One thing is for sure, Apple's move to their own silicon is going to make Mac hardware updates a hell of a lot more interesting. When Apple is dependent on Intel there's only so much they can do to differentiate their product hardware from competing products that were launched in the same time frame. This always took a little wind out of Apple's sails and led to component-level comparisons that obscured the substantial system-level and ecosystem-level benefits that the Mac has always had over competing products.
The other thing I'd like to add is: don't for a second get lulled into believing that Apple's big silicon move is going to be all halcyon days of sunshine going forward. Intel, AMD, and emerging players who've not made their presence known yet are not going to roll over and let Apple get out ahead of the pack. They are going to fight tooth and nail to make Apple doubt its decision. If Apple can't deliver the kind of performance needed from their homegrown solutions they'll be pummeled in the market.
Yes, Apple is now being throttled by Intel, but so are Apple's competitors. There is nothing that points to this being a permanent state of affairs. Once Apple separates itself from Intel they won't be able to count on everyone else suffering with the same issues they have related to Intel. If Intel arrives at a huge breakthrough, Apple will be on the outside looking in.
The point is that we have to be aware that this is a risky move by Apple and they will have to prove to their customers for years to come that this was the right long-term strategy. Right now it seems brilliant, but sustaining that shine will take a hell of a lot of work. We just need to be realistic, temper the unbridled enthusiasm that some of us feel today, and be prepared for a long war of attrition against very capable adversaries. Nothing is a given and nothing will come easy. Personally, I see no other company being better positioned to meet the kind of challenges that are ahead for Apple.
Really, what has set Apple apart from the Windows World is its software and ecosystem, not the hardware.
I seriously doubt that the move to "Apple Silicon" will change that.
At most it will enable them to better customize the hardware for their own needs. But regardless, like the iPhone, the major differentiator will be the software and ecosystem -- and now, for the Mac, they will be better able to take advantage of those things (such as, for example, opening up iOS apps to Mac).
But, that said, I agree that their presentations will gain additional WOW! factor.
That's the biggest reason Apple is transitioning the Mac to its own custom silicon. Jony Srouji mentioned it specifically in his portion of the keynote. Along with their SoC's offering "industry leading'" performance per watt.
Unlike what the article says, Apple didn’t say what the machines they were using for their demos had in the way of RAM or storage. They did say how much would be in the developer machines. And I would also like to point out that the iPad Pro has 6GB RAM, not 16GB, so the developer machines are not outfitted the same. That’s not surprising, because the Mac needs more resources than iOS devices.
Good point; but since they obviously had this transition in mind when they did the A12Z, I submit they designed an External memory bus into the SoC, allowing the internal RAM and Flash to be supplanted/replaced with external resources.
Many Microcontrollers have supported such an expansion bus for decades; in fact, before EPROM and flash-based microcontrollers were the norm, it was the only way to do a “Development Version” of most microcontrollers.
Afterall, when you’ve got a BGA package with a few hundred pads, what’s another 96 or so for an expansion bus?
Are you stating that the A12Z, and by extension, all Apple SoC’s have their RAM on chip, when you say internal? Because that’s not true.
Linux is supported, why not Windows in VM? That's an ARM distribution of Linux. Docker and MONO also have ARM distributions.
Remember, VIrtualization (VirtualBox, Parallels, etc.) and compatibility layers (WINE) require that the underlying binaries run directly on the CPU. WINE exists for ARM now, but that's not going to run Intel binaries.
For PowerPC to Intel, Steve Jobs put up a chart showing the increase in performance per watt making it clear it was a good business decision. They're saying this is also the motivation here. But, I want to see real-world benchmarks.
