If that is accurate, and there is need/desire to continue to run legacy x86 -- it seems that Apple and Intel could collaborate on ARM chips to do this efficiently.
While they could in theory, why would they in practice? So iPads and iPhones - mobile devices - could run x86 applications meant for desktops (laptops) and have a thoroughly unsuitable UX/UI as a result? Or so Macs could use Ax processors even though the power/heat limitations that render x86 processors undesirable in mobile do not exist in the MacBook and Mac form factors that have physical cords, much larger batteries and cooling measures?
Before Apple starts running desktop applications on devices with 5 inch screens, or starts putting mobile SOCs in laptops when far more powerful and versatile desktop SOCs not only exist but cost about the same you are going to have to come up with either a business or technical reason why this can or should be done.
Microsoft and before them Canonical (Ubuntu) have spent years doing work in this area, creating versions of their operating system that would run on both desktop CPUs and mobile SOCs and apps that would work on 5' smartphone screens as well as 50' desktop monitors. But Microsoft and Ubuntu did this because they were desperate for relevance and market share, and in the case of Microsoft attempting to compensate for their mobile app store being a barren wasteland compared to competing app stores by not only iOS but even Google and Amazon. (Our mobile users don't have any mobile apps? Well let them use desktop applications to replace them!) It is similar to Google bringing their Google Play mobile apps to Chrome OS desktop devices, many of whom do not even have touchscreens: a response to plummeting Android tablet sales as well as the fact that Chromebooks never really took off to begin with.
Desperation often drives a ton of bad tech and business decisions. Apple is not desperate - nowhere close - so trying to mix and match hardware and software between their mobile and desktop lines isn't something that should be on their radar.
If that is accurate, and there is need/desire to continue to run legacy x86 -- it seems that Apple and Intel could collaborate on ARM chips to do this efficiently.
While they could in theory, why would they in practice? So iPads and iPhones - mobile devices - could run x86 applications meant for desktops (laptops) and have a thoroughly unsuitable UX/UI as a result? Or so Macs could use Ax processors even though the power/heat limitations that render x86 processors undesirable in mobile do not exist in the MacBook and Mac form factors that have physical cords, much larger batteries and cooling measures?
Before Apple starts running desktop applications on devices with 5 inch screens, or starts putting mobile SOCs in laptops when far more powerful and versatile desktop SOCs not only exist but cost about the same you are going to have to come up with either a business or technical reason why this can or should be done.
Microsoft and before them Canonical (Ubuntu) have spent years doing work in this area, creating versions of their operating system that would run on both desktop CPUs and mobile SOCs and apps that would work on 5' smartphone screens as well as 50' desktop monitors. But Microsoft and Ubuntu did this because they were desperate for relevance and market share, and in the case of Microsoft attempting to compensate for their mobile app store being a barren wasteland compared to competing app stores by not only iOS but even Google and Amazon. (Our mobile users don't have any mobile apps? Well let them use desktop applications to replace them!) It is similar to Google bringing their Google Play mobile apps to Chrome OS desktop devices, many of whom do not even have touchscreens: a response to plummeting Android tablet sales as well as the fact that Chromebooks never really took off to begin with.
Desperation often drives a ton of bad tech and business decisions. Apple is not desperate - nowhere close - so trying to mix and match hardware and software between their mobile and desktop lines isn't something that should be on their radar.
I understand what you say and mostly agree. However, I was thinking more about the server side of things. The bulk of the servers run on Intel chips, but ARM servers are making inroads:
Digitimes Research: ARM server shipments to break 1 million units in 2017
Danny Kuo, DIGITIMES Research, Taipei [Monday 16 December 2013]
ARM-based server shipments worldwide are expected to reach 20,000 units in 2013, accounting for 0.2% of total server shipments and the volume is estimated to grow to 1.06 million by 2017 with a compound annual growth rate (CAGR) of 170%.
ARM-based servers started becoming available at the end of 2012 and as more vendors are promoting their related products in 2013, Digitimes Research expects ARM-based server shipments to gradually increase every year.
