Also, look at the sales numbers for Parallels Desktop and VMWare Fusion. If emulation of Windows weren't important to people, those products wouldn't be selling nearly as well as they are.
do yourself a favor and google correlation vs causation.
what are the sales numbers youre referring to? and who are they going to -- pros? enterprise? how does that relate to consumer-level macbooks and imacs?
Which process technology are their latest processors running on? 32nm. That's right! They are two full generations behind intel. Hard to believe. Their top units need a full 220 watts. Those are compared to intel's 125 watts for the equivalent units.
Nah! Nothing for them now.
The 9370 FX is for enthusiasts. The standard FX-8350 that I'm using is a TDP max of 125 W. The APU Excavator core of their last Bulldozer line is 30W-45W with a Radeon R9-300 series GPGPU SoC on it.
The Zen project is Keller's return and a complete move in a direction that will merge his experience creating Apple's A4 with his vast experience from DEC Alpha, AMD and more.
Also, look at the sales numbers for Parallels Desktop and VMWare Fusion. If emulation of Windows weren't important to people, those products wouldn't be selling nearly as well as they are.
1) It's virtualization, not emulation. Yiu really should understand the difference as their distinction is highly relevant to computing.
2) With the change from PPC to Intel we got a major refresh to the overall HW and we got much faster speeds over the lackluster advancements coming from IBM. I know I was certainly debating on getting another PowerBook because of the shortcomings of the PPC CPU performavce.
Then there was a n iconic Get A Mac ad campaign. On top of that we had some major iPod changes around that time and both iTunes for Windows and the iTunes Music Store were still new to many Windows users but showing the level of quality Apple could bring to users.
Finally, there were plenty of cultural and general industry changes that helped sell "PCs". That's not to say that the option to run Windows on a Mac (which didn't arrive until mid-2006) didn't help sell some additional Macs, but there is no logical way you can say running Windows on a Mac is the reason.
Either Intel licenses it to Apple or Apple has to use DP for their external display support and forego TB on what would be a low-cost "PC" by Apple. I don't think TB is something people with sub $800 "PCs" care about.
1) It's virtualization, not emulation. Yiu really should understand the difference as their distinction is highly relevant to computing.
I'm a software developer — I know the difference between virtualization and emulation, thanks. It is not highly relevant to this discussion, which is why I'm opting for simplicity in this case. The average Joe lumps every technology which runs one operating system on top of another under the umbrella of "emulation."
2) With the change from PPC to Intel we got a major refresh to the overall HW and we got much faster speeds over the lackluster advancements coming from IBM. I know I was certainly debating on getting another PowerBook because of the shortcomings of the PPC CPU performavce.
With a change from Intel to ARM, we would get much slower speeds. Think about how your debating would have gone if Apple had switched the other direction at that time, from Intel to PPC, because performance-wise that's a better analogy here. Would you have upgraded at all?
If the entire Mac line (note: I am not discounting the possibility of a Chromebook-type ultra-low-end model that doesn't run native apps) were converted to Intel, I doubt many existing users would upgrade at all until their existing machines died, given that it will most likely be years before ARM even manages to catch up to Intel's current processors, never mind what Intel will have by then. What's the benefit? Saving a little bit of money on the parts? Running a bunch of apps designed for a tiny touch screen with a completely different UI paradigm which could easily be emulated on an Intel processor anyway? This is just describing the Surface RT, and look what that did to the perception of the Surface line. Heck, just look at the ads MS is putting out now, and how desperate they are to make you forget the Surface RT ever existed. They're all, "Hey, look at the Surface Pro! It's just as powerful as that Mac! No really, we swear! And you can run your Windows apps on it!" Do you really think Apple will want to follow this example?
Finally, there were plenty of cultural and general industry changes that helped sell "PCs". That's not to say that the option to run Windows on a Mac (which didn't arrive until mid-2006) didn't help sell some additional Macs, but there is no logical way you can say running Windows on a Mac is the reason.
Running Windows on a Mac isn't the reason. Switching to Intel was the reason. Apple's sales exploded immediately after the Intel switch, and you can't really write that off as just being due to a marketing campaign. The Intel switch clearly made a difference. It had major advantages, chief among which was performance. That being said, running Windows certainly is a reason. Just go in an Apple store sometime and see how prominently Parallels is displayed. Watch a sales pitch given to a potential PC switcher. Heck, look at the software section of the Apple Store web site. The first thing you see is MS Office. The second one is Parallels. This is a major Mac benefit, which brings lots of customers, and Apple would be insane to lose it.
