Appreciate the info. So it doesn’t seem like there are as many roadblocks as many would think. Challenges for sure, but I am optimistic.
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.
Apple Plans to Ditch Intel and Use Custom Mac Chips Starting in 2020
- Thread starter MacRumors
- Start date
- Sort by reaction score
Appreciate the info. So it doesn’t seem like there are as many roadblocks as many would think. Challenges for sure, but I am optimistic.
sorry for taking so long, work and all'
many DB engines don't store data as simple plain text, and when you're using it in binary formats (backups for example), how that file is created is specific to the platform and OS it is created on.
the one that I'm dealing with daily is called Progress Openedge. DB backups are not portable between CPU types, nor even OS's. you cannot take a simple backup from a Linux and restore it on windows. Heck, you can'ttake 32bit linux DB backup and restore it to 64bit linux.
this isn't the only DB engine like this. it has to do with the ways that data is written by the engine. in my work, I don't particular need to know the engineering behind it, But I spent the last 7 years supporting a few hundred clients on this platform. Almost all of them major financial institutions.
A switch to ARM has a direct impact on what people do both professionally and personally. It's not about ARM, a switch to new technology, or Apple trying to make more money. There is nothing "old" in trying to use the best tool available to meet your objectives.. that is until those tools no longer work for your purpose. If the swap to ARM works for you, great. It won't work for a bunch of people (no idea how many).
If all you do is run applications on the iPad, this could be a nice swap.... that assumes people keep developing those apps. It will be interesting to see the impact on their Mac sales as well as the requisite impact on those building iPhone and iPad applications.
We'll just have to see how successful they are developing a laptop/desktop chip and what products they move to ARM and what they don't. I think it's much more likely only their low end Mac would get converted over for a number of years. It's a lot harder, costs a lot more money, and is more time consuming than most think to develop the CPU and supporting chipsets.
Apple still has to prove they can compete on the desktop. Benchmarks on a phone are not equivalent to benchmarks on a desktop.
Was thinking about how Apple could actually do this. Would be interesting if they used many different chips to make up a single processor. A T2 chip for various chipset things, an A11 for Wifi, bluetooth, basic OS functions, and a 15watt processor of just cores, and a GPU. That would spread out the heat and mean higher yields (cheaper chips) since each is smaller. They can then just scale up the power by adding more of the 15w chips. A macbook pro 15 may be 3 for 45w total, while an imac might get 6.
don't apple already make their own chips for their iphones and ipads, this rumors is just for Macs, could this be the end of hackintosh?, many hackintosh users buy apps from the apps store, so I don't know if Tim wants to throw that money away, the day that I won't be able to build a hackintosh then I buy a real Mac but the problem is customization and of course the price, I don't mind paying the money but give me new hardware that is worth the money that I'm paying, don't steal my money with old obsolete hardware, just look at the pre historic Mac Pro 6.1 from 2013 and how much apple still ask for it, so much for 5 years obsolete rule from apple, hello Tim is 2018, where is the new Mac Pro 7.1 , oh yeah iPhone X SE version 2 , iPhone X S plus, give me a break
[doublepost=1522805797][/doublepost]this will eventually might happen one day but I'll be hearing this same old story for quite some time
so if it happens , happens
but until then I will not believe a word
the funny thing is that I hear this every April 1 every year
so I guess you already know what that means
[doublepost=1522805797][/doublepost]this will eventually might happen one day but I'll be hearing this same old story for quite some time
so if it happens , happens
but until then I will not believe a word
the funny thing is that I hear this every April 1 every year
so I guess you already know what that means
Apple uses enterprise servers like every other corporation. They are not running macOS Server.
Apple is "The iPhone Company". It makes sense for them to ditch macOS altogether and focus on iOS. I expect that within a short time, Apple will offer macOS on only 2 products... a high-end notebook and high-end desktop. Everything else for consumers will be iOS-based.
Just look at how many folk have been able to switch from a macbook to an iPad as their primary computer. If their use cases haven't changed, then it shows that they "over-bought" by buying a macbook. They are able to do everything they were doing before (on the macbook), but now on a much more restrictive platform (ipad).
My iMac and 11" Macbook Air are the 2 best personal computers that I've owned over the 40+ years I've owned personal computers. But I will not follow Apple over to ARM-land should that happen... and if it happens, I know that I'll be in the small minority who will not follow. I'm part of a handful of people who actually dislike a notch on a smartphone's screen. I am that odd.
iCloud is running on Google's cloud hosting, which is unix (dunno what variant) based.