ARM has unveiled its objectives for the server business, including having ARM-based server shipments account for 5-10% of global volume by 2016 and 10-15% by 2017.
Digitimes Research believes that Windows Server operating system's support for the ARM architecture is the key factor deciding the growth of ARM-based server shipments.
Poorer than expected support could result in ARM-based server shipments only achieving a CAGR (2013-2017) of only 150%. However, stronger than expected support could result in a CAGR of 180%.
As I understand it, ARM servers can easily handle things like web, email, RSS, etc. * -- while most SQL DB servers run on Intel chips. Would it be worthwhile to have ARM servers that could run x86 code. The UI/UX issues would not apply to servers.
* Many Big Data NoSQL DBs (including Apple's FoundationDB) can run on ARM.
Now, Apple sold ~230 million ARM A9 devices in 2015, and, likely, the 12-month sales of A10 devices will be similar.
It is rumored that Apple is vastly expanding its server capacity. It seems to me that Apple could dogfood its own chips in many of those servers. This should realize savings: original cost**, operational cost (power, cooling, etc).
** why buy hundreds of thousands of Intel chips -- rather have Intel and others build Apple's ARM chips at a larger scale (and assumed lower net cost to Apple).
They have their back to the wall with missing out on the mobile market for processor technology, partly because the tied their wagon to the horses called Microsoft. They now think the solution to their problem is to become a commodity FAB house for ARM processors for companies who can not get in toTMSC and Samsung since Apple has those suppliers all wrapped around the Apple wagon wheel.
Yeah this is going to work out well for Intel margins, I wonder how the Intel Investors will feel in a could of years when Intel tells them Margin are suffering because the low cost competitors like TMSC eating their margin lunch. I guaranty the market will said this is the best thing that every happen to Intel get people to buy in and then the big guys dump out before the bad news.
It wasn't due to their tying their wagon to Microsoft at all. Instead, go back to ancient history: 10 whole years ago. Back then the PC, workstation and server/mainframe markets were huge, where all the money and prestige were, and they were dominated by Intel, AMD, Sun and IBM CPUs. (About this RISC versus CISC thing ... most RISC CPUs went into workstations and servers in those days). CPUs that went into smaller devices like mobile phones were much lower profile, earned much less revenue and were left to the likes of Samsung and Qualcomm, who were unable to break into the PC market. Case in point: Qualcomm made SOCs for Blackberry and Nokia devices, and Samsung provided the CPUs for a ton of dumb and feature phones, as did Motorola.
When Apple created the iPhone, they created their own custom design of the small device (low power/low heat) chips that were already going into Blackberry, Windows CE, Java Mobile and similar, so it is not as if they went to Intel asking them to design or manufacture a mobile CPU for them and Intel said no. Why would they, as Intel was already supplying CPUs for Macs by this time, as Apple had long abandoned the PowerPC IBM-Motorola RISC chips created by then?
You could say that Intel was slow to react to the Ax line and to mobile in general. Let's remember: Intel's problem wasn't missing out on the billions that comes from manufacturing mobile SOCs. Instead, it was that the mobile device market caused a huge - though not proportional! - decline in PC sales. Intel's problem isn't that they are not a major iPhone supplier, but rather the existence of the iPhone is why they aren't selling nearly as many PC CPUs as before. Intel had no way of knowing in advance that the iPhone was going to be as big as it was, and they certainly had no idea that the iPhone's success was going to come at the expense of PCs, because mobile phones, feature phones and even early smart phones - which were very common years before the iPhone - did not trigger a drop in PC sales.
Intel's decision to compete with ARM with low power x86 chips was not a bad one, as they provided as good or better performance to all but the very best ARM SOCs. Instead the problem was Intel's strategy. What Intel should have done was court Google to convince them to switch from ARM to x86 for their hardware. This would not have caused conflict with Microsoft, as Microsoft and their manufacturers were already using ARM chips for Windows Mobile hardware, and Microsoft under Ballmer had no desire to unify their mobile and PC platforms in any meaningful way. I believe that Intel could have won over Google by promoting the ability to use the same hardware platform for both Android and Chrome OS - meaning phones, tablets, laptops, TV boxes and whatever other hardware Google had in mind - had Intel made this pitch before Google and their partners standardized mainstream Android architecture with the Jellybean release.