Every time I see a story like this, all you need to do is look at the use-case, and it falls flat on it's face.
1) Existing A-series and ARM based parts are still Pentium 3-era speeds on comparable workloads
2) Intel's Atom parts are barely faster than Pentium 3-era parts
3) Intel's onboard video has always been a joke, but mobile isn't much better.
I could see Apple eventually producing a CPU/GPU that is more capable than Intel's weak mobile offerings, but Apple is unlikely to create an environment where they have to produce two versions of software for any extended period of time (see Motorola vs PPC, and PPC vs Intel.) Also look at how Microsoft fell on it's face with the Surface. Even if Apple produced a full version of OS X for an ARM based iMac or MacBook, the lack of being unable to just run every previous version of software will be felt and it will fail. Apple had an advantage earlier on when it switched from PPC to Intel because it was switching from a platform that only it was really using for desktops, so it could effectively push everyone off it. Not true with switching from x86 (look again at Microsoft, who also produced ARM, MIPS and ALPHA versions of Windows NT or CE at previous points in time.)
It's extremely unlikely that Apple would produce a part that is anywhere performance parity of the Xeon series in the Mac Pro.
There's also been some grumbling lately that OS X is getting too dumbed down that even people who normally like Apple are considering alternatives. Apple might be at a point in it's life where it might be more arrogant than Microsoft in making unnecessary changes to the operating system.
Exactly. The only reason our company has any Macs is cause they can natively run x86 software and other operating systems. If Apple were to move off of x86, there would no longer be any reason for us to buy Apple hardware.
I agree concerning the dumbing down of MacOS. It concerns me greatly. It's already hard enough to run MacOS in our organization, but as more and more basic fuctionality (interoperability with Windows & Linux, i.e. SMB) is broken in each release of MacOS, there may come a point where we abandon the platform entirely. Apple really needs to think through many of their recent design decisions, both HW and SW.
Every time I see a story like this, all you need to do is look at the use-case, and it falls flat on it's face.
1) Existing A-series and ARM based parts are still Pentium 3-era speeds on comparable workloads.
Can you provide some evidence that Apple's ARM designs would not be able to compete with any workload on an Intel chip with a comparable TDP?
2. Windows compatibility isn't a big deal for most users. It matters a great deal for a vocal minority, but for most users it's a non-issue. I can see Apple deciding that losing that niche is an acceptable loss. Also, emulation can be a solution for some of those users.
It seems to have been important to somebody:
Also, look at the sales numbers for Parallels Desktop and VMWare Fusion. If emulation of Windows weren't important to people, those products wouldn't be selling nearly as well as they are.
I'm not sure sales numbers of Parallels Desktop and VMWare Fusion are available, but the purchase prices have dropped substantially, I've been offered both for free with several Mac sofware packages so sales numbers would be bogus anyway.
Apple's mac sales had been in the dumps before the Intel macs hit the stores. This was due to PPC chips falling further and further behind in the race with Intel. By switching to Intel, Apple's Mac products were once again as zippy as their Wintel rivals. In addition, Apple also offered a free program called, "Boot camp," to lessen the pain of anyone wanting to switch to Mac OS.
Sine then (and 2006 was a half-century ago in dog-years) Parallels Desktop and VMWare Fusion have made switching between platforms even less daunting. In addition multi-core computers are far more common today, and along with more RAM, make emulation a snap compared to 2006. None of this means users demand or want to run WIndows, it just means buying a Mac is an easier decision to make.
We are about two years away from iOS-based devices and computers outselling Wintel boxes... WIndows compatibly is a fading issue. Even Ballmer has moved on from offering a friendly reach around to Wintel users.
On the plus side, if they did switch to ARM, we might get another "no new features" performance focussed release of OS X.
The releases seemed to have ramped up with the change of management - Forstall leaving, putting Federighi and Ive in his place. Now that they've implemented the larger changes, I hope they dial back on the OS releases and focus more on stability. I think it's right to push ahead with new software requirements but one thing they did recently was requiring that new apps in the App Store have to use the iOS 8 SDK by February 2015 despite only being released September 2014. That's only 5 months for people to get all new software updates and test them all and it prevents any OS pre-Mavericks being used for app development. The process keeps things moving but they should allow for reasonable update cycles. Businesses need stable software and updating to bleeding edge beta software isn't always feasible. OS releases every 2 years would be fine if there's nothing major to add to them and it makes the releases more interesting. Then they can gives software update requirements a year vs < 6 months.
