Using an "alternative" to Thunderbolt still means giving up on Thunderbolt.
Not that I imagine it would really matter to many users. The rate at which it's being adopted might be described as "glacial" and the cost "exorbitant." I suspect that for all but a very few super-power users, a USB3 world makes Thunderbolt a solution in search of a problem.
Mac Pro (Early 2013) will be run on two A6 chips. Clocked at a whopping 1.4GHz and utilizing a full 2GB of RAM, they'll be the powerhouse that no one wanted. But thin!
Joking aside, I do like that Apple's getting into chip design themselves. Years ago, I imagined that doing that would be the final step in truly optimizing a hardware-software ecosystem.
I agree with you accept Apple is depending on Samsung to produce the chips.
Apple's recent attempts at re-inventing the wheel haven't been exactly smooth so I'm afraid of a product that can't have its "growing pains" patched with software. (Speaking of that, has anyone else found that iMessage is not doing a very good job of staying in sync between Mac and iPhone?)
Besides, I imagine there's gotta be a big difference between designing a chip capable of running a telephone-level computer and one that can replace a Xeon. That doesn't mean they CAN'T, but might mean not YET. Then again, they didn't let lack of readiness stop them from releasing Maps, so who knows?
Apple doesn't reinvent the wheel. It imagines where the wheel will be in 10 years, and designs a wheel for that. Same wheel, just without all the cruft of 9 years of incremental improvements.
This started at NeXT. The fact they could pivot onto Intel as cleanly as they did, is a testament to their ability to think ahead and ponder how to make things work, and what they can't make work they say 'No' to. Look at Windows 8's feeble attempt to climb onto ARM with Surface. Compare to the fork in Mac OSX (which prior to PPC was ran on 4 processors in production at the SAME time... 68K, HP-PaRISC, Solaris, and Intel Pentium) and iOS,
The move will be easy, and my guess is Apple sees the power/performance curves of the custom ARM chip it 'can' make and the path Intel 'can make' and is posturing against Intel to 'step up and partner, or see us move off your chipset' (because Apple has shown them they can... 6 times in the past).
I'm figuring the former (partner up), but in order to 'negotiate' you gotta have a position of strength... custom ARM and the massive growth of iOS is that lever point.
The rate at which it's being adopted might be described as "glacial" and the cost "exorbitant."
I just bought 4 new Thunderbolt adapters at $30 each. A DVI, VGA, Firewire and Ethernet. So yeah, the computer is thin but you have to carry around the adapters which are bulky.
Apple doesn't reinvent the wheel. It imagines where the wheel will be in 10 years, and designs a wheel for that. Same wheel, just without all the cruft of 9 years of incremental improvements.
And for 9 years Apple users go without peripherals and apps from smaller vendors while they wait for the rest of the world to catch up! (j/k)
Mac Pro (Early 2013) will be run on two A6 chips. Clocked at a whopping 1.4GHz and utilizing a full 2GB of RAM, they'll be the powerhouse that no one wanted. But thin!
Joking aside, I do like that Apple's getting into chip design themselves. Years ago, I imagined that doing that would be the final step in truly optimizing a hardware-software ecosystem.
When apple went to intel, it enabled millions of windows users to comfortably move to mac because they knew they could run windows in a VM. Not sure if that's still the case but I do think it would scare some people.
I don't see a safe landing space for Apple here. ARM is years away from having consumer class hardware beyond the mobile space.
AMD could be purchased for a pittance but does Apple really want to get into head to head competition with Intel?
Lastly virtualization is very key to Apple products being sold in Enterprise. Not saying it cannot be done with Apple homegrown solutions but a different architecture makes that endeavor much harder.
Talk about a blade computer. Make it 1U height and rackable but have it come with feet on the bottom and a stand that lets it also run vertically as you have it and…
…oh, and drop the ODD entirely and…
…got yourself an interesting computer there. Though it would leave GPU to be desired.
I can see Apple doing this for MacBook Air. Didn't they already report that Air ran "surprisingly well" on ARM a while ago.