Instead Intel decided to be one of the founding partners and major funders Tizen in order to compete with Android. This decision to do so was ridiculous, no so much because Tizen had no chance of competing with Android - as the original plan was for Tizen to be able to run Linux, web-based and possibly even Android and Microsoft apps - but because Tizen does not require x86 hardware! So though Intel invested years of effort and tens of millions of dollars in the Tizen project, the only Tizen devices on the planet run on Samsung's ARM SOCs, making Intel's position worse off than it would have been had they never spearheaded and helped fund the effort to create Tizen in the first place. So Intel's ambitions in mobile failed for the same reason that Nokia's did: they chose competing with Google and Android over collaborating with them. By the time Intel realized that Tizen was a bust (for everyone but Samsung) and Android would succeed long term and started marketing their x86 chips to Android OEMs as a result, it was too late.
Intel's decision to compete with ARM with low power x86 chips was not a bad one, as they provided as good or better performance to all but the very best ARM SOCs. Instead the problem was Intel's strategy. What Intel should have done was court Google to convince them to switch from ARM to x86 for their hardware. This would not have caused conflict with Microsoft, as Microsoft and their manufacturers were already using ARM chips for Windows Mobile hardware, and Microsoft under Ballmer had no desire to unify their mobile and PC platforms in any meaningful way. I believe that Intel could have won over Google by promoting the ability to use the same hardware platform for both Android and Chrome OS - meaning phones, tablets, laptops, TV boxes and whatever other hardware Google had in mind - had Intel made this pitch before Google and their partners standardized mainstream Android architecture with the Jellybean release.
Were the Intel chips of 2007-2009 competitive (size, power, heat dissipation) with the ARM chips then available?
It is doubtful that Intel will get Apple's business readily. TSMC has developed InFO which allows for thinner chip profiles. In addition, TSMC is about to move into the lead with respect to advanced manufacturing on smaller nodes.
While integrating the modem as part of the SoC allows for certain advantages, TSMC has other advantages over Intel.
Time will tell, but it is doubtful that Apple moves to Intel foundries anytime soon.
Fabrication is a very capital intensive business, much like chip design, and the moves on the chess board must be made years in advance of the results-- so Apple has spent years getting away from Samsung. Getting dependent on TSMC isn't a huge improvement, Apple has made itself independent- both by in-housing chip design and also by making their designs flexible for manufacturing.
So a few years down the line, Intel could win some business from TSMC. Apple will always keep multiple foundries fighting for its business, and thus they will need to compete on process.
Intel and TSMC both are great at process, though intel is ahead.
Legacy x86 will never run on ARM chips, except in emulation. What makes ARM into an ARM chip is its instruction set (ARM stands for "Advanced RISC Machines" and RISC stands for "Reduced Instruction Set Computing". x86 is CISC or "Complex Instruction Set Computing")
The ARM Instruction set is what gives it superior power-per-watt vs Intel approach of chasing superior-power-at-any-watts. To reduce power per watt , Intel can improve process-- and they have-- or go to another instruction set -- and if they do the latter it will no longer be x86.
But take the superior instruction set of an ARM, and the superior SoC Design of Apple and add Intel's process advantages, and that would put Apple even further ahead of everyone else. Literally nobody would be able to compete, even other ARM licensees (because LG is using off the shelf ARM IP, while Apple is doing better stuff in house.)
I honestly can't see Intel making SOTA 10nm SoCs for Apple at a price Apple would be willing to pay.