I'm not sure sales numbers of Parallels Desktop and VMWare Fusion are available, but the purchase prices have dropped substantially, I've been offered both for free with several Mac sofware packages so sales numbers would be bogus anyway.
The prices in 2007 for both products was $79.99 (source). Today, the price for Parallels is still $79.99, and the price for VMWare Fusion has dropped a whopping $10 to $69.99.
Both products of course frequently participate in promotions in order to compete with each other and secure market share.
Apple's mac sales had been in the dumps before the Intel macs hit the stores. This was due to PPC chips falling further and further behind in the race with Intel. By switching to Intel, Apple's Mac products were once again as zippy as their Wintel rivals. In addition, Apple also offered a free program called, "Boot camp," to lessen the pain of anyone wanting to switch to Mac OS.
Sine then (and 2006 was a half-century ago in dog-years) Parallels Desktop and VMWare Fusion have made switching between platforms even less daunting. In addition multi-core computers are far more common today, and along with more RAM, make emulation a snap compared to 2006. None of this means users demand or want to run WIndows, it just means buying a Mac is an easier decision to make.
All of which would be negated by switching to ARM.
We are about two years away from iOS-based devices and computers outselling Wintel boxes... WIndows compatibly [sic] is a fading issue. Even Ballmer has moved on from offering a friendly reach around to Wintel users.
Ballmer hasn't been in charge of MS for almost a year.
Windows still controls something like 90% of the PC market. Windows compatibility is still a very big deal. There are many people out there that have to run Windows software for their jobs, since business is still largely Windows-based, and many of these companies have custom in-house software that isn't cross-platform. These requirements tend to be a bit non-negotiable for people who have these requirements, so losing Windows compatibility means losing those customers. There's also the gaming crowd, who also tend to want Windows for obvious reasons.
I think it would be quite cool to have a zillion ARM cores in a desktop or laptop. Potentially a lot more power than a single quad core intel chip.
However, having made prototypes means nothing about future plans. NeXT supposedly had a many-core PPC version running for a long time in the labs before there was any speculation at all of any tie-up with Apple. And Apple supposedly had classic MacOS running in the lab on intel for years but never did anything commercial with it. Companies go down various paths to provide alternative routes if things don't go the way they expect with their main plans, or to try stuff out, or otherwise get experience with alternatives and/or contingency plans.
I have speculated that core volume could be the differentiator.
Moore's law has slowed down with Intel and with all the companies making ARM based parts, certainly on of the big silicon players is going to make a 64core (or more) ARM based chip. And it will have more raw compute power than the biggest Intel chips, probably slower single thread speed though, at least for a generation or two. There will be interesting problems sets for that kind of hardware.
In the consumer market, are we post-PC? One disadvantage Apple had with PowerPC was that it was really challenging to do a direct comparison to a Windows Intel machine, there were a lot of variables and the mass consensus was that PowerPC would have an edge for a little while and then get left behind. Running the same hardware, its just user experience to compare; Photoshop will run just as fast on OS X as Windows.
The 9370 FX is for enthusiasts. The standard FX-8350 that I'm using is a TDP max of 125 W. The APU Excavator core of their last Bulldozer line is 30W-45W with a Radeon R9-300 series GPGPU SoC on it.
The Zen project is Keller's return and a complete move in a direction that will merge his experience creating Apple's A4 with his vast experience from DEC Alpha, AMD and more.
Go to Anandtech. They just reviewed some of AMD's newest chips. Terrible, really. They hope that AMD will do well, so they try to give them the benefit of the doubt, but they're slowly dying. And yes, 220 watts is where their top chips are. Even the newest chip at 95 watts doesn't perform as well as a number of intels chips at half the power, for most tasks. And while the over locking is impressive, to get all the way up there requires taking the chip all the way to an astounding 260 watts.
Sorry, but this simply isn't acceptable. AMD was always a crap company, and it remains a crap company. They had almost three years in the sun when intel lost their way with the latter stages of Netburst. But then Intel turned around, and AMD has been left in the dust since with several crap designs. It's too late for them.