But they would incur a huge performance penalty if they did this across the board, making their products essentially non-competitive with similarly priced alternatives.
Re: "Such a shift would be difficult and isn't imminent, though it would allow Apple to further distinguish its laptops and desktops from competitors that run Intel's chips and Microsoft's Windows software," ...
Difficult? Yes. But Apple has mastered "difficult transitions" over the years. E.g. 68K processors to RISC, RISC to CISC (Intel), OS 9 to OS X, ad infinitum. If Apple had botched any of those transitions badly, they'd have died as a result.
Imminent? No. Not without a 64-bit ARM CPU. But the ARMv8 64-bit architecture spec was released in late 2011, it is a 64-bit architecture, and it will be able to run 32-bit apps. The 2014 iPhone could, in theory, use a 64-bit Apple-designed ARM SoC.
And so could 2014 iMacs. It's nearly inevitable. There are too many advantages to using ARM over Intel. No more boutique pricing for off-the-shelf Intel CPUs, which would bring Apple's MacBook Air pricing advantage to the iMac line. The CPU is one of the more expensive components of any computer, and if Apple can use its own ARM designs, Mac users will avoid The Intel Tax (tm). (Apple is a relatively small consumer of Intel CPUs, so they don't get the same deals they do for flash memory, touchscreens, etc.) Using ARM chips could potentially vastly reduce energy consumption and heat production, leading to reduced electrical and mechanical complexity (no fans, smaller power supplies etc.) Not to mention total safety from Intel-based malware.
Disadvantages? Very few. Sure, the 2014 Intel CPUs will be more powerful, but consumers won't care. The MHz / GHz race is over. Irrelevant. A 64-bit ARM chip would already be overkill for the typical emailing / browsing / Facebooking / Tweeting / FaceTiming / YouTubing consumer. If you want to do hypersonic aerodynamics simulations, get a Mac Pro (which could still run Intel chips for the few "pro" users that need performance at any cost.)
Oh, and the tired old "won't run Windows" argument is equally irrelevant. If you're locked into Windows and Office for some reason, you'll probably still be able to buy a Wintel PC. Even in 2014. We'll all pity you.
Mac Pro (Early 2013) will be run on two A6 chips. Clocked at a whopping 1.4GHz and utilizing a full 2GB of RAM, they'll be the powerhouse that no one wanted. But thin!
Joking aside, I do like that Apple's getting into chip design themselves. Years ago, I imagined that doing that would be the final step in truly optimizing a hardware-software ecosystem.
Actually, there is a server that Dell is playing around with that utilizes a bunch of card slots and on each card slot contains a bunch of ARM chips. The code name is Copper.
Think of what Apple could do with jamming in a bunch of high end ARM processors in the same space as the two XEONS in a MacPro, or in the space of a laptop or iMac? They could easily fit more of them inside since they don't generate the same level of heat and draw nearly as much power.
It wouldn't surprise me if Apple has some research labs of multi processor ARM laptops and desktops they are playing around with.
It would obviously be a lot easier for Apple to change to RISC processing before Microsoft due to the Windows having to be re-written for ARM, which they are doing with WIndows 8RT and the Windows phones.
Maybe good, but mostly a DIFFICULT read. The "democratization" of publishing seems to mean that people who have no business being writers try anyway. Knowing about a subject and being able to write about have always been two different things, and still are.
What I miss most is copy editors who caught most typing and grammar errors. AI could really use one, too.
I can testify for the validity of this report, as I've been informed by insider sources myself, however it's not Intel Apple plans to move from, it's the x86 architecture.
Some of you are mentioning the "fall back" aspect of current architecture for windows software... But remember... The next version of windows will in fact be ARM compatible.
Comments
Quote:
Originally Posted by BulletSpongeRTR
http://www.techpowerup.com/158480/AMD-Demonstrates-Trinity-APU-Its-Own-Thunderbolt-Alternative.html
Using an "alternative" to Thunderbolt still means giving up on Thunderbolt.
Not that I imagine it would really matter to many users. The rate at which it's being adopted might be described as "glacial" and the cost "exorbitant." I suspect that for all but a very few super-power users, a USB3 world makes Thunderbolt a solution in search of a problem.
Maybe a BTO version with an auxiliary Intel chip is the answer for those of us that want VMs.
Quote:
Originally Posted by Tallest Skil
Mac Pro (Early 2013) will be run on two A6 chips. Clocked at a whopping 1.4GHz and utilizing a full 2GB of RAM, they'll be the powerhouse that no one wanted. But thin!
Joking aside, I do like that Apple's getting into chip design themselves. Years ago, I imagined that doing that would be the final step in truly optimizing a hardware-software ecosystem.
I agree with you accept Apple is depending on Samsung to produce the chips.
Quote:
Originally Posted by v5v
Apple's recent attempts at re-inventing the wheel haven't been exactly smooth so I'm afraid of a product that can't have its "growing pains" patched with software. (Speaking of that, has anyone else found that iMessage is not doing a very good job of staying in sync between Mac and iPhone?)
Besides, I imagine there's gotta be a big difference between designing a chip capable of running a telephone-level computer and one that can replace a Xeon. That doesn't mean they CAN'T, but might mean not YET. Then again, they didn't let lack of readiness stop them from releasing Maps, so who knows?
Apple doesn't reinvent the wheel. It imagines where the wheel will be in 10 years, and designs a wheel for that. Same wheel, just without all the cruft of 9 years of incremental improvements.
This started at NeXT. The fact they could pivot onto Intel as cleanly as they did, is a testament to their ability to think ahead and ponder how to make things work, and what they can't make work they say 'No' to. Look at Windows 8's feeble attempt to climb onto ARM with Surface. Compare to the fork in Mac OSX (which prior to PPC was ran on 4 processors in production at the SAME time... 68K, HP-PaRISC, Solaris, and Intel Pentium) and iOS,
The move will be easy, and my guess is Apple sees the power/performance curves of the custom ARM chip it 'can' make and the path Intel 'can make' and is posturing against Intel to 'step up and partner, or see us move off your chipset' (because Apple has shown them they can... 6 times in the past).
I'm figuring the former (partner up), but in order to 'negotiate' you gotta have a position of strength... custom ARM and the massive growth of iOS is that lever point.
Quote:
Originally Posted by v5v
Thunderbolt...
The rate at which it's being adopted might be described as "glacial" and the cost "exorbitant."
I just bought 4 new Thunderbolt adapters at $30 each. A DVI, VGA, Firewire and Ethernet. So yeah, the computer is thin but you have to carry around the adapters which are bulky.
Quote:
Originally Posted by TheOtherGeoff
Apple doesn't reinvent the wheel. It imagines where the wheel will be in 10 years, and designs a wheel for that. Same wheel, just without all the cruft of 9 years of incremental improvements.
And for 9 years Apple users go without peripherals and apps from smaller vendors while they wait for the rest of the world to catch up! (j/k)
Quote:
Originally Posted by Tallest Skil
Mac Pro (Early 2013) will be run on two A6 chips. Clocked at a whopping 1.4GHz and utilizing a full 2GB of RAM, they'll be the powerhouse that no one wanted. But thin!
Joking aside, I do like that Apple's getting into chip design themselves. Years ago, I imagined that doing that would be the final step in truly optimizing a hardware-software ecosystem.
Quote:
Originally Posted by hmurchison
I don't see a safe landing space for Apple here. ARM is years away from having consumer class hardware beyond the mobile space.
AMD could be purchased for a pittance but does Apple really want to get into head to head competition with Intel?
Lastly virtualization is very key to Apple products being sold in Enterprise. Not saying it cannot be done with Apple homegrown solutions but a different architecture makes that endeavor much harder.
Once again spot on post. /endofthread
Originally Posted by Shunnabunich
*low whistle, slow nod*
Talk about a blade computer. Make it 1U height and rackable but have it come with feet on the bottom and a stand that lets it also run vertically as you have it and…
…oh, and drop the ODD entirely and…
…got yourself an interesting computer there. Though it would leave GPU to be desired.
But they would incur a huge performance penalty if they did this across the board, making their products essentially non-competitive with similarly priced alternatives.
Difficult? Yes. But Apple has mastered "difficult transitions" over the years. E.g. 68K processors to RISC, RISC to CISC (Intel), OS 9 to OS X, ad infinitum. If Apple had botched any of those transitions badly, they'd have died as a result.
Imminent? No. Not without a 64-bit ARM CPU. But the ARMv8 64-bit architecture spec was released in late 2011, it is a 64-bit architecture, and it will be able to run 32-bit apps. The 2014 iPhone could, in theory, use a 64-bit Apple-designed ARM SoC.
And so could 2014 iMacs. It's nearly inevitable. There are too many advantages to using ARM over Intel. No more boutique pricing for off-the-shelf Intel CPUs, which would bring Apple's MacBook Air pricing advantage to the iMac line. The CPU is one of the more expensive components of any computer, and if Apple can use its own ARM designs, Mac users will avoid The Intel Tax (tm). (Apple is a relatively small consumer of Intel CPUs, so they don't get the same deals they do for flash memory, touchscreens, etc.) Using ARM chips could potentially vastly reduce energy consumption and heat production, leading to reduced electrical and mechanical complexity (no fans, smaller power supplies etc.) Not to mention total safety from Intel-based malware.
Disadvantages? Very few. Sure, the 2014 Intel CPUs will be more powerful, but consumers won't care. The MHz / GHz race is over. Irrelevant. A 64-bit ARM chip would already be overkill for the typical emailing / browsing / Facebooking / Tweeting / FaceTiming / YouTubing consumer. If you want to do hypersonic aerodynamics simulations, get a Mac Pro (which could still run Intel chips for the few "pro" users that need performance at any cost.)
Oh, and the tired old "won't run Windows" argument is equally irrelevant. If you're locked into Windows and Office for some reason, you'll probably still be able to buy a Wintel PC. Even in 2014. We'll all pity you.
Quote:
Originally Posted by Tallest Skil
Mac Pro (Early 2013) will be run on two A6 chips. Clocked at a whopping 1.4GHz and utilizing a full 2GB of RAM, they'll be the powerhouse that no one wanted. But thin!
Joking aside, I do like that Apple's getting into chip design themselves. Years ago, I imagined that doing that would be the final step in truly optimizing a hardware-software ecosystem.
Actually, there is a server that Dell is playing around with that utilizes a bunch of card slots and on each card slot contains a bunch of ARM chips. The code name is Copper.
They are using Intel Atom chips.
Here is a story on the Dell ARM server.
http://gigaom.com/cloud/see-what-cloud-can-do-dell-unveils-arm-servers/
Think of what Apple could do with jamming in a bunch of high end ARM processors in the same space as the two XEONS in a MacPro, or in the space of a laptop or iMac? They could easily fit more of them inside since they don't generate the same level of heat and draw nearly as much power.
It wouldn't surprise me if Apple has some research labs of multi processor ARM laptops and desktops they are playing around with.
It would obviously be a lot easier for Apple to change to RISC processing before Microsoft due to the Windows having to be re-written for ARM, which they are doing with WIndows 8RT and the Windows phones.
Quote:
Originally Posted by UltraSPARC
Good read: http://semiaccurate.com/2012/09/24/an-update-on-apple-moving-away-from-intel/
Maybe good, but mostly a DIFFICULT read. The "democratization" of publishing seems to mean that people who have no business being writers try anyway. Knowing about a subject and being able to write about have always been two different things, and still are.
What I miss most is copy editors who caught most typing and grammar errors. AI could really use one, too.
Some of you are mentioning the "fall back" aspect of current architecture for windows software... But remember... The next version of windows will in fact be ARM compatible.
Just thought I would mention.