The estimated combined cost of Apple's new A8 processor and Qualcomm's MDM9625M modem account for $59.50
There are around 2 Billion transistors in an A8 and half a billion in an Intel Core i3 priced at $109
This costly investment (and loss of face by management) may signal Intel's expectation that ARM processors will eventually be used in laptops. It has been reported that Apple runs a version of macOS on ARM in the lab, and if Apple makes the switch to ARM it will be difficult to prevent copy-cat PC makers from emulating that move. ARM processors are a lot less costly than processors in laptops and desktops, so this switch may represent the biggest cost saving still to be achieved.
The S-series is still Apple custom designs, not ARM reference designs. It's also significantly more complex to make than the A-series.
This news does not preclude Intel from fabbing apple's custom designs. What it shows is Intel's willingness to fab SOCs that aren't x86 and aren't designed by Intel. That's a big strategic shift that could open the door to a deal with Apple.
im pretty sure intel can handle the complexity of the A-chips.
Except they need permission from ARM to fabricate non-standard implementations of the ARM instruction set, which Apple falls under. Perhaps consider reading the article next time.
Intel doesn't need any. Apple has a 64 bit Architecture License and Intel is just the fab in this scenario. TSMC doesn't have an ARM license either.
Intel has a 32 bit Architecture License so it could make any custom variant on the older 32 bit ARM cores and it certainly can buy a 64-bit Architecture License if it every wanted to.
Should be interesting what happens if SoftBank gets approval for purchasing ARM.
They have their back to the wall with missing out on the mobile market for processor technology, partly because the tied their wagon to the horses called Microsoft. They now think the solution to their problem is to become a commodity FAB house for ARM processors for companies who can not get in toTMSC and Samsung since Apple has those suppliers all wrapped around the Apple wagon wheel.
Yeah this is going to work out well for Intel margins, I wonder how the Intel Investors will feel in a could of years when Intel tells them Margin are suffering because the low cost competitors like TMSC eating their margin lunch. I guaranty the market will said this is the best thing that every happen to Intel get people to buy in and then the big guys dump out before the bad news.
It wasn't due to their tying their wagon to Microsoft at all. Instead, go back to ancient history: 10 whole years ago. Back then the PC, workstation and server/mainframe markets were huge, where all the money and prestige were, and they were dominated by Intel, AMD, Sun and IBM CPUs. (About this RISC versus CISC thing ... most RISC CPUs went into workstations and servers in those days). CPUs that went into smaller devices like mobile phones were much lower profile, earned much less revenue and were left to the likes of Samsung and Qualcomm, who were unable to break into the PC market. Case in point: Qualcomm made SOCs for Blackberry and Nokia devices, and Samsung provided the CPUs for a ton of dumb and feature phones, as did Motorola.
When Apple created the iPhone, they created their own custom design of the small device (low power/low heat) chips that were already going into Blackberry, Windows CE, Java Mobile and similar, so it is not as if they went to Intel asking them to design or manufacture a mobile CPU for them and Intel said no. Why would they, as Intel was already supplying CPUs for Macs by this time, as Apple had long abandoned the PowerPC IBM-Motorola RISC chips created by then?
You could say that Intel was slow to react to the Ax line and to mobile in general. Let's remember: Intel's problem wasn't missing out on the billions that comes from manufacturing mobile SOCs. Instead, it was that the mobile device market caused a huge - though not proportional! - decline in PC sales. Intel's problem isn't that they are not a major iPhone supplier, but rather the existence of the iPhone is why they aren't selling nearly as many PC CPUs as before. Intel had no way of knowing in advance that the iPhone was going to be as big as it was, and they certainly had no idea that the iPhone's success was going to come at the expense of PCs, because mobile phones, feature phones and even early smart phones - which were very common years before the iPhone - did not trigger a drop in PC sales.