You’re right. An i9 core run in the thermal envelope of an iPhone 8 would get tromped by an A11.
Who knows what Apple could do with an newer ARM core circuit designed for a 100W heat sink.
So they are running linux?
[doublepost=1522811681][/doublepost]Most people are concerned if their favorite App makers Adobe would recompile their apps for the new platform? I say they would, because they are not doing a service for the Mac market, they too love to use Mac products in their respective companies. Most creative professionals love Apple products and use them extensively. Microsoft makes their products too for the same reason it makes sense for them to sell on Mac. Apart from that Microsoft has already shown how to make Windows apps work on ARM processors with little to no penalty.
Exactly.
I don't know why people are expecting Apple to put a smartphone processor into a laptop or desktop.
What I gathered from this announcement is that Apple will expand their chipmaking skills beyond phones and tablets... and will create new chips that are appropriate for higher-wattage Mac laptops and desktops.
That's just silly; things simply don't scale that way.
A Saturn V rocket would fail catastrophically strapped to the back of a bicycle. Who knows what NASA could do getting people to the moon on pedal power. Makes as much sense as your comment anyway.
[doublepost=1522813550][/doublepost]
Maybe because Timmy has been saying for years his gimpy little iPhone CPUs offer "desktop" CPU and GPU performance.
That smartphone processor **is** Apple's idea of desktop power.
I sensed this with iPhone X. For my surprise people are paying price for a computer to get a clumsy and heavy design with UX flaws and pseudo security Face Id, and consider it a good investment. Obviously Apple marketing department is very good. Young millennials are spoiled vanity cult and addicted to fashion, they don't care for the price, even in my country with low general income they are willing to give everything only to be seen successful in the eyes of others. This is the primary target now. Second target are people with low computing needs that share the same deep emotional need. This will be the final shift to mobile only Apple. Now the move to cut off Mac OS department makes sense. So for semi pro users (designer, musician, etc. wannabes) Apple will have some iMac Pros and dumbed down products. But high end pros will be going to Linux and Windows in a heartbeat. This is the end of an inspiring era in witch intelligence, aesthetics and user experience was made by smart people for smart people. RIP.
Apple is "The iPhone Company". It makes sense for them to ditch macOS altogether and focus on iOS. I expect that within a short time, Apple will offer macOS on only 2 products... a high-end notebook and high-end desktop. Everything else for consumers will be iOS-based.
Just look at how many folk have been able to switch from a macbook to an iPad as their primary computer. If their use cases haven't changed, then it shows that they "over-bought" by buying a macbook. They are able to do everything they were doing before (on the macbook), but now on a much more restrictive platform (ipad).
My iMac and 11" Macbook Air are the 2 best personal computers that I've owned over the 40+ years I've owned personal computers. But I will not follow Apple over to ARM-land should that happen... and if it happens, I know that I'll be in the small minority who will not follow. I'm part of a handful of people who actually dislike a notch on a smartphone's screen. I am that odd.
That is probably their direction. I also believe ultimately it will be detrimental to their business. Their goal is to sell as many devices for as much as possible and become the high quality chrome book company.... ala ipod and iphone. My guess is that with the change to ARM (if they fully convert) will result in a reduction in apps and business.
I'm just hoping they develop chips for their low end ipad/mini-macbook type devices and leave the power required for development on x86. It will be years before they have a competitive desktop chip on ARM (more than 2 anyway).
A $35 Raspberry PI runs linux on a quad core ARM CPU designed for the smartphone market.
With the Apple tax that should bring the cost for a new Mac up to about $50. That would still be overpriced for an ARM-based Mac considering how limited it will be.
I guess the people high on Kool-Aid believe the $20 Ax chip drawing 5 Watts in their iToys is comparable to a $400 Intel CPU plus a $400 NVidia GPU collectively drawing 400 Watts. The only difference between the systems are price gouging and inefficient power use, right? And when Apple replaces the expensive CPU with a cheap smartphone chip, they'll pass the savings on to the customer, right?
With the Apple tax that should bring the cost for a new Mac up to about $50. That would still be overpriced for an ARM-based Mac considering how limited it will be.
I guess the people high on Kool-Aid believe the $20 Ax chip drawing 5 Watts in their iToys is comparable to a $400 Intel CPU plus a $400 NVidia GPU collectively drawing 400 Watts. The only difference between the systems are price gouging and inefficient power use, right? And when Apple replaces the expensive CPU with a cheap smartphone chip, they'll pass the savings on to the customer, right?