Back then, there were the commercial ppc chips apple was using, and the new chips Intel were selling to Apple. Those were not out yet (the Yonah), but Apple had to make a really good case for it because of the put-downs they had going for several years.
this is different. Most Mac users are happy with this decision. Apple likely isn’t ready to give any info about what the new chips can do. Since they’re not going to sell these chips, there’s no reason to give out specs now to get OEM’s ready.
but, considering the performance that we saw in the demo/s using the A12Z, which is now an old chip, you can bet that the Mac versions, and yes, they said that this would be a line of chips designed for the Mac, will be MUCH better. Consider that the A13, out late last year is better than the A12Z in integer, and the new A14 out later this year, will be better. Same thing for the graphics cores. And they won’t be the Mac chip line.
One thing is for sure, Apple's move to their own silicon is going to make Mac hardware updates a hell of a lot more interesting. When Apple is dependent on Intel there's only so much they can do to differentiate their product hardware from competing products that were launched in the same time frame. This always took a little wind out of Apple's sails and led to component-level comparisons that obscured the substantial system-level and ecosystem-level benefits that the Mac has always had over competing products.
The other thing I'd like to add is: don't for a second get lulled into believing that Apple's big silicon move is going to be all halcyon days of sunshine going forward. Intel, AMD, and emerging players who've not made their presence known yet are not going to roll over and let Apple get out ahead of the pack. They are going to fight tooth and nail to make Apple doubt its decision. If Apple can't deliver the kind of performance needed from their homegrown solutions they'll be pummeled in the market.
Yes, Apple is now being throttled by Intel, but so are Apple's competitors. There is nothing that points to this being a permanent state of affairs. Once Apple separates itself from Intel they won't be able to count on everyone else suffering with the same issues they have related to Intel. If Intel arrives at a huge breakthrough, Apple will be on the outside looking in.
The point is that we have to be aware that this is a risky move by Apple and they will have to prove to their customers for years to come that this was the right long-term strategy. Right now it seems brilliant, but sustaining that shine will take a hell of a lot of work. We just need to be realistic, temper the unbridled enthusiasm that some of us feel today, and be prepared for a long war of attrition against very capable adversaries. Nothing is a given and nothing will come easy. Personally, I see no other company being better positioned to meet the kind of challenges that are ahead for Apple.
Really, what has set Apple apart from the Windows World is its software and ecosystem, not the hardware.
I seriously doubt that the move to "Apple Silicon" will change that.
At most it will enable them to better customize the hardware for their own needs. But regardless, like the iPhone, the major differentiator will be the software and ecosystem -- and now, for the Mac, they will be better able to take advantage of those things (such as, for example, opening up iOS apps to Mac).
But, that said, I agree that their presentations will gain additional WOW! factor.
That's the biggest reason Apple is transitioning the Mac to its own custom silicon. Jony Srouji mentioned it specifically in his portion of the keynote. Along with their SoC's offering "industry leading'" performance per watt.
Actually no need for a Mac, whatever processor is planted in, but the iPad battery life is too short and not user replaceable. Why Apple doesn't design the iPhone and iPad to have the battery swap possible at user level?
Nobody is doing that anymore. The battery life is not too short, though longer is always better.
swappable batteries are, by design necessity, smaller, and give less battery life, unless you want a thick, heavy battery, which almost no one wants.
Your points don't really have anything to do with anything.
It should be possible to disagree without being rude, and I'm not even sure we do disagree. Well, on style we do.
Apple's drop in the K-12 market is because they chose not to develop a machine cheap enough to keep their share at 60%. ...
There really is no point bleating on about market share, because for Apple it's of secondary concern when up against profit.
You're saying that after 40 years of Apple in Education then schools decided that "cheap" was the only thing. Do you really believe the focus on cost+value is new in K-12? It has been like that forever and Apple did profit from K-12 - not on margins but on volume. When it comes to value then K-12 is the canary in the coal mine.
Jobs once asked a CEO: "Do you want to sell sugar water for the rest of your life, or do you want to come with me and change the world?". Tim Cook is thrilled with selling sugar water like animojis, widgets, and UI skins. He is not moving us forward on how we work or gather/structure/validate our knowledge.