Intel's decision to compete with ARM with low power x86 chips was not a bad one, as they provided as good or better performance to all but the very best ARM SOCs. Instead the problem was Intel's strategy. What Intel should have done was court Google to convince them to switch from ARM to x86 for their hardware. This would not have caused conflict with Microsoft, as Microsoft and their manufacturers were already using ARM chips for Windows Mobile hardware, and Microsoft under Ballmer had no desire to unify their mobile and PC platforms in any meaningful way. I believe that Intel could have won over Google by promoting the ability to use the same hardware platform for both Android and Chrome OS - meaning phones, tablets, laptops, TV boxes and whatever other hardware Google had in mind - had Intel made this pitch before Google and their partners standardized mainstream Android architecture with the Jellybean release.
Instead Intel decided to be one of the founding partners and major funders Tizen in order to compete with Android. This decision to do so was ridiculous, no so much because Tizen had no chance of competing with Android - as the original plan was for Tizen to be able to run Linux, web-based and possibly even Android and Microsoft apps - but because Tizen does not require x86 hardware! So though Intel invested years of effort and tens of millions of dollars in the Tizen project, the only Tizen devices on the planet run on Samsung's ARM SOCs, making Intel's position worse off than it would have been had they never spearheaded and helped fund the effort to create Tizen in the first place. So Intel's ambitions in mobile failed for the same reason that Nokia's did: they chose competing with Google and Android over collaborating with them. By the time Intel realized that Tizen was a bust (for everyone but Samsung) and Android would succeed long term and started marketing their x86 chips to Android OEMs as a result, it was too late.
My point is and still valid, Intel business model was based on MS creating software which require PC to be upgrade which allowed Intel to sell more and more PC year after year. When Processor tech stop getting faster year after year MS could no long make power hungary programs and what we all had was good enough. Intel missed lots of things and they missed them because they were linked with MS with out MS they could not sell Processors into all the various markets. Apple shown they can run their software on any processor platform.
You do not need to give me a lesson on computer I been in the computer industry since the 80s and saw the evolution first hand. PPC was a better chip and Apple new it, low power and better perform, issue was Motorola failed to keep it fresh, The same was true with the 68K processors from Motorola, but they failed to keep it going forward. Apple this time is controlling their future with ARM, they not relying on anyone to keep it moving forward, they ahead of the game.
Comments
Before Apple starts running desktop applications on devices with 5 inch screens, or starts putting mobile SOCs in laptops when far more powerful and versatile desktop SOCs not only exist but cost about the same you are going to have to come up with either a business or technical reason why this can or should be done.
Microsoft and before them Canonical (Ubuntu) have spent years doing work in this area, creating versions of their operating system that would run on both desktop CPUs and mobile SOCs and apps that would work on 5' smartphone screens as well as 50' desktop monitors. But Microsoft and Ubuntu did this because they were desperate for relevance and market share, and in the case of Microsoft attempting to compensate for their mobile app store being a barren wasteland compared to competing app stores by not only iOS but even Google and Amazon. (Our mobile users don't have any mobile apps? Well let them use desktop applications to replace them!) It is similar to Google bringing their Google Play mobile apps to Chrome OS desktop devices, many of whom do not even have touchscreens: a response to plummeting Android tablet sales as well as the fact that Chromebooks never really took off to begin with.
Desperation often drives a ton of bad tech and business decisions. Apple is not desperate - nowhere close - so trying to mix and match hardware and software between their mobile and desktop lines isn't something that should be on their radar.
I understand what you say and mostly agree. However, I was thinking more about the server side of things. The bulk of the servers run on Intel chips, but ARM servers are making inroads:
http://www.digitimes.com/news/a20131216PD207.html?chid=2
As I understand it, ARM servers can easily handle things like web, email, RSS, etc. * -- while most SQL DB servers run on Intel chips. Would it be worthwhile to have ARM servers that could run x86 code. The UI/UX issues would not apply to servers.
* Many Big Data NoSQL DBs (including Apple's FoundationDB) can run on ARM.
Now, Apple sold ~230 million ARM A9 devices in 2015, and, likely, the 12-month sales of A10 devices will be similar.
It is rumored that Apple is vastly expanding its server capacity. It seems to me that Apple could dogfood its own chips in many of those servers. This should realize savings: original cost**, operational cost (power, cooling, etc).