The problem with this train of thought is this: PC hardware is much cheaper for equal or better performance. I work in Hollywood and I'm beginning to see a great deal of previously Mac-only shops migrating over to PCs because the current Apple hardware is just NOT fast enough. This is not an opinion; it's been actively happening for about 2 years now, and it's accelerating.
To address your post (and others') further, ARM compatibility is still NOT x86 (or x86_64) compatibility. Even if Windows ran well on ARM, no programs would, as they're not compiled for ARM. Unfortunately, there's also an annoying habit software and hardware companies have where they'll provide drivers for only the latest stuff they make, leaving your mouse (I'm looking at you, Logitech), sound card (Creative Labs), and so on without full support (this happened on Windows when they moved from XP to Vista and completely changed the way the OS handled drivers).
And for those harping about how great the move from PPC to intel was (and it was), and how this will be awesome again, keep in mind that Apple reached critical mass with the Mac by, arguably 2010 (definitely by 2013 -- half a decade ago). When they moved from PPC, all of 8 people owned Macs -- they're now a household name and Apple's in a position to piss off a lot of their users.
Sadly, the opposite viewpoint could bite them, as well -- macOS comprises less than 10% of all PCs... I'm wondering just how much of an incentive most companies (especially dual-OS, like Adobe) will have to re-write their code. Worse, what if they simply 'port' the software over? It'll work, but it'll work like crap (it reminds me of 'ported' software back in the Amiga days).
As an ex-DEC guy, I think I remember that Intel ended up licensing some of the Alpha architecture. And didn't they have a lawsuit of some kind first? Where DEC claimed that Pentium was infringing? I think the Alpha lives on in China, too. Aidenshaw would know for sure.
I can see developers building to some kind of byte code - abstracted from the underlying OS, and Apple supporting building and deploying bites on each different platforms (iOS, macOS, tvOS, watchOS) via the store. Apple can do that in a 'closed' ecosystem - where all software is installed via an AppStore - but the Mac App store is pretty crappy and developers/users will balk - unless the trade off isn't bad. I'd imagine their direction will become more transparent at WWDC.
If they really want to show that an iPad or armPad can do everything - they need a version of XCode that runs on the iPad, allows a developer to build an app, run it, and submit it to the App Store.
They'd also need (IMO) a terminal as good as the mac - where you can configure your environment (shell, tmux, emacs/vim/neovim, etc) to be as robust as the current environment.
Seems like a piece of cake.
Last edited:
With the background you mentioned in another post, you should know that RISC does not mean that for example a Subtract instruction is removed (reduced) because there are Add and Negate instructions. Instead, RISC means a Load/Store architecture with a large number of registers. Memory operands can only be loaded to or stored from registers. Even immediate operands don't exist except for small constants baked into instructions. RISC means that instructions don't have to be split into pieces for the scheduler.
x86 is a 1970's architecture. The 8086/8088 were introduced in 1978. The architecture was heavily influenced by 1973's 8080 - it was designed so that 8080 assembly code could be 99% machine translated to 8086 code. The 8080 itself is an enhancement of the earlier 8008.
You're wrong. It will be great for Apple's business because then they will only have to support one type of CPU rather than two and it will be great for developers in the long run because they will only have to write and support applications for essentially one platform. It just doesn't make sense for Apple to continue to support a platform with a dedicated team whose revenue is dropping every year. The revenue from sales of iOS devices is more than 10× that of Macs in Q1 2018.
I'm not explaining myself well, but your comment is proving my point.
Why should Apple's business model be sympathetic to their competitors? If Windows is so important to your work, then you'd either need two laptops or to ditch OS X. OR ditch Windows and do all your work in OS X.
But by building their own chips, Apple could take the Mac in a different direction - one that integrates even better into the Apple ecosystem (which is what Apple is all about after all).
I don't know too much about computers, but comparing this situation to smartphones, no-one complains that they can't boot their phone into both iOS and Android. They have two phones if having both OSs is important to them.
I'm not a big Apple fan, and since Jobs died I feel Apple have been following the crowd. But I like this idea - Apple are leading again. They're not worrying about how this will impact people but focusing on all the benefits of having custom chips in their Macs, knowing that people will scream and moan, but ultimately they'll adapt to the change and prefer it.
In the same way that Apple dropped CD drives and persuaded users to find another way, rather than waiting for a new technology to become adopted and then removing CD drives.
Can you imagine if Apple were still using PPCs...?