I was involved with education here in NYC for many years, being on the technology committee that made these decisions, abet as an advisor. Yes, school systems tend to go to “good enough” if it’s cheap enough. Chromebooks were half the price of an iPad, and included a keyboard. Simplistic, yes, but again, cheap. Apple countered with the new, bigger $329 iPad. But without a keyboard at that price, it came out to over 50% more. School districts have been more short of cash than every before. First the Bush recession was disastrous to school budgets everywhere. But things were looking better for a while, until congress turned to the dark, I mean, Republican side, and the administration couldn’t get past their hatred of federal education support. And then, we got Trump.
apple has actually clawed back some k-12 marketshare. We’ll have to see how far that goes.
I remain skeptical. Not because I don't think Apple can pull this off. They've done it 2x before so clearly they can. Consumers probably won't notice any difference other than better battery life, etc. Will large corporations buy any laptop that doesn't have an x86 compatible processor? Clearly they buy lots of iPhones and iPads so perhaps. Fortunately for Apple the Mac is a small portion of their overall revenue so the risk is relatively small. Especially compared to their last major transition when Mac was the majority of their revenue. This is a multi-year adventure so it will be interesting to follow Apple's progress.
I think we're at the point where for "desktop class" systems corporations aren't going to give a damn.
And they've done this more than twice before. Remember, what we call Apple and macOS today is really NeXT and NeXTSTEP rebranded. So, they started out on Motorola silicon, then switched to Intel (first instance of Fat Binaries aka Universal Apps). [We'll now skip a few niche architectures that NeXTSTEP also ran on.] After NeXT took over Apple, they moved to PowerPC and then, eventually, back to Intel. Each time the process was more seamless than the last, and this transition to Apple Silicon looks to be even more seamless for developers and users than the PowerPC to Intel transition was. They know how to do this.
I suspect the corporate acceptance will take a lot of time. It's already difficult convincing companies to consider an AMD processor rather than Intel. This is several degrees beyond that.
I’m not so sure about that. Mac use in corporations has been rising at a good pace the past few years. Almost all software they use is Mac, not Windows. These days are over.
with Craig stating that all mac x86 software will run, making a point that plug-ins would work too, a point of contention during the older moves, I don’t see a lot of problems for corporate use. These machines also offer better security, always a sore point.
Given that big organizations generally give at least a year before moving to a new version of Windows, I see about the same delay here. Not a problem.
Apple hit problems with 68K Apple hit problems with PowerPC Apple hit problems with Intel
At some point you have realise that only you care enough about what you’re trying to do to make sure you have all the pieces in place to do it.
Apple failing to deliver more than 2 updates to Mac Pro in 10 years this isn't due to silicon. The problems Apple hit on the Mac - like going from 60% to 20% market share in K-12 (mac+iPad) were not due to silicon. Mac shipments dropped 21% from 2019 to 2020 - with all other vendors on x86 it simply isn't about the silicon.
Your points don't really have anything to do with anything. Apple's drop in the K-12 market is because they chose not to develop a machine cheap enough to keep their share at 60%. What they did was build a machine that allowed them to keep the most profitable 20%. Probably because they realised that the machine kids use at school doesn't seem to influence the machine they eventually use when they're in college or go to work.
There really is no point bleating on about market share, because for Apple it's of secondary concern when up against profit. So with that in mind, let's look at the problem they're trying to actually solve (their desktops and laptops running on a stagnating architecture), and what does the switch give them:
A machine that's cheaper to build. An ecosystem that is much simpler and cheaper to build software for, which benefits Apple's developers as well as Apple Machines that are optimised for the operating systems and software they're running. Machines that can be recycled and stripped down to power lower-end machines and devices later on.
I remember, years ago, when Apple bought PA-Semi, and we all thought that they were going to build their own PowerPC chips. The move to Intel was seen with much the same hysteria we're seeing from a few people here now. But we can see now that Intel was only ever a stop-gap. Complete control and integration was always the plan. I suspect that Apple's Silicon engineers knew precisely when Intel was going to hit the wall, and Apple's management knew that the company would do little about it because they have a near monopoly, so there was no real incentive for them to try harder.
I disagree with that statement. Apple's drop in the K-12 to market had little to do with Apple's product pricing and a lot to do with them not offering a software / service infrastructure that made the management and administration of those devices easy for teachers and IT staff. Apple wasn't beaten by cheap Chromebooks but by Google's well done Google Classroom cloud service.
That’s not true. There was a great amount of service infrastructure support. There was, and is still almost none for Chromebooks, as well as a dearth of professionally created, and approved curriculum standard software for them, whereas there were loads for the Mac, and later, for the iPad.
At the moment, this is little more than a loss of prestige for Intel.
The problem will come if Microsoft decides to put some effort into its own ARM strategy.
Apple doesn't have close to the silicon team that Apple has to pull this off. What we saw yesterday was how central Apple's silicon team is to Apple's competitive advantage. MS would have to buy a major silicon company to pull this off. Who knows, maybe they'll buy AMD (*sarcasm*).
Microsoft can participate in ARM's Cortex X program, or work with a company that is.
The Cortex X1 and next year's ARMv9 Matterhorn core (Cortex X2?) are being designed without previous constraints that are on cores like the Cortex A78.
The Cortex-X1 here is projected to use 1.5x the power of an A78. This might end up slightly lower but I’m being overly cautious here and prefer to be on the more pessimistic side. Here’s the real kicker though: the X1 could very well use up to 2x the power of a Cortex-A77/A78 and it would still be able to compete with Apple’s cores in terms of energy efficiency – the core’s increased performance largely makes up for its increased power draw, meaning its energy efficiency at the projected power would roughly only be 23% worse than an A78, and only 11-14% worse than say a current generation Snapdragon 865. Arm has such a big leeway in power efficiency at the moment that I just don’t see any scenario where the X1 would end up disappointing.
...
Meanwhile the Cortex-X1 is a big change for Arm. And that change has less to do with the technology of the cores, and more with the business decisions that it now opens up for the company, although both are intertwined. For years many people were wondering why the company didn't design a core that could more closely compete with what Apple had built. In my view, one of the reasons for that was that Arm has always been constrained by the need to create a “one core fits all” design that could fit all of their customers’ needs – and not just the few flagship SoC designs.
The Cortex-X program here effectively unshackles Arm from these business limitations, and it allows the company to provide the best of both worlds. As a result, the A78 continues the company’s bread & butter design philosophy of power-performance-area leadership, whilst the X1 and its successors can now aim for the stars in terms of performance, without such strict area usage or power consumption limitations.
In this regard, the X1 seems really, really impressive. The 30% IPC improvement over the A77 is astounding and not something I had expected from the company this generation. The company has been incessantly beating the drum of their annual projected 20-25% improvements in performance – a pace which is currently well beyond what the competition has been able to achieve. These most recent projected performance figures are getting crazy close to the best that what we’ve seeing from the x86 players out there right now. That’s exciting for Arm, and should be worrying for the competition.
That quote makes no sense. The latest Qualcomm chip is well behind Apple’s chip of two years ago in performance and efficiency. If this will be 10% behind that, then it’s no competition to Apple at all. That’s according to Anandtech testing. The chart you see here shows, not an actual Cortex-X1 chip test, but an estimated number, which is indicated by the word—projection. That means nothing. It’s also an estimate of what it will be on 5nm. Useless.
At the moment, this is little more than a loss of prestige for Intel.
The problem will come if Microsoft decides to put some effort into its own ARM strategy.
Apple doesn't have close to the silicon team that Apple has to pull this off. What we saw yesterday was how central Apple's silicon team is to Apple's competitive advantage. MS would have to buy a major silicon company to pull this off. Who knows, maybe they'll buy AMD (*sarcasm*).
Microsoft can participate in ARM's Cortex X program, or work with a company that is...
Apple's A series processors aren't just about energy efficiency and flops. In large part it's about integrating all these other things,
into the processor and being able to customize it precisely to meet your needs. The graphic above represents years of R&D, customizing everything for exactly what Apple wants it to do. You can't have this overnight.
Of course it's more than that. You've just posted their image of what makes up their SoC. These components already exist, not only as ARM reference designs, but as components that have been worked on for years by various manufacturers.
Here's a block diagram from Qualcomm's Snapdragon 865 for example:
Even Microsoft has done custom silicon work alongside Qualcomm to fit their needs. In the HoloLens 2, Microsoft created a custom multiprocessor the HPU (Holographic Processing Unit) to accompany the Snapdragon 850. In the Surface Pro X, they branded the Microsoft SQ1 - going even further to take a Snapdragon 8cx and build onto it with a higher CPU clock, faster GPU (2.1 TFLOPS) and custom AI engine (9 TOPS).
I'm not sure what your point is, but Qualcomm SOC's aren't known to be as performant as Apple's, and more to the point, Qualcomm's SOC for Windows 10 ARM was not up to the hype. Qualcomm SOC's don't suck for smartphone use, but I don't see Qualcomm as going all in to a desktop class processor as Apple is doing either.
Qualcomm is just one company that Microsoft has worked with. A big part of the reason why Qualcomm's SoCs aren't competing with Apple is due to their use of medium sized CPU cores and a reliance on ARM's outdated small cores. The Cortex X program solves that for them. They finally have a big core on ARM's roadmap they can use. Qualcomm has been pushing towards the laptop space for a few years now - this was a big piece of the puzzle missing for them.
My point is that Qualcomm is going to remain behind Apple in SOC performance, both in mobile and in notebook/desktop. The fact that Qualcomm has an architectural license yet has waited for Cortex X is telling, all the while Apple is innovating in the SOC space with the ARM ISA license. Apple's advantage is that it can provide synergy between its OS, development system and SOC designs, and other hardware.
That's not definite. The market has only recently made it possible for other ARM solutions to compete with x86. Qualcomm would still have had to compete with Intel and AMD.
Cortex X program allows partners to create custom products working with ARM to fit there needs. These are not on the ARM roadmap and have no set release timeline.
Synergy works in some cases, but doesn't guarantee superiority in an open market - especially one where the major productivity ecosystem lives with Windows and on x86-64.
At the moment, this is little more than a loss of prestige for Intel.
The problem will come if Microsoft decides to put some effort into its own ARM strategy.
Apple doesn't have close to the silicon team that Apple has to pull this off. What we saw yesterday was how central Apple's silicon team is to Apple's competitive advantage. MS would have to buy a major silicon company to pull this off. Who knows, maybe they'll buy AMD (*sarcasm*).
Microsoft can participate in ARM's Cortex X program, or work with a company that is...
Apple's A series processors aren't just about energy efficiency and flops. In large part it's about integrating all these other things,
into the processor and being able to customize it precisely to meet your needs. The graphic above represents years of R&D, customizing everything for exactly what Apple wants it to do. You can't have this overnight.
Of course it's more than that. You've just posted their image of what makes up their SoC. These components already exist, not only as ARM reference designs, but as components that have been worked on for years by various manufacturers.
Here's a block diagram from Qualcomm's Snapdragon 865 for example:
Even Microsoft has done custom silicon work alongside Qualcomm to fit their needs. In the HoloLens 2, Microsoft created a custom multiprocessor the HPU (Holographic Processing Unit) to accompany the Snapdragon 850. In the Surface Pro X, they branded the Microsoft SQ1 - going even further to take a Snapdragon 8cx and build onto it with a higher CPU clock, faster GPU (2.1 TFLOPS) and custom AI engine (9 TOPS).
Compare the diagrams, and the actual capabilities, and I think you've simply reinforced my point.
No, you just need to do a better job in understanding what all of the components in the Qualcomm diagram actually do and their functions.
You seem to be missing the point of having custom logic that you’ve spent years optimizing for your own needs. Yes, all the parts of Apple’s SoC can be found elsewhere, but Apple has designed, and redesigned them to squeeze all the performance from them by being able to customize everything for EXACTLY what they need. As an example, their GPU’s directly support Metal calls, which removes a layer of overhead. They’re free to extend and customize ARM’s ISA as well, as there’s no need for their own silicon to be compatible with anything outside their own ecosystem.
This assumes Apple has the best parts, integration and ecosystem. That's not always the case. Other manufacturers have done considerable work on integration and the major productivity ecosystem is not with Apple. The point of the Cortex X program allows partners to have custom ARM solutions made for their need. This is not a roadmap product and is not something everyone has access to and isn't restricted to a product roadmap or the current ISA version being used by ARM's mainstream designs.
I remain skeptical. Not because I don't think Apple can pull this off. They've done it 2x before so clearly they can. Consumers probably won't notice any difference other than better battery life, etc. Will large corporations buy any laptop that doesn't have an x86 compatible processor? Clearly they buy lots of iPhones and iPads so perhaps. Fortunately for Apple the Mac is a small portion of their overall revenue so the risk is relatively small. Especially compared to their last major transition when Mac was the majority of their revenue. This is a multi-year adventure so it will be interesting to follow Apple's progress.
I think we're at the point where for "desktop class" systems corporations aren't going to give a damn.
And they've done this more than twice before. Remember, what we call Apple and macOS today is really NeXT and NeXTSTEP rebranded. So, they started out on Motorola silicon, then switched to Intel (first instance of Fat Binaries aka Universal Apps). [We'll now skip a few niche architectures that NeXTSTEP also ran on.] After NeXT took over Apple, they moved to PowerPC and then, eventually, back to Intel. Each time the process was more seamless than the last, and this transition to Apple Silicon looks to be even more seamless for developers and users than the PowerPC to Intel transition was. They know how to do this.
I suspect the corporate acceptance will take a lot of time. It's already difficult convincing companies to consider an AMD processor rather than Intel. This is several degrees beyond that.
I’m not so sure about that. Mac use in corporations has been rising at a good pace the past few years. Almost all software they use is Mac, not Windows. These days are over.
with Craig stating that all mac x86 software will run, making a point that plug-ins would work too, a point of contention during the older moves, I don’t see a lot of problems for corporate use. These machines also offer better security, always a sore point.
Given that big organizations generally give at least a year before moving to a new version of Windows, I see about the same delay here. Not a problem.
My experience in industry is the opposite: Almost all software was designed for Windows and implemented by a Windows centric IT department trained in Windows.
At my last IT position at a regional Red Cross center, only the CEO used Apple products and was thrilled to have somebody with even a basic understanding. The only reason she was able to use Apple products was because she was the boss. Everybody else was pushed into Windows and Android stuff.
At the moment, this is little more than a loss of prestige for Intel.
The problem will come if Microsoft decides to put some effort into its own ARM strategy.
Apple doesn't have close to the silicon team that Apple has to pull this off. What we saw yesterday was how central Apple's silicon team is to Apple's competitive advantage. MS would have to buy a major silicon company to pull this off. Who knows, maybe they'll buy AMD (*sarcasm*).
Microsoft can participate in ARM's Cortex X program, or work with a company that is.
The Cortex X1 and next year's ARMv9 Matterhorn core (Cortex X2?) are being designed without previous constraints that are on cores like the Cortex A78.
The Cortex-X1 here is projected to use 1.5x the power of an A78. This might end up slightly lower but I’m being overly cautious here and prefer to be on the more pessimistic side. Here’s the real kicker though: the X1 could very well use up to 2x the power of a Cortex-A77/A78 and it would still be able to compete with Apple’s cores in terms of energy efficiency – the core’s increased performance largely makes up for its increased power draw, meaning its energy efficiency at the projected power would roughly only be 23% worse than an A78, and only 11-14% worse than say a current generation Snapdragon 865. Arm has such a big leeway in power efficiency at the moment that I just don’t see any scenario where the X1 would end up disappointing.
...
Meanwhile the Cortex-X1 is a big change for Arm. And that change has less to do with the technology of the cores, and more with the business decisions that it now opens up for the company, although both are intertwined. For years many people were wondering why the company didn't design a core that could more closely compete with what Apple had built. In my view, one of the reasons for that was that Arm has always been constrained by the need to create a “one core fits all” design that could fit all of their customers’ needs – and not just the few flagship SoC designs.
The Cortex-X program here effectively unshackles Arm from these business limitations, and it allows the company to provide the best of both worlds. As a result, the A78 continues the company’s bread & butter design philosophy of power-performance-area leadership, whilst the X1 and its successors can now aim for the stars in terms of performance, without such strict area usage or power consumption limitations.
In this regard, the X1 seems really, really impressive. The 30% IPC improvement over the A77 is astounding and not something I had expected from the company this generation. The company has been incessantly beating the drum of their annual projected 20-25% improvements in performance – a pace which is currently well beyond what the competition has been able to achieve. These most recent projected performance figures are getting crazy close to the best that what we’ve seeing from the x86 players out there right now. That’s exciting for Arm, and should be worrying for the competition.
That quote makes no sense. The latest Qualcomm chip is well behind Apple’s chip of two years ago in performance and efficiency. If this will be 10% behind that, then it’s no competition to Apple at all. That’s according to Anandtech testing. The chart you see here shows, not an actual Cortex-X1 chip test, but an estimated number, which is indicated by the word—projection. That means nothing. It’s also an estimate of what it will be on 5nm. Useless.
It's not useless. It's based on what ARM presented from someone with an excellent track record.
Again, as a big note – these figures are largely my own projections based on the various data-points that Arm has presented. This can end up differently in actual products, but in the past our predictions of the A76 and A77 ended up extremely close to the actual silicon, if not even pessimistically worse than what the real figures ended up at.
Another key point from Arm:
Cortex-X1 is the very first example of a Cortex CPU that the CXC program can produce. It extends the digital immersion capabilities of smartphones through new levels of performance, making Cortex-X1 Arm’s most powerful CPU to date.
As part of the CXC program, subscribed partners collaborate with Arm to define custom CPUs that push performance at an envelope outside of the Cortex-A PPA. As a result, partners will have a CPU that is specific to their market needs and shows differentiation beyond roadmap Cortex-A CPUs. Through the CXC program, we are meeting the needs of the ever-expanding ecosystem, taking the best of Arm and applying it to the next level.
This is not simply a roadmap product. The Cortex X program allows partners to get the product they want for their needs.
Unlike what the article says, Apple didn’t say what the machines they were using for their demos had in the way of RAM or storage. They did say how much would be in the developer machines. And I would also like to point out that the iPad Pro has 6GB RAM, not 16GB, so the developer machines are not outfitted the same. That’s not surprising, because the Mac needs more resources than iOS devices.
Good point; but since they obviously had this transition in mind when they did the A12Z, I submit they designed an External memory bus into the SoC, allowing the internal RAM and Flash to be supplanted/replaced with external resources.
Many Microcontrollers have supported such an expansion bus for decades; in fact, before EPROM and flash-based microcontrollers were the norm, it was the only way to do a “Development Version” of most microcontrollers.
Afterall, when you’ve got a BGA package with a few hundred pads, what’s another 96 or so for an expansion bus?
Are you stating that the A12Z, and by extension, all Apple SoC’s have their RAM on chip, when you say internal? Because that’s not true.
Unlike what the article says, Apple didn’t say what the machines they were using for their demos had in the way of RAM or storage. They did say how much would be in the developer machines. And I would also like to point out that the iPad Pro has 6GB RAM, not 16GB, so the developer machines are not outfitted the same. That’s not surprising, because the Mac needs more resources than iOS devices.
Good point; but since they obviously had this transition in mind when they did the A12Z, I submit they designed an External memory bus into the SoC, allowing the internal RAM and Flash to be supplanted/replaced with external resources.
Many Microcontrollers have supported such an expansion bus for decades; in fact, before EPROM and flash-based microcontrollers were the norm, it was the only way to do a “Development Version” of most microcontrollers.
Afterall, when you’ve got a BGA package with a few hundred pads, what’s another 96 or so for an expansion bus?
Are you stating that the A12Z, and by extension, all Apple SoC’s have their RAM on chip, when you say internal? Because that’s not true.
Duh!
🤦♂️
You are of course, correct! I’ve seen enough iFixit teardowns not to know better, sorry!
My point is that Qualcomm is going to remain behind Apple in SOC performance, both in mobile and in notebook/desktop. The fact that Qualcomm has an architectural license yet has waited for Cortex X is telling, all the while Apple is innovating in the SOC space with the ARM ISA license. Apple's advantage is that it can provide synergy between its OS, development system and SOC designs, and other hardware.
Bzzt! Wrong!
Apple has held an ARM Architectural License since at least 2008. One of about 15 such licenses in the world.
Comments
this is different. Most Mac users are happy with this decision. Apple likely isn’t ready to give any info about what the new chips can do. Since they’re not going to sell these chips, there’s no reason to give out specs now to get OEM’s ready.
but, considering the performance that we saw in the demo/s using the A12Z, which is now an old chip, you can bet that the Mac versions, and yes, they said that this would be a line of chips designed for the Mac, will be MUCH better. Consider that the A13, out late last year is better than the A12Z in integer, and the new A14 out later this year, will be better. Same thing for the graphics cores. And they won’t be the Mac chip line.
swappable batteries are, by design necessity, smaller, and give less battery life, unless you want a thick, heavy battery, which almost no one wants.
But it is hardly an insurmountable development effort; even for a one-person shop.
So what was your point, again?
I was involved with education here in NYC for many years, being on the technology committee that made these decisions, abet as an advisor. Yes, school systems tend to go to “good enough” if it’s cheap enough. Chromebooks were half the price of an iPad, and included a keyboard. Simplistic, yes, but again, cheap. Apple countered with the new, bigger $329 iPad. But without a keyboard at that price, it came out to over 50% more. School districts have been more short of cash than every before. First the Bush recession was disastrous to school budgets everywhere. But things were looking better for a while, until congress turned to the dark, I mean, Republican side, and the administration couldn’t get past their hatred of federal education support. And then, we got Trump.
apple has actually clawed back some k-12 marketshare. We’ll have to see how far that goes.
I’m not so sure about that. Mac use in corporations has been rising at a good pace the past few years. Almost all software they use is Mac, not Windows. These days are over.
with Craig stating that all mac x86 software will run, making a point that plug-ins would work too, a point of contention during the older moves, I don’t see a lot of problems for corporate use. These machines also offer better security, always a sore point.
That’s not true. There was a great amount of service infrastructure support. There was, and is still almost none for Chromebooks, as well as a dearth of professionally created, and approved curriculum standard software for them, whereas there were loads for the Mac, and later, for the iPad.
Cortex X program allows partners to create custom products working with ARM to fit there needs. These are not on the ARM roadmap and have no set release timeline.
Synergy works in some cases, but doesn't guarantee superiority in an open market - especially one where the major productivity ecosystem lives with Windows and on x86-64.
Another key point from Arm:
This is not simply a roadmap product. The Cortex X program allows partners to get the product they want for their needs.
🤦♂️
You are of course, correct! I’ve seen enough iFixit teardowns not to know better, sorry!
Well, that made it much easier to implement...
(Slinks off into the darkness)...
Apple has held an ARM Architectural License since at least 2008. One of about 15 such licenses in the world.
https://en.m.wikipedia.org/wiki/Arm_Holdings