** why buy hundreds of thousands of Intel chips -- rather have Intel and others build Apple's ARM chips at a larger scale (and assumed lower net cost to Apple).
When Apple created the iPhone, they created their own custom design of the small device (low power/low heat) chips that were already going into Blackberry, Windows CE, Java Mobile and similar, so it is not as if they went to Intel asking them to design or manufacture a mobile CPU for them and Intel said no. Why would they, as Intel was already supplying CPUs for Macs by this time, as Apple had long abandoned the PowerPC IBM-Motorola RISC chips created by then?
You could say that Intel was slow to react to the Ax line and to mobile in general. Let's remember: Intel's problem wasn't missing out on the billions that comes from manufacturing mobile SOCs. Instead, it was that the mobile device market caused a huge - though not proportional! - decline in PC sales. Intel's problem isn't that they are not a major iPhone supplier, but rather the existence of the iPhone is why they aren't selling nearly as many PC CPUs as before. Intel had no way of knowing in advance that the iPhone was going to be as big as it was, and they certainly had no idea that the iPhone's success was going to come at the expense of PCs, because mobile phones, feature phones and even early smart phones - which were very common years before the iPhone - did not trigger a drop in PC sales.
Intel's decision to compete with ARM with low power x86 chips was not a bad one, as they provided as good or better performance to all but the very best ARM SOCs. Instead the problem was Intel's strategy. What Intel should have done was court Google to convince them to switch from ARM to x86 for their hardware. This would not have caused conflict with Microsoft, as Microsoft and their manufacturers were already using ARM chips for Windows Mobile hardware, and Microsoft under Ballmer had no desire to unify their mobile and PC platforms in any meaningful way. I believe that Intel could have won over Google by promoting the ability to use the same hardware platform for both Android and Chrome OS - meaning phones, tablets, laptops, TV boxes and whatever other hardware Google had in mind - had Intel made this pitch before Google and their partners standardized mainstream Android architecture with the Jellybean release.
Instead Intel decided to be one of the founding partners and major funders Tizen in order to compete with Android. This decision to do so was ridiculous, no so much because Tizen had no chance of competing with Android - as the original plan was for Tizen to be able to run Linux, web-based and possibly even Android and Microsoft apps - but because Tizen does not require x86 hardware! So though Intel invested years of effort and tens of millions of dollars in the Tizen project, the only Tizen devices on the planet run on Samsung's ARM SOCs, making Intel's position worse off than it would have been had they never spearheaded and helped fund the effort to create Tizen in the first place. So Intel's ambitions in mobile failed for the same reason that Nokia's did: they chose competing with Google and Android over collaborating with them. By the time Intel realized that Tizen was a bust (for everyone but Samsung) and Android would succeed long term and started marketing their x86 chips to Android OEMs as a result, it was too late.
Were the Intel chips of 2007-2009 competitive (size, power, heat dissipation) with the ARM chips then available?
There are around 2 Billion transistors in an A8 and half a billion in an Intel Core i3 priced at $109
Intel has a 32 bit Architecture License so it could make any custom variant on the older 32 bit ARM cores and it certainly can buy a 64-bit Architecture License if it every wanted to.
Should be interesting what happens if SoftBank gets approval for purchasing ARM.
My point is and still valid, Intel business model was based on MS creating software which require PC to be upgrade which allowed Intel to sell more and more PC year after year. When Processor tech stop getting faster year after year MS could no long make power hungary programs and what we all had was good enough. Intel missed lots of things and they missed them because they were linked with MS with out MS they could not sell Processors into all the various markets. Apple shown they can run their software on any processor platform.
You do not need to give me a lesson on computer I been in the computer industry since the 80s and saw the evolution first hand. PPC was a better chip and Apple new it, low power and better perform, issue was Motorola failed to keep it fresh, The same was true with the 68K processors from Motorola, but they failed to keep it going forward. Apple this time is controlling their future with ARM, they not relying on anyone to keep it moving forward, they ahead of the game.