I'm a software developer — I know the difference between virtualization and emulation, thanks. It is not highly relevant to this discussion, which is why I'm opting for simplicity in this case. The average Joe lumps every technology which runs one operating system on top of another under the umbrella of "emulation."
With a change from Intel to ARM, we would get much slower speeds. Think about how your debating would have gone if Apple had switched the other direction at that time, from Intel to PPC, because performance-wise that's a better analogy here. Would you have upgraded at all?
If the entire Mac line (note: I am not discounting the possibility of a Chromebook-type ultra-low-end model that doesn't run native apps) were converted to Intel, I doubt many existing users would upgrade at all until their existing machines died, given that it will most likely be years before ARM even manages to catch up to Intel's current processors, never mind what Intel will have by then. What's the benefit? Saving a little bit of money on the parts? Running a bunch of apps designed for a tiny touch screen with a completely different UI paradigm which could easily be emulated on an Intel processor anyway? This is just describing the Surface RT, and look what that did to the perception of the Surface line. Heck, just look at the ads MS is putting out now, and how desperate they are to make you forget the Surface RT ever existed. They're all, "Hey, look at the Surface Pro! It's just as powerful as that Mac! No really, we swear! And you can run your Windows apps on it!" Do you really think Apple will want to follow this example?
Running Windows on a Mac isn't the reason. Switching to Intel was the reason. Apple's sales exploded immediately after the Intel switch, and you can't really write that off as just being due to a marketing campaign. The Intel switch clearly made a difference. It had major advantages, chief among which was performance. That being said, running Windows certainly is a reason. Just go in an Apple store sometime and see how prominently Parallels is displayed. Watch a sales pitch given to a potential PC switcher. Heck, look at the software section of the Apple Store web site. The first thing you see is MS Office. The second one is Parallels. This is a major Mac benefit, which brings lots of customers, and Apple would be insane to lose it.
This is a difficult issue for everyone to discuss. Moving to x86 was a wrenching experience. Watching intel dissolve with Netburst, and then having Jobs come out on stage saying that performance/watts was why Apple was moving to x86 was like a dream. How could that be right? I assumed that he had seen things at intel that no one else knew about, except, possibly, some other big intel customers. After all, while the G5 was slightly behind x86 in performance, it was actually gaining on it. I estimated that in less than a year, it would surpass it in performance.
But Apple moved for two reasons. One major one, and quite possibly THE major problem at the time, was that laptop sales were exceeding Desktop sales, and that looked to be increasing. Apple had no "modern" solution for that. IBM had failed to deliver a mobile G5, and Apple needed a mobile chip that could be comparable to intels best. Two G4's weren't a real solution.
The second problem was that PPC pricing was moving up because Apple bought 75% of the G5's IBM made. That wasn't sustainable.
The last reason was that Apple wanted people to continue switching from Windows. But that is always difficult because of the need to buy new software—a major hurdle for most people, and even businesses. Moving to x86 meant that those moving could use Windows at full speed until they could wean themselves from their few required Windows programs. But part of that was also that it's easier to port apps over if the same chip is used. There's less work involved. So Apple figured that more software would come to the platform, and indeed, that's what happened.
As I say, this was a difficult issue, and psychology was part of it. Having "Intel Inside" was a comfort to many Windows people who wanted to move. We can never forget about the power of psychology.
As far as moving to ARM, it's possible. But Apple would need to ensure that all OS X software that now runs well on their current lowest end Macbook Air would run on this about as well. If they can make app development and use transparent, then it will work. If not, then it won't. They don't need to look at their top performances machines. They can stay on intel. But if Apple can shave off $100 from the cost of the now $899 machine, then that would be a win for them.if they could drop the price to $849, that would be a win. If they could go down to $799, that would generate many more sales. The lowering of prices across the line last year is credited with enhancing sales.
Right now, the A8X is better than the top Atom. It's at about the performance level of the bottom i3 low power. Apple uses the i5 low power as their weakest chip, so ARM has a way to go. But if Apple produced a four core A9, or A10 with close to double the GPU performance, and an additional 30% increase per CPU core, they would be very close to i5. You can look up the Geekbench scores to see that. If they also do what I've been suggesting, which is to add those few x86 instructions to the CPU that take up 80% of the emulation time, they would have that problem solved as well. That can be done. The OS could switch to them when a native x86 app asks for them. Experts I've spoken to say that it can be done.
I wouldn't put it past Apple to be seriously considering this.
I'm not sure sales numbers of Parallels Desktop and VMWare Fusion are available, but the purchase prices have dropped substantially, I've been offered both for free with several Mac sofware packages so sales numbers would be bogus anyway.
Apple's mac sales had been in the dumps before the Intel macs hit the stores. This was due to PPC chips falling further and further behind in the race with Intel. By switching to Intel, Apple's Mac products were once again as zippy as their Wintel rivals. In addition, Apple also offered a free program called, "Boot camp," to lessen the pain of anyone wanting to switch to Mac OS.
Sine then (and 2006 was a half-century ago in dog-years) Parallels Desktop and VMWare Fusion have made switching between platforms even less daunting. In addition multi-core computers are far more common today, and along with more RAM, make emulation a snap compared to 2006. None of this means users demand or want to run WIndows, it just means buying a Mac is an easier decision to make.
We are about two years away from iOS-based devices and computers outselling Wintel boxes... WIndows compatibly is a fading issue. Even Ballmer has moved on from offering a friendly reach around to Wintel users.
Apple's sales were increasing nicely before the switch. The fear at the time is that sales would slow down because of it. I remember the talk about that very well. What actually happened was that they sped up.
Our friend Rob Enderely, in an article that I chided him about, said that the move to x86 would cause the increasing Mac sales to "tank". I still bother him about that when we correspond.
I don't think they would do Intel emulation if they did switch to ARM (as others have said, for speed reasons). The ARM version of OS X would only run apps from the Mac App Store (same as iOS has always done). And the App Store would know to only display apps for download that had an ARM version in their fat binary.
Apple's sales were increasing nicely before the switch. The fear at the time is that sales would slow down because of it. I remember the talk about that very well. What actually happened was that they sped up.
Our friend Rob Enderely, in an article that I chided him about, said that the move to x86 would cause the increasing Mac sales to "tank". I still bother him about that when we correspond.
Thanks for the correction. I remember bits of that period of time and how the road looking forward for the RISC chips looked grim as Intel specs were on a meteoric climb. No one had switched, like Apple did, and survived it didn't look god for Apple at the time. I got my timeline a bit blurred.
I don't think they would do Intel emulation if they did switch to ARM (as others have said, for speed reasons). The ARM version of OS X would only run apps from the Mac App Store (same as iOS has always done). And the App Store would know to only display apps for download that had an ARM version in their fat binary.
I hardly see the point to it then. That's basically an iPad with keyboard.
I hardly see the point to it then. That's basically an iPad with keyboard.
What do you mean, it's still basically a Mac. You can run Terminal, open multiple programs in multiple windows, run Xcode, Final Cut Pro, Logic Pro, Photoshop Elements. The only problem would be if the particular program you want to run is not on the App Store, but presumably if Apple were to release such a Mac, they would lobby more software developers to put their apps on the App Store. And the developers themselves, seeing which way the wind is blowing, might also bite the bullet and upload their apps.
Comments
It seems to have been important to somebody:
Also, look at the sales numbers for Parallels Desktop and VMWare Fusion. If emulation of Windows weren't important to people, those products wouldn't be selling nearly as well as they are.
do yourself a favor and google correlation vs causation.
what are the sales numbers youre referring to? and who are they going to -- pros? enterprise? how does that relate to consumer-level macbooks and imacs?
Not happening.
Every time I see a story like this, all you need to do is look at the use-case, and it falls flat on it's face.
1) Existing A-series and ARM based parts are still Pentium 3-era speeds on comparable workloads
Very wrong of course. Try this test and you will see.
The latest Intel Atom is outperformed by a similar clocked A15.
I've been doing tests with the A8X. It can do 5 instructions per clock cycle like 2 loads and 3 SIMD.
The Intel Atom can decode only 2 instructions and in 64 bit or with SSE4 it can't even do that for a lot of instructions.
(never mind micro coded ones).
At the same clock speed the A8X is about 2.5 times faster as Atom.
It is on par with the latest big cores of _Intel.
The 9370 FX is for enthusiasts. The standard FX-8350 that I'm using is a TDP max of 125 W. The APU Excavator core of their last Bulldozer line is 30W-45W with a Radeon R9-300 series GPGPU SoC on it.
The Zen project is Keller's return and a complete move in a direction that will merge his experience creating Apple's A4 with his vast experience from DEC Alpha, AMD and more.
1) It's virtualization, not emulation. Yiu really should understand the difference as their distinction is highly relevant to computing.
2) With the change from PPC to Intel we got a major refresh to the overall HW and we got much faster speeds over the lackluster advancements coming from IBM. I know I was certainly debating on getting another PowerBook because of the shortcomings of the PPC CPU performavce.
Then there was a n iconic Get A Mac ad campaign. On top of that we had some major iPod changes around that time and both iTunes for Windows and the iTunes Music Store were still new to many Windows users but showing the level of quality Apple could bring to users.
Finally, there were plenty of cultural and general industry changes that helped sell "PCs". That's not to say that the option to run Windows on a Mac (which didn't arrive until mid-2006) didn't help sell some additional Macs, but there is no logical way you can say running Windows on a Mac is the reason.
Either Intel licenses it to Apple or Apple has to use DP for their external display support and forego TB on what would be a low-cost "PC" by Apple. I don't think TB is something people with sub $800 "PCs" care about.
With a change from Intel to ARM, we would get much slower speeds. Think about how your debating would have gone if Apple had switched the other direction at that time, from Intel to PPC, because performance-wise that's a better analogy here. Would you have upgraded at all?
If the entire Mac line (note: I am not discounting the possibility of a Chromebook-type ultra-low-end model that doesn't run native apps) were converted to Intel, I doubt many existing users would upgrade at all until their existing machines died, given that it will most likely be years before ARM even manages to catch up to Intel's current processors, never mind what Intel will have by then. What's the benefit? Saving a little bit of money on the parts? Running a bunch of apps designed for a tiny touch screen with a completely different UI paradigm which could easily be emulated on an Intel processor anyway? This is just describing the Surface RT, and look what that did to the perception of the Surface line. Heck, just look at the ads MS is putting out now, and how desperate they are to make you forget the Surface RT ever existed. They're all, "Hey, look at the Surface Pro! It's just as powerful as that Mac! No really, we swear! And you can run your Windows apps on it!" Do you really think Apple will want to follow this example?
Running Windows on a Mac isn't the reason. Switching to Intel was the reason. Apple's sales exploded immediately after the Intel switch, and you can't really write that off as just being due to a marketing campaign. The Intel switch clearly made a difference. It had major advantages, chief among which was performance. That being said, running Windows certainly is a reason. Just go in an Apple store sometime and see how prominently Parallels is displayed. Watch a sales pitch given to a potential PC switcher. Heck, look at the software section of the Apple Store web site. The first thing you see is MS Office. The second one is Parallels. This is a major Mac benefit, which brings lots of customers, and Apple would be insane to lose it.
Not happening.
Every time I see a story like this, all you need to do is look at the use-case, and it falls flat on it's face.
1) Existing A-series and ARM based parts are still Pentium 3-era speeds on comparable workloads
2) Intel's Atom parts are barely faster than Pentium 3-era parts
3) Intel's onboard video has always been a joke, but mobile isn't much better.
I could see Apple eventually producing a CPU/GPU that is more capable than Intel's weak mobile offerings, but Apple is unlikely to create an environment where they have to produce two versions of software for any extended period of time (see Motorola vs PPC, and PPC vs Intel.) Also look at how Microsoft fell on it's face with the Surface. Even if Apple produced a full version of OS X for an ARM based iMac or MacBook, the lack of being unable to just run every previous version of software will be felt and it will fail. Apple had an advantage earlier on when it switched from PPC to Intel because it was switching from a platform that only it was really using for desktops, so it could effectively push everyone off it. Not true with switching from x86 (look again at Microsoft, who also produced ARM, MIPS and ALPHA versions of Windows NT or CE at previous points in time.)
It's extremely unlikely that Apple would produce a part that is anywhere performance parity of the Xeon series in the Mac Pro.
There's also been some grumbling lately that OS X is getting too dumbed down that even people who normally like Apple are considering alternatives. Apple might be at a point in it's life where it might be more arrogant than Microsoft in making unnecessary changes to the operating system.
Exactly. The only reason our company has any Macs is cause they can natively run x86 software and other operating systems. If Apple were to move off of x86, there would no longer be any reason for us to buy Apple hardware.
I agree concerning the dumbing down of MacOS. It concerns me greatly. It's already hard enough to run MacOS in our organization, but as more and more basic fuctionality (interoperability with Windows & Linux, i.e. SMB) is broken in each release of MacOS, there may come a point where we abandon the platform entirely. Apple really needs to think through many of their recent design decisions, both HW and SW.
Can you provide some evidence that Apple's ARM designs would not be able to compete with any workload on an Intel chip with a comparable TDP?
I'm not sure sales numbers of Parallels Desktop and VMWare Fusion are available, but the purchase prices have dropped substantially, I've been offered both for free with several Mac sofware packages so sales numbers would be bogus anyway.
Apple's mac sales had been in the dumps before the Intel macs hit the stores. This was due to PPC chips falling further and further behind in the race with Intel. By switching to Intel, Apple's Mac products were once again as zippy as their Wintel rivals. In addition, Apple also offered a free program called, "Boot camp," to lessen the pain of anyone wanting to switch to Mac OS.
Sine then (and 2006 was a half-century ago in dog-years) Parallels Desktop and VMWare Fusion have made switching between platforms even less daunting. In addition multi-core computers are far more common today, and along with more RAM, make emulation a snap compared to 2006. None of this means users demand or want to run WIndows, it just means buying a Mac is an easier decision to make.
We are about two years away from iOS-based devices and computers outselling Wintel boxes... WIndows compatibly is a fading issue. Even Ballmer has moved on from offering a friendly reach around to Wintel users.
I was expecting this really appreciated.
On the plus side, if they did switch to ARM, we might get another "no new features" performance focussed release of OS X.
The releases seemed to have ramped up with the change of management - Forstall leaving, putting Federighi and Ive in his place. Now that they've implemented the larger changes, I hope they dial back on the OS releases and focus more on stability. I think it's right to push ahead with new software requirements but one thing they did recently was requiring that new apps in the App Store have to use the iOS 8 SDK by February 2015 despite only being released September 2014. That's only 5 months for people to get all new software updates and test them all and it prevents any OS pre-Mavericks being used for app development. The process keeps things moving but they should allow for reasonable update cycles. Businesses need stable software and updating to bleeding edge beta software isn't always feasible. OS releases every 2 years would be fine if there's nothing major to add to them and it makes the releases more interesting. Then they can gives software update requirements a year vs < 6 months.
Both products of course frequently participate in promotions in order to compete with each other and secure market share.
All of which would be negated by switching to ARM.
Ballmer hasn't been in charge of MS for almost a year.
Windows still controls something like 90% of the PC market. Windows compatibility is still a very big deal. There are many people out there that have to run Windows software for their jobs, since business is still largely Windows-based, and many of these companies have custom in-house software that isn't cross-platform. These requirements tend to be a bit non-negotiable for people who have these requirements, so losing Windows compatibility means losing those customers. There's also the gaming crowd, who also tend to want Windows for obvious reasons.
I think it would be quite cool to have a zillion ARM cores in a desktop or laptop. Potentially a lot more power than a single quad core intel chip.
However, having made prototypes means nothing about future plans. NeXT supposedly had a many-core PPC version running for a long time in the labs before there was any speculation at all of any tie-up with Apple. And Apple supposedly had classic MacOS running in the lab on intel for years but never did anything commercial with it. Companies go down various paths to provide alternative routes if things don't go the way they expect with their main plans, or to try stuff out, or otherwise get experience with alternatives and/or contingency plans.
I have speculated that core volume could be the differentiator.
Moore's law has slowed down with Intel and with all the companies making ARM based parts, certainly on of the big silicon players is going to make a 64core (or more) ARM based chip. And it will have more raw compute power than the biggest Intel chips, probably slower single thread speed though, at least for a generation or two. There will be interesting problems sets for that kind of hardware.
In the consumer market, are we post-PC? One disadvantage Apple had with PowerPC was that it was really challenging to do a direct comparison to a Windows Intel machine, there were a lot of variables and the mass consensus was that PowerPC would have an edge for a little while and then get left behind. Running the same hardware, its just user experience to compare; Photoshop will run just as fast on OS X as Windows.
Go to Anandtech. They just reviewed some of AMD's newest chips. Terrible, really. They hope that AMD will do well, so they try to give them the benefit of the doubt, but they're slowly dying. And yes, 220 watts is where their top chips are. Even the newest chip at 95 watts doesn't perform as well as a number of intels chips at half the power, for most tasks. And while the over locking is impressive, to get all the way up there requires taking the chip all the way to an astounding 260 watts.
