a benchmark on an I3 with CLI Environnement vs I7 in Windows with Skyrim running in the background and conclude that I3 are more powerful than I7
I get your point. I'm editing the post by adding the 0.23W consumption of the iPad in standby. This brings the total to 5.84W.
Last edited: