Qualcomm 5G modems are not on a separate chip, so that's not really any reason to stay with Qualcomm anymore. To your point, Apple's silicon give them a competitive advantage that I do not expect Apple to readily share with others.
That would be antithetical to Apple's philosophy. Apple is, according to rumors, specifically tailoring their future ARM SoCs to macOS.
The A14 and other chips are Apple's greatest technical advantage. An advantage that helped kill Microsoft's own flailing mobile platform.
Apple doesn't need Microsoft. They're literally abandoning Intel: Microsoft is (unfortunately) a non-starter in the mobile, low-power space, in volume and in quality.
Apple will never give up their "keys to the castle" to another trillion-dollar conglomerate.
Someone said Microsoft is the Hooli of our time and that is a bitter truth: generic incompetence, a helping of corporate arrogance, and a mountain of bloat.
I still believe they will have dedicated Mac ARM SoCs - the rumours seem to suggest using the plain jane A14X chip the iPad Pro will use, however I read them as using the same ARM core technology, in a new SoC.
I imagine the SoC will have a (sustained, can be exceeded for short periods of time) TDP target of 10W (MacBook Air), 25W (14" MacBook) and 45W (16" MacBook), and maybe 65W (iMac). This is probably double the iPad Pro TDP target. The big question is what the turbo clocks will be at these higher TDPs, in order to compete favourably with Intel (10nm) at the same TDP.
The GPU is another interesting factor - I'm sure they will be targeting 2-3 TFLOPS at least (FP32), i.e., around the same power as Renoir or Tiger Lake. That'll be a lot larger than the GPU in an iPad - although that could be where the TDP headroom comes in handy.
Hopefully they announce some techy details about this processor today, and don't leave us to wait.
You do realise the chip you are describing won't exist, right? It's simply not economical to scale from 10W to 65W with one chip. Nobody will do that. Ever.
A vanilla A12Z-like chip can efficiently scale to 15W easily, all they have to do is to enable 2.67GHz on all big cores and boost single core to 3GHz with 1GHz GPU. That's without designing the chip to fit both low-end laptop and pad in mind.
It's much more economical to design a chip to suite 7.5W-15W TDP then another one to suite 25W-60W. I would say they are releasing A14X a bit more powerful than an X part would normally be and clock it much lower for iPad Pro, boost up to 3GHz/1GHz for up to 15W laptop.
That is literally what AMD are doing with Renoir. You can cut some cores, reduce the clocks and get a lower-end, lower TDP chip. Or boost the clocks beyond that sweet spot in the perf/W curve and get a more powerful higher-TDP product.
Apple may, of course, have multiple silicon designs for different product areas. iMac, for example, can run entirely on the discrete GPU they include, so why bother with a [powerful] integrated one. Just stick 16 (or more!) "big A14 cores" on a chip [and some small cores and a small GPU for low power tasks] and call it a day.
What part of not economical do you not understand? AMD is literally SELLING the APU directly based on testing, and they have 14 SKUs so far. Apple cannot do that, Apple makes maybe 3-5 SKUs out of the whole range. Even with 14 SKUs, Renoir does NOT have the TDP range you are implying because one single chip simply can't do that.
Apple will share A14X between iPad and Macbook, that's a given. So you will be wrong. They will also make a more powerful chip to serve the >25W segment.
I'd say what psychobriggsy said turned out to be pretty accurate. The M1 essentially boosts to 3.2GHz on the big Firestorm cores in all of the Macs and iPads that it goes into. The differentiation factor lies in which ones can sustain those clock speeds, which results in them getting the same general system snappiness in daily usage, but a little bit different multicore performance in heavy multithreaded tasks. The M1 iPad Pro's power draw, though, is naturally quite a bit higher than the A12X/Z iPad Pro (8W sustained on the whole device vs. 10W) (The whole power draw of the device can peak at 20W in the new iPad Pros when not throttling). The GPU boosts to about 1.1-1.2GHz. For the 14 & 16" Macbook Pro, the higher-tier Space Gray Mac mini and bigger iMacs though, I expect them to up the TDP to 45-60W, not with the same chip, of course, but with a bigger chip with more cores (10-20 CPU cores, 16 big 4 little cores, 16-64 GPU cores).
My guess is that you'll see a range of A14 based variants. I would expect all MacBook variants to start with an A14x. However, on the desktop, Apple is going to have to be able to produce a more powerful variant with significantly more cores. Maybe multiple A14x chips or a base A14x with some sort of network fabric to multiple slave chips with just CPU cores? Who knows...
I wouldn't be surprised to see them keep selling at least some Intel-based models for two years after introducing ARM-based replacements. They'll definitely have refurbished Intel Macs for sale for at least that long.
Two years is the transition from selling Intel machines to selling only ARM machines. They obviously won't suddenly stop supporting Intel macs soon after they stop selling them, that would be idiotic. They will keep updating the OS for Intel for at least a few years after that, and application developers will likely keep releasing Intel updates for even more years.
Okay, now that we've seen the keynote... Tim said the transition will be complete in 2 years. That means no more Intel devices for sale at that time. Let's talk about what that means... 1. I think that is a very conservative timeframe. Apple will be happy to pat themselves on the back when the transition is complete in 1+ years from a hardware perspective. 2. Expect that transition complete signal they no longer need Rosetto 2. The keynote in 2022 will likely announce the end of that translation layer as it's no longer needed.
They're not going to drop Rosetta 2 that soon. The first version of OS X that dropped the original Rosetta was released about 5 years after the Intel transition.
Exactly and I agree with nearly everything you've said: they already separate iPhone vs iPad SoCs. I doubt they'd use the same SoC for macOS. A third, specific-to-mac-OS-and-not-Windows variant seems more likely.
My silly, uninformed guess on the clocks, at boost, will be in the ~3 GHz range (the iPad Pro's A12X/Z already boosts to 2.5 GHz single-core on the big cores). If Apple can eke out anything over 4 GHz in a 25W or lower TDP, I'll be impressed. They do need to add a fair amount of cores, if we're maintaining the multi-tasking capabilities of macOS.
That's another question: will Apple's laptop ARM chips even run on the latest Catalina macOS or perhaps only on next year's macOS? That is, only future versions of macOS will only be ARM-native, so they might not waste time porting Catalina (10.15) to ARM-native.
So perhaps there will be some major performance-limiting changes to macOS that make it more palatable to ARM-based workloads. But that wouldn't be very "Pro", either.
Unless they are imminently going to release hardware, they won't announce any hardware details, I bet, unfortunately. Software, they test openly. Hardware? It's a tight ship at Apple.
If they're 5nm then I could see them having a high base clock of around 3GHz, with a turbo to around 4GHz in the 25W TDP. It just depends on a lot of things we simply don't know about regarding Apple's A14 core design.
The low-power cores are going to give MacBooks a nice battery life advantage, as they'll be adequate for many tasks.
Apple didn't really hurt Microsoft in the mobile space. It was Google anti-competitive behaviors, such as preventing Google services from running on Windows mobile, buying Waze and shutting down the Windows Mobile app, etc. And it was Android phones competing with Windows Phones. Of course there was some overlap with the high end Nokia phones and iPhones, but Microsoft needed to thrive at the low end and middle tier phones to gain a foothold in the market.
It was a combination of Apple and Google that sunk Microsoft in mobile. Apple took the high end of the market (mostly) and Android took the mid to low end of the market. They also made it impossible for Microsoft to sell their OS as Google was giving theirs away for free. Either way, it was the developers not supporting Microsoft that ultimately made their platform no longer viable. The market will only support 2 major platforms apparently. The iPhone turned the market upside-down with the iPhone. Android was first to copy. Microsoft is the odd man out.
Google services on Windows? No thanks, don't even want it on Android, as surprising few apps require the spyware, such as Line. Microsoft messed up Windows Phone on its own: outdated processors and extremely buggy OS trying to run on too diverse hardware each with a few units sold. Nokia was spluttering the market with crap products when it should be focusing on optimizing each one and equipping them with competent hardware in the first place.
Do you really believe that either Apple or Microsoft would let any of that nonsense stand between them and lots of moolah? Microsoft has long abandoned "Wintel" , and Apple has an interest in not forcing its iMac buying customers to have to learn a new suite of office software. Both companies are in this to make money, so they will align when it suits them and (importantly) makes money, and fight each other when it doesn't.
if Apple could split out hardware division and let Apple SOC support windows/android, I would be happier as Apple shareholder. That would increase potential big time.
"Also a new widget, called the smart stack, which can flip between multiple widgets"
AKA Siri Watch Face (which is not a criticism! I find the Siri Watch Face very useful, albeit not perfect). I love it when Apple makes these (only obvious in retrospect) cross-device borrowings.
While I would agree with you that Apple leaving Intel isn't going to devastate Intel in any way. Obviously, it's not a good thing, but not a catastrophic event in and of itself.
However, Apple didn't make this move knowing there wasn't going to be a major benefit in doing so. This will come both in terms of power per watt and also in terms of overall performance. This will lead to more powerful devices getting better battery life than Wintel counterparts. This will put pressure on the Wintel platform to eventually move to WinARM as well. Think of the Mac transition more of a leading indicator for the rest of the industry. Yes, Microsoft has flirted with ARM in the past, dipping your toe in the water as they have is very different from diving in head first as Apple does. Once Apple demonstrates the benefit to others in a big way, the rest will follow. Of course, once the client side switches, the same goes for the server side. This will take quite a few years to play out, but the writing is on the wall for Intel.
Yeah, fair question... I don't know so much that Qualcomm is a joke as the Android market is a joke. Qualcomm is supplying chips to meet the needs of their market. While there are indeed flagship Android phones using an SD865 chip, that's a very small part of the Android market and not where Qualcomm's "bread and butter" is. Qualcomm is an obvious contender, but they'd need a partner to commit to large scale volume purchases. Microsoft could help get that started. We'd just need to see more than previous efforts to date in this area. All the same, the direction is inevitable. How it will play out does remain a question though.
Apple always tries to excite, but their need to make massive profits and their tendency to ignore and not listen to their users will continue.
For example, we don't need a Mac Mini development kit. It is just running an A12Z. You could cheaply put it in a Apple TV and add a second display connector and some USB ports. But then you'd realize they can easily sell it for less than $400 dollars, since the Apple TV is $230. They want you to pay $1000 USD... that's their objective, which will most likely ruin the platform. I'm hoping this will give Microsoft a kick in the pants and we will finally start to see an ARM desktop platform develop for Windows, including DIY.
Recently we've seen $140 PSU's, $400 motherboards, and all the other overprice DIY parts ruining the ecosystem. ARM can bring those costs way down. Imagine $100 8 core CPUs, and $100 motherboards that don't require $140 PSUs, just run on silent direct 12V adapters etc.
Apple fanboy detected. Just because people use Apple doesn't mean they don't dislike how Apple continually doesn't listen to them. They likely use Apple that it's the least combo of compromises for their desired use case.
T9 dialing, don't end the call when turning off the screen, how about make Maps usable in sunlight, instead of dark? I have to pull up Google maps to use it in the sunshine...
Maybe let people use USB tethering without needing to DOWNLOAD iTunes? Lots of small issues that Apple never seems to fix.
Competetive ARM (m)ATX/ITX SoCs would indeed be fascinating, but theres no guarantee the other associated components would be any cheaper. A high power CPU is a high power CPU, not a Raspberry Pi.
Well. So we have the transition strategy detailed, all the software anounced.
But no new hardware. That is just so cumbersome. Other companies would have anounced the hardware a year before releasing it to public. But Apple is simply secretive all around always.
Just a development kit based on a 20 month old SoC. What about some benchmarks and reviews of this ?
Apple just has to be faster in the workloads that mac pro users actually care about.
Apple doesn't have to win in general purpose compute, or HPC. With GPU transcoding finally coming to Adobe Premier, even the traditional excuse of "CPU encoding" is starting to fade away.
The primary holdback is the lack of really performant arm javascript engines. Even V8 (Chromium) is largely unoptimized on arm chips.
Other main failing of existing SoC designs (sure, they are for cheaper devices than a $10k server, largely speaking) is storage subsystem performance, which traditionally hasn't been a problem with Apple SoCs.
While I, of course, don't have any hard evidence for it, I have (had) the suspicion that Microsoft making an ARM native version of Office, especially WORD and EXCEL, didn't exactly hurt Apple's decision to go "native" regarding their CPUs. Yes, Apple's design is based on ARM (and very much customized), but the shared features of ARM v8.x must have made it much easier for Microsoft to port Office to Apple's own CPUs. And, let's face it: many iMac users do use and are familiar with MS Office for Mac, and the more applications are native for any CPU and OS, the better the user experience.
So, industry watchers, please check if TSMC gets really busy again with Apple orders just after they finished the next iPhone SoC. Because that'll be the CPU (SoC) for the first iMac with home grown CPU. Actually, something like an improved Mac Air would make a lot of sense, especially if it has a 5G modem integrated or as an option. Always on and always connected is good for a premium ultraportable.
No idea why people here want more walled garden BS. And also want M$, Google to move to ARM as well for what ? x86 is old ? fucking all of these goddamned sites, every single thing that you use on the planet. majority of it is powered by Intel x86 and AMD x86 Intel marketshare is over 90% and AMD is over 4%, less than 5-6% is ARM that is on Datacenter technologies.
And now the next part is that garbage Tombraider shitty demo looks worse than a console Xb1 GFX fidelity, those consoles have a processor which is slower than Q6600 Intel CPU, and the GPU technology is at AMD Radeon 7000 series class to Polaris Sub RX480 class. That is for the top PS4 Pro and XB1X.
And then we have the DIY market, you choose the CPU, Memory, SSD, PSU, Mobo, GPU, Case, Display fucking OS, every single thing is a CHOICE, this is what innovation means, putting computers into people's hand from IBM bullshit rooms of MF systems. Now salivating at Apple's proprietary soldered trash and scummy businesses like locking out users from even repairing their own stuff.
It's unfortuante how many people just suck up that Apple's thumb like a little kid. That's what caused this whole Mobile market generating the same sheep minded mentality to refresh a phone very year, and gaming market like fortnite and other trash.
Excellent retort! No doubt resplendent of the depth of your reason and general comprehension. Snide little remarks from little people revelling in diminishing other peoples' genuine concerns with sarcastic immaturity.
Walled garden? yes. Benefits of ARM vs x86? yet to be determined if at all and don't bother rsdurrecting the old, tired and increasingly irrelevant risc vs cisc.debate. People seem to genuinely think ARM/apple are magically capable of out engineering the incumbents (Intel, amd, nvidia) at their own game, via marketing bluster and some ludicrous synthetic benchmarks. Tight integration may result in some efficientcy gains but ARM/apple are not going to create anything revolutionary in terms of general compute performance compared to amd etc. Where do these expectations originate? Mindless cult thinking?
Altered phone socs might be exceptional for MacBook air like computing but on the desktop where outright performance matters? I fail to see the benefit to anyone but Apple and their desire for absolute control of the Mac ecosystem and user eexperience. Ofcourse all for your own good since people seem incapable of being anything but salivating corporate slaves.
Why is it exciting for everyone? Why would ARM be any better than existing AMD x86 chips for example in the DIY workstation market where a vibrant market already exists? Established performance cpus exist on an existing architecture which are only going to improve with a vast existing largely optimised software ecosytem; the x86 instruction decoding baggage cost diminishing with every iteration and process shrink. ARM doesn't offer any significant benefit besides power efficiency in phone soc guise which is likely to vanish when cores are optimized for performance.
They did. But on a separate page for iOS. iPadOS today has them on the front page.
New is not their existence, it's - the variable sizes - the intermixing with app icons on the same page - the "smart stack" widgets that surface (hopefully mostly accurately) the most useful of a collection of widgets at a particular time
The interesting technological aspects are - how you, the developer, support the variable sizes - how you, the developer, indicate just how "relevant" a widget is at a particular time, so that the system can surface the most relevant one(s).
If you're going to be snarky, it helps to at least know the baseline issue about which you are snarking.
I suppose this is the end of the road for the Iris spec Intel CPUs? I don't think anyone else would be willing to pay Apple prices for those chips, now that AMD has made such a good APU.
in this pandamic no need to worry just start internet marketing and learn SEARCH ENGINE OPTIMIZATION SERVICES IN LONDON https://www.kanootesoft.co.uk/services/internet-ma... SEO is a way of search engine marketing that will help promote your website’s rank on SERPs by getting you better visibility and lesser bounce rates. There are a lot of ways you could do more efficient search engine marketing (SEM), such as SEO, paid inclusion or placement method, and contextual advertising. Through SEO, you have a way of bettering your volume and traffic quality in order to improve organic traffic from different search engines, thereby increasing your rank on not just Google, but other search engine pages, such as Yahoo and MSN, as well. Not just that, SEO gives you a chance to improve your click through rates as well.
in this pandamic no need to worry just start internet marketing and learn SEARCH ENGINE OPTIMIZATION SERVICES IN LONDON https://www.kanootesoft.co.uk/services/internet-ma... SEO is a way of search engine marketing that will help promote your website’s rank on SERPs by getting you better visibility and lesser bounce rates. There are a lot of ways you could do more efficient search engine marketing (SEM), such as SEO, paid inclusion or placement method, and contextual advertising. Through SEO, you have a way of bettering your volume and traffic quality in order to improve organic traffic from different search engines, thereby increasing your rank on not just Google, but other search engine pages, such as Yahoo and MSN, as well. Not just that, SEO gives you a chance to improve your click through rates as well.
in this pandamic no need to worry just start internet marketing and learn SEARCH ENGINE OPTIMIZATION SERVICES IN LONDON https://www.kanootesoft.co.uk/services/internet-ma... SEO is a way of search engine marketing that will help promote your website’s rank on SERPs by getting you better visibility and lesser bounce rates. There are a lot of ways you could do more efficient search engine marketing (SEM), such as SEO, paid inclusion or placement method, and contextual advertising. Through SEO, you have a way of bettering your volume and traffic quality in order to improve organic traffic from different search engines, thereby increasing your rank on not just Google, but other search engine pages, such as Yahoo and MSN, as well. Not just that, SEO gives you a chance to improve your click through rates as well.
my buddy's friend makes $96 hourly on the internet. She has been with out artwork for five months however final month her charge emerge as $12747 really on foot on the internet for some hours. study greater on this net internet site.....
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
Facebook
Twitter
RSS
78 Comments
Back to Article
sorten - Monday, June 22, 2020 - link
It would be great if Apple and Microsoft could become frenemies like Google and Microsoft so that the next Surface Pro X could be built on the A14.sorten - Monday, June 22, 2020 - link
Now that Visual Studio Code has a preview release on Windows on ARM, I have nothing stopping me from making the switch.Smell This - Monday, June 22, 2020 - link
I think you may well be in fantasy La-La Land.
Neither MS nor Apple will cede ground on brand or marketing, and the significance of the Qualcomm/Snapdragon SOC & modem cannot be discounted.
'Intel on ARM' is more likely (rolling eyes) ...
techconc - Monday, June 22, 2020 - link
Qualcomm 5G modems are not on a separate chip, so that's not really any reason to stay with Qualcomm anymore. To your point, Apple's silicon give them a competitive advantage that I do not expect Apple to readily share with others.techconc - Monday, June 22, 2020 - link
Type... I meant "now" on a separate chip.sorten - Monday, June 22, 2020 - link
Of course I am. LOL. I don't THINK it will happen. I just WISH it would happen.ikjadoon - Monday, June 22, 2020 - link
That would be antithetical to Apple's philosophy. Apple is, according to rumors, specifically tailoring their future ARM SoCs to macOS.The A14 and other chips are Apple's greatest technical advantage. An advantage that helped kill Microsoft's own flailing mobile platform.
Apple doesn't need Microsoft. They're literally abandoning Intel: Microsoft is (unfortunately) a non-starter in the mobile, low-power space, in volume and in quality.
Apple will never give up their "keys to the castle" to another trillion-dollar conglomerate.
Someone said Microsoft is the Hooli of our time and that is a bitter truth: generic incompetence, a helping of corporate arrogance, and a mountain of bloat.
psychobriggsy - Monday, June 22, 2020 - link
I still believe they will have dedicated Mac ARM SoCs - the rumours seem to suggest using the plain jane A14X chip the iPad Pro will use, however I read them as using the same ARM core technology, in a new SoC.I imagine the SoC will have a (sustained, can be exceeded for short periods of time) TDP target of 10W (MacBook Air), 25W (14" MacBook) and 45W (16" MacBook), and maybe 65W (iMac). This is probably double the iPad Pro TDP target. The big question is what the turbo clocks will be at these higher TDPs, in order to compete favourably with Intel (10nm) at the same TDP.
The GPU is another interesting factor - I'm sure they will be targeting 2-3 TFLOPS at least (FP32), i.e., around the same power as Renoir or Tiger Lake. That'll be a lot larger than the GPU in an iPad - although that could be where the TDP headroom comes in handy.
Hopefully they announce some techy details about this processor today, and don't leave us to wait.
dotjaz - Monday, June 22, 2020 - link
You do realise the chip you are describing won't exist, right? It's simply not economical to scale from 10W to 65W with one chip. Nobody will do that. Ever.A vanilla A12Z-like chip can efficiently scale to 15W easily, all they have to do is to enable 2.67GHz on all big cores and boost single core to 3GHz with 1GHz GPU. That's without designing the chip to fit both low-end laptop and pad in mind.
It's much more economical to design a chip to suite 7.5W-15W TDP then another one to suite 25W-60W. I would say they are releasing A14X a bit more powerful than an X part would normally be and clock it much lower for iPad Pro, boost up to 3GHz/1GHz for up to 15W laptop.
psychobriggsy - Monday, June 22, 2020 - link
That is literally what AMD are doing with Renoir. You can cut some cores, reduce the clocks and get a lower-end, lower TDP chip. Or boost the clocks beyond that sweet spot in the perf/W curve and get a more powerful higher-TDP product.Apple may, of course, have multiple silicon designs for different product areas. iMac, for example, can run entirely on the discrete GPU they include, so why bother with a [powerful] integrated one. Just stick 16 (or more!) "big A14 cores" on a chip [and some small cores and a small GPU for low power tasks] and call it a day.
dotjaz - Monday, June 22, 2020 - link
What part of not economical do you not understand? AMD is literally SELLING the APU directly based on testing, and they have 14 SKUs so far. Apple cannot do that, Apple makes maybe 3-5 SKUs out of the whole range. Even with 14 SKUs, Renoir does NOT have the TDP range you are implying because one single chip simply can't do that.dotjaz - Monday, June 22, 2020 - link
Apple will share A14X between iPad and Macbook, that's a given. So you will be wrong. They will also make a more powerful chip to serve the >25W segment.Jorgp2 - Monday, June 22, 2020 - link
Lol, no.Renoir is not scaling down to 10w, that's why they built a whole new die for their APUs.
And even then, those APUs won't go down to tens of watts
caribbeanblue - Thursday, July 29, 2021 - link
I'd say what psychobriggsy said turned out to be pretty accurate. The M1 essentially boosts to 3.2GHz on the big Firestorm cores in all of the Macs and iPads that it goes into. The differentiation factor lies in which ones can sustain those clock speeds, which results in them getting the same general system snappiness in daily usage, but a little bit different multicore performance in heavy multithreaded tasks. The M1 iPad Pro's power draw, though, is naturally quite a bit higher than the A12X/Z iPad Pro (8W sustained on the whole device vs. 10W) (The whole power draw of the device can peak at 20W in the new iPad Pros when not throttling). The GPU boosts to about 1.1-1.2GHz. For the 14 & 16" Macbook Pro, the higher-tier Space Gray Mac mini and bigger iMacs though, I expect them to up the TDP to 45-60W, not with the same chip, of course, but with a bigger chip with more cores (10-20 CPU cores, 16 big 4 little cores, 16-64 GPU cores).techconc - Monday, June 22, 2020 - link
My guess is that you'll see a range of A14 based variants. I would expect all MacBook variants to start with an A14x. However, on the desktop, Apple is going to have to be able to produce a more powerful variant with significantly more cores. Maybe multiple A14x chips or a base A14x with some sort of network fabric to multiple slave chips with just CPU cores? Who knows...eek2121 - Monday, June 22, 2020 - link
My guess is that we will continue to see x86 used in Macs for a long time to come.SarahKerrigan - Monday, June 22, 2020 - link
Tim Cook said the transition will be completed within two years. So depending on your definition of "a long time to come", not really.Billy Tallis - Monday, June 22, 2020 - link
I wouldn't be surprised to see them keep selling at least some Intel-based models for two years after introducing ARM-based replacements. They'll definitely have refurbished Intel Macs for sale for at least that long.prisonerX - Monday, June 22, 2020 - link
Two years is the transition from selling Intel machines to selling only ARM machines. They obviously won't suddenly stop supporting Intel macs soon after they stop selling them, that would be idiotic. They will keep updating the OS for Intel for at least a few years after that, and application developers will likely keep releasing Intel updates for even more years.techconc - Monday, June 22, 2020 - link
Okay, now that we've seen the keynote... Tim said the transition will be complete in 2 years. That means no more Intel devices for sale at that time. Let's talk about what that means...1. I think that is a very conservative timeframe. Apple will be happy to pat themselves on the back when the transition is complete in 1+ years from a hardware perspective.
2. Expect that transition complete signal they no longer need Rosetto 2. The keynote in 2022 will likely announce the end of that translation layer as it's no longer needed.
Billy Tallis - Monday, June 22, 2020 - link
They're not going to drop Rosetta 2 that soon. The first version of OS X that dropped the original Rosetta was released about 5 years after the Intel transition.ikjadoon - Monday, June 22, 2020 - link
Exactly and I agree with nearly everything you've said: they already separate iPhone vs iPad SoCs. I doubt they'd use the same SoC for macOS. A third, specific-to-mac-OS-and-not-Windows variant seems more likely.My silly, uninformed guess on the clocks, at boost, will be in the ~3 GHz range (the iPad Pro's A12X/Z already boosts to 2.5 GHz single-core on the big cores). If Apple can eke out anything over 4 GHz in a 25W or lower TDP, I'll be impressed. They do need to add a fair amount of cores, if we're maintaining the multi-tasking capabilities of macOS.
That's another question: will Apple's laptop ARM chips even run on the latest Catalina macOS or perhaps only on next year's macOS? That is, only future versions of macOS will only be ARM-native, so they might not waste time porting Catalina (10.15) to ARM-native.
So perhaps there will be some major performance-limiting changes to macOS that make it more palatable to ARM-based workloads. But that wouldn't be very "Pro", either.
Unless they are imminently going to release hardware, they won't announce any hardware details, I bet, unfortunately. Software, they test openly. Hardware? It's a tight ship at Apple.
psychobriggsy - Monday, June 22, 2020 - link
If they're 5nm then I could see them having a high base clock of around 3GHz, with a turbo to around 4GHz in the 25W TDP. It just depends on a lot of things we simply don't know about regarding Apple's A14 core design.The low-power cores are going to give MacBooks a nice battery life advantage, as they'll be adequate for many tasks.
sorten - Monday, June 22, 2020 - link
Apple didn't really hurt Microsoft in the mobile space. It was Google anti-competitive behaviors, such as preventing Google services from running on Windows mobile, buying Waze and shutting down the Windows Mobile app, etc. And it was Android phones competing with Windows Phones. Of course there was some overlap with the high end Nokia phones and iPhones, but Microsoft needed to thrive at the low end and middle tier phones to gain a foothold in the market.techconc - Monday, June 22, 2020 - link
It was a combination of Apple and Google that sunk Microsoft in mobile. Apple took the high end of the market (mostly) and Android took the mid to low end of the market. They also made it impossible for Microsoft to sell their OS as Google was giving theirs away for free. Either way, it was the developers not supporting Microsoft that ultimately made their platform no longer viable. The market will only support 2 major platforms apparently. The iPhone turned the market upside-down with the iPhone. Android was first to copy. Microsoft is the odd man out.fmcjw - Monday, June 22, 2020 - link
Google services on Windows? No thanks, don't even want it on Android, as surprising few apps require the spyware, such as Line. Microsoft messed up Windows Phone on its own: outdated processors and extremely buggy OS trying to run on too diverse hardware each with a few units sold. Nokia was spluttering the market with crap products when it should be focusing on optimizing each one and equipping them with competent hardware in the first place.Fulljack - Monday, June 22, 2020 - link
I agree with you. why would Apple want to sell or even license their custom A14 core for other companies to use?that doesn't make any sense, they want everyone in line for their ecosystem.
eastcoast_pete - Monday, June 22, 2020 - link
Do you really believe that either Apple or Microsoft would let any of that nonsense stand between them and lots of moolah? Microsoft has long abandoned "Wintel" , and Apple has an interest in not forcing its iMac buying customers to have to learn a new suite of office software. Both companies are in this to make money, so they will align when it suits them and (importantly) makes money, and fight each other when it doesn't.trivik12 - Monday, June 22, 2020 - link
if Apple could split out hardware division and let Apple SOC support windows/android, I would be happier as Apple shareholder. That would increase potential big time.psychobriggsy - Monday, June 22, 2020 - link
6pm UK - just when I have to get the boy's dinner ready :(7pm Europe
10.30pm India
jeremyshaw - Monday, June 22, 2020 - link
Get everything ready earlier - a nice 5:30PM dinner never hurt :Dikjadoon - Monday, June 22, 2020 - link
1 pm US EST12 pm US Central
For other timezone-illiterate Americans like me.
WaltC - Monday, June 22, 2020 - link
More excitement in the Walled Garden...;) Oh, joy.name99 - Monday, June 22, 2020 - link
"Also a new widget, called the smart stack, which can flip between multiple widgets"AKA Siri Watch Face (which is not a criticism! I find the Siri Watch Face very useful, albeit not perfect). I love it when Apple makes these (only obvious in retrospect) cross-device borrowings.
ikjadoon - Monday, June 22, 2020 - link
Default email and browser apps can NOW be set!https://imgur.com/a/8q1kvEc
This was NOT covered by the speakers, but it's right there. Wow. Wowwww.
name99 - Monday, June 22, 2020 - link
"Now on to Home technology"Hey, I never knew Michael Che worked for Apple now :-)
Chaitanya - Monday, June 22, 2020 - link
Intel is going to be in a lot of pain after this move.Bubba2020 - Monday, June 22, 2020 - link
Why?Meteor2 - Monday, June 22, 2020 - link
Because it indicates the direction of the future. It’s ARM, not x86.Bubba2020 - Monday, June 22, 2020 - link
Apple is only a few percent of Intel's sales.techconc - Monday, June 22, 2020 - link
While I would agree with you that Apple leaving Intel isn't going to devastate Intel in any way. Obviously, it's not a good thing, but not a catastrophic event in and of itself.However, Apple didn't make this move knowing there wasn't going to be a major benefit in doing so. This will come both in terms of power per watt and also in terms of overall performance. This will lead to more powerful devices getting better battery life than Wintel counterparts. This will put pressure on the Wintel platform to eventually move to WinARM as well. Think of the Mac transition more of a leading indicator for the rest of the industry. Yes, Microsoft has flirted with ARM in the past, dipping your toe in the water as they have is very different from diving in head first as Apple does. Once Apple demonstrates the benefit to others in a big way, the rest will follow. Of course, once the client side switches, the same goes for the server side. This will take quite a few years to play out, but the writing is on the wall for Intel.
techconc - Monday, June 22, 2020 - link
I'd add that Stratechery has a nice writeup covering this angle as well.https://stratechery.com/2020/apple-arm-and-intel/
Bubba2020 - Monday, June 22, 2020 - link
Who is going to make these CPU's that will run Windows? Qualcomm is a joke.techconc - Monday, June 22, 2020 - link
Yeah, fair question... I don't know so much that Qualcomm is a joke as the Android market is a joke. Qualcomm is supplying chips to meet the needs of their market. While there are indeed flagship Android phones using an SD865 chip, that's a very small part of the Android market and not where Qualcomm's "bread and butter" is. Qualcomm is an obvious contender, but they'd need a partner to commit to large scale volume purchases. Microsoft could help get that started. We'd just need to see more than previous efforts to date in this area. All the same, the direction is inevitable. How it will play out does remain a question though.alufan - Monday, June 22, 2020 - link
guess where jim keller will pop up next?Teckk - Monday, June 22, 2020 - link
What about Windows on Mac with Apple Silicon?eek2121 - Monday, June 22, 2020 - link
Intel’s stock is about to bomb.Bubba2020 - Monday, June 22, 2020 - link
It spiked upward at the announcementPeterCollier - Monday, June 22, 2020 - link
And let me guess: you're now a millionaire because you actually believe what you typed and shorted Intel just before it started to "bomb."Did I get it right, or are you still stuck in your mother's basement?
Alistair - Monday, June 22, 2020 - link
Apple always tries to excite, but their need to make massive profits and their tendency to ignore and not listen to their users will continue.For example, we don't need a Mac Mini development kit. It is just running an A12Z. You could cheaply put it in a Apple TV and add a second display connector and some USB ports. But then you'd realize they can easily sell it for less than $400 dollars, since the Apple TV is $230. They want you to pay $1000 USD... that's their objective, which will most likely ruin the platform. I'm hoping this will give Microsoft a kick in the pants and we will finally start to see an ARM desktop platform develop for Windows, including DIY.
Recently we've seen $140 PSU's, $400 motherboards, and all the other overprice DIY parts ruining the ecosystem. ARM can bring those costs way down. Imagine $100 8 core CPUs, and $100 motherboards that don't require $140 PSUs, just run on silent direct 12V adapters etc.
It's time for DIY to mature to ARM.
dendem - Monday, June 22, 2020 - link
$500. The ARM MacMini is $500. A better deal than some boogied up Apple TV.Alistair - Monday, June 22, 2020 - link
no it is not $500, you have to return it afterwards, it is $500 to rent it and get access to the programMeteor2 - Monday, June 22, 2020 - link
“Apple doesn’t listen to its users but makes massive profits.” Interesting dissonance you must be feeling thereHifihedgehog - Monday, June 22, 2020 - link
Apple fanboy detected. Just because people use Apple doesn't mean they don't dislike how Apple continually doesn't listen to them. They likely use Apple that it's the least combo of compromises for their desired use case.Hifihedgehog - Monday, June 22, 2020 - link
*because it's the least combo...Bubba2020 - Monday, June 22, 2020 - link
Apple hater detected :)Alistair - Monday, June 22, 2020 - link
T9 dialing, don't end the call when turning off the screen, how about make Maps usable in sunlight, instead of dark? I have to pull up Google maps to use it in the sunshine...Maybe let people use USB tethering without needing to DOWNLOAD iTunes? Lots of small issues that Apple never seems to fix.
brucethemoose - Monday, June 22, 2020 - link
Uhh, I don't think dev kits are that simple.Competetive ARM (m)ATX/ITX SoCs would indeed be fascinating, but theres no guarantee the other associated components would be any cheaper. A high power CPU is a high power CPU, not a Raspberry Pi.
Meteor2 - Monday, June 22, 2020 - link
I’ll never understand British shareholders selling ARM to a Japanese company. Foregoing vast future profits for a few quid today.GC2:CS - Monday, June 22, 2020 - link
Well. So we have the transition strategy detailed, all the software anounced.But no new hardware. That is just so cumbersome. Other companies would have anounced the hardware a year before releasing it to public. But Apple is simply secretive all around always.
Just a development kit based on a 20 month old SoC. What about some benchmarks and reviews of this ?
kgardas - Monday, June 22, 2020 - link
I'm really curious how in 2 years they will beat Xeon W-32xx in current Mac PRO. If not, then next alternative is to kill Mac PRO line completely.jeremyshaw - Monday, June 22, 2020 - link
Apple just has to be faster in the workloads that mac pro users actually care about.Apple doesn't have to win in general purpose compute, or HPC. With GPU transcoding finally coming to Adobe Premier, even the traditional excuse of "CPU encoding" is starting to fade away.
The primary holdback is the lack of really performant arm javascript engines. Even V8 (Chromium) is largely unoptimized on arm chips.
Other main failing of existing SoC designs (sure, they are for cheaper devices than a $10k server, largely speaking) is storage subsystem performance, which traditionally hasn't been a problem with Apple SoCs.
eastcoast_pete - Monday, June 22, 2020 - link
While I, of course, don't have any hard evidence for it, I have (had) the suspicion that Microsoft making an ARM native version of Office, especially WORD and EXCEL, didn't exactly hurt Apple's decision to go "native" regarding their CPUs. Yes, Apple's design is based on ARM (and very much customized), but the shared features of ARM v8.x must have made it much easier for Microsoft to port Office to Apple's own CPUs. And, let's face it: many iMac users do use and are familiar with MS Office for Mac, and the more applications are native for any CPU and OS, the better the user experience.eastcoast_pete - Monday, June 22, 2020 - link
So, industry watchers, please check if TSMC gets really busy again with Apple orders just after they finished the next iPhone SoC. Because that'll be the CPU (SoC) for the first iMac with home grown CPU.Actually, something like an improved Mac Air would make a lot of sense, especially if it has a 5G modem integrated or as an option. Always on and always connected is good for a premium ultraportable.
brucethemoose - Monday, June 22, 2020 - link
Its a good time to be ARM.I'd bet good money that this will kickstart MS and Samsung ARM laptop/desktop efforts.
Quantumz0d - Monday, June 22, 2020 - link
No idea why people here want more walled garden BS. And also want M$, Google to move to ARM as well for what ? x86 is old ? fucking all of these goddamned sites, every single thing that you use on the planet. majority of it is powered by Intel x86 and AMD x86 Intel marketshare is over 90% and AMD is over 4%, less than 5-6% is ARM that is on Datacenter technologies.And now the next part is that garbage Tombraider shitty demo looks worse than a console Xb1 GFX fidelity, those consoles have a processor which is slower than Q6600 Intel CPU, and the GPU technology is at AMD Radeon 7000 series class to Polaris Sub RX480 class. That is for the top PS4 Pro and XB1X.
And then we have the DIY market, you choose the CPU, Memory, SSD, PSU, Mobo, GPU, Case, Display fucking OS, every single thing is a CHOICE, this is what innovation means, putting computers into people's hand from IBM bullshit rooms of MF systems. Now salivating at Apple's proprietary soldered trash and scummy businesses like locking out users from even repairing their own stuff.
It's unfortuante how many people just suck up that Apple's thumb like a little kid. That's what caused this whole Mobile market generating the same sheep minded mentality to refresh a phone very year, and gaming market like fortnite and other trash.
I didn't even mention other BS aspects of this.
biigD - Monday, June 22, 2020 - link
Someone needs a hug!sparcplug - Monday, June 22, 2020 - link
Excellent retort! No doubt resplendent of the depth of your reason and general comprehension. Snide little remarks from little people revelling in diminishing other peoples' genuine concerns with sarcastic immaturity.Walled garden? yes.
Benefits of ARM vs x86? yet to be determined if at all and don't bother rsdurrecting the old, tired and increasingly irrelevant risc vs cisc.debate. People seem to genuinely think ARM/apple are magically capable of out engineering the incumbents (Intel, amd, nvidia) at their own game, via marketing bluster and some ludicrous synthetic benchmarks. Tight integration may result in some efficientcy gains but ARM/apple are not going to create anything revolutionary in terms of general compute performance compared to amd etc. Where do these expectations originate? Mindless cult thinking?
Altered phone socs might be exceptional for MacBook air like computing but on the desktop where outright performance matters? I fail to see the benefit to anyone but Apple and their desire for absolute control of the Mac ecosystem and user eexperience. Ofcourse all for your own good since people seem incapable of being anything but salivating corporate slaves.
Alistair - Monday, June 22, 2020 - link
you are missing how Apple will be a leader in this area, this is exciting for everyone, and will bring ARM to DYI and Windows faster alsosparcplug - Monday, June 22, 2020 - link
Why is it exciting for everyone? Why would ARM be any better than existing AMD x86 chips for example in the DIY workstation market where a vibrant market already exists? Established performance cpus exist on an existing architecture which are only going to improve with a vast existing largely optimised software ecosytem; the x86 instruction decoding baggage cost diminishing with every iteration and process shrink. ARM doesn't offer any significant benefit besides power efficiency in phone soc guise which is likely to vanish when cores are optimized for performance.LilDenuchi - Monday, June 22, 2020 - link
Dear god. I thought iOS had app drawers and widgets all this time.name99 - Monday, June 22, 2020 - link
They did. But on a separate page for iOS. iPadOS today has them on the front page.New is not their existence, it's
- the variable sizes
- the intermixing with app icons on the same page
- the "smart stack" widgets that surface (hopefully mostly accurately) the most useful of a collection of widgets at a particular time
The interesting technological aspects are
- how you, the developer, support the variable sizes
- how you, the developer, indicate just how "relevant" a widget is at a particular time, so that the system can surface the most relevant one(s).
If you're going to be snarky, it helps to at least know the baseline issue about which you are snarking.
ads295 - Monday, June 22, 2020 - link
I suppose this is the end of the road for the Iris spec Intel CPUs? I don't think anyone else would be willing to pay Apple prices for those chips, now that AMD has made such a good APU.kanootesoft - Wednesday, June 24, 2020 - link
in this pandamic no need to worry just start internet marketing and learnSEARCH ENGINE OPTIMIZATION SERVICES IN LONDON
https://www.kanootesoft.co.uk/services/internet-ma...
SEO is a way of search engine marketing that will help promote your website’s rank on SERPs by getting you better visibility and lesser bounce rates. There are a lot of ways you could do more efficient search engine marketing (SEM), such as SEO, paid inclusion or placement method, and contextual advertising. Through SEO, you have a way of bettering your volume and traffic quality in order to improve organic traffic from different search engines, thereby increasing your rank on not just Google, but other search engine pages, such as Yahoo and MSN, as well. Not just that, SEO gives you a chance to improve your click through rates as well.
kanootesoft - Wednesday, June 24, 2020 - link
in this pandamic no need to worry just start internet marketing and learnSEARCH ENGINE OPTIMIZATION SERVICES IN LONDON
https://www.kanootesoft.co.uk/services/internet-ma...
SEO is a way of search engine marketing that will help promote your website’s rank on SERPs by getting you better visibility and lesser bounce rates. There are a lot of ways you could do more efficient search engine marketing (SEM), such as SEO, paid inclusion or placement method, and contextual advertising. Through SEO, you have a way of bettering your volume and traffic quality in order to improve organic traffic from different search engines, thereby increasing your rank on not just Google, but other search engine pages, such as Yahoo and MSN, as well. Not just that, SEO gives you a chance to improve your click through rates as well.
kanootesoft - Wednesday, June 24, 2020 - link
in this pandamic no need to worry just start internet marketing and learnSEARCH ENGINE OPTIMIZATION SERVICES IN LONDON
https://www.kanootesoft.co.uk/services/internet-ma...
SEO is a way of search engine marketing that will help promote your website’s rank on SERPs by getting you better visibility and lesser bounce rates. There are a lot of ways you could do more efficient search engine marketing (SEM), such as SEO, paid inclusion or placement method, and contextual advertising. Through SEO, you have a way of bettering your volume and traffic quality in order to improve organic traffic from different search engines, thereby increasing your rank on not just Google, but other search engine pages, such as Yahoo and MSN, as well. Not just that, SEO gives you a chance to improve your click through rates as well.
kanootesoft - Wednesday, June 24, 2020 - link
yesAffor - Tuesday, July 7, 2020 - link
my buddy's friend makes $96 hourly on the internet. She has been with out artwork for five months however final month her charge emerge as $12747 really on foot on the internet for some hours. study greater on this net internet site.....HERE☛ www.worknet8.com