Sorry, but this simply isn't acceptable. AMD was always a crap company, and it remains a crap company. They had almost three years in the sun when intel lost their way with the latter stages of Netburst. But then Intel turned around, and AMD has been left in the dust since with several crap designs. It's too late for them.
This is a difficult issue for everyone to discuss. Moving to x86 was a wrenching experience. Watching intel dissolve with Netburst, and then having Jobs come out on stage saying that performance/watts was why Apple was moving to x86 was like a dream. How could that be right? I assumed that he had seen things at intel that no one else knew about, except, possibly, some other big intel customers. After all, while the G5 was slightly behind x86 in performance, it was actually gaining on it. I estimated that in less than a year, it would surpass it in performance.
But Apple moved for two reasons. One major one, and quite possibly THE major problem at the time, was that laptop sales were exceeding Desktop sales, and that looked to be increasing. Apple had no "modern" solution for that. IBM had failed to deliver a mobile G5, and Apple needed a mobile chip that could be comparable to intels best. Two G4's weren't a real solution.
The second problem was that PPC pricing was moving up because Apple bought 75% of the G5's IBM made. That wasn't sustainable.
The last reason was that Apple wanted people to continue switching from Windows. But that is always difficult because of the need to buy new software—a major hurdle for most people, and even businesses. Moving to x86 meant that those moving could use Windows at full speed until they could wean themselves from their few required Windows programs. But part of that was also that it's easier to port apps over if the same chip is used. There's less work involved. So Apple figured that more software would come to the platform, and indeed, that's what happened.
As I say, this was a difficult issue, and psychology was part of it. Having "Intel Inside" was a comfort to many Windows people who wanted to move. We can never forget about the power of psychology.
As far as moving to ARM, it's possible. But Apple would need to ensure that all OS X software that now runs well on their current lowest end Macbook Air would run on this about as well. If they can make app development and use transparent, then it will work. If not, then it won't. They don't need to look at their top performances machines. They can stay on intel. But if Apple can shave off $100 from the cost of the now $899 machine, then that would be a win for them.if they could drop the price to $849, that would be a win. If they could go down to $799, that would generate many more sales. The lowering of prices across the line last year is credited with enhancing sales.
Right now, the A8X is better than the top Atom. It's at about the performance level of the bottom i3 low power. Apple uses the i5 low power as their weakest chip, so ARM has a way to go. But if Apple produced a four core A9, or A10 with close to double the GPU performance, and an additional 30% increase per CPU core, they would be very close to i5. You can look up the Geekbench scores to see that. If they also do what I've been suggesting, which is to add those few x86 instructions to the CPU that take up 80% of the emulation time, they would have that problem solved as well. That can be done. The OS could switch to them when a native x86 app asks for them. Experts I've spoken to say that it can be done.
I wouldn't put it past Apple to be seriously considering this.
Apple's sales were increasing nicely before the switch. The fear at the time is that sales would slow down because of it. I remember the talk about that very well. What actually happened was that they sped up.
Our friend Rob Enderely, in an article that I chided him about, said that the move to x86 would cause the increasing Mac sales to "tank". I still bother him about that when we correspond.
I don't think they would do Intel emulation if they did switch to ARM (as others have said, for speed reasons). The ARM version of OS X would only run apps from the Mac App Store (same as iOS has always done). And the App Store would know to only display apps for download that had an ARM version in their fat binary.
Thanks for the correction. I remember bits of that period of time and how the road looking forward for the RISC chips looked grim as Intel specs were on a meteoric climb. No one had switched, like Apple did, and survived it didn't look god for Apple at the time. I got my timeline a bit blurred.
I hardly see the point to it then. That's basically an iPad with keyboard.
I hardly see the point to it then. That's basically an iPad with keyboard.
What do you mean, it's still basically a Mac. You can run Terminal, open multiple programs in multiple windows, run Xcode, Final Cut Pro, Logic Pro, Photoshop Elements. The only problem would be if the particular program you want to run is not on the App Store, but presumably if Apple were to release such a Mac, they would lobby more software developers to put their apps on the App Store. And the developers themselves, seeing which way the wind is blowing, might also bite the bullet and upload their apps.