I think this transition will be done for their consumer products only. Most consumers nowadays just buy MacBook Pros as their laptop even though they don't need the raw speed it offers. They don't really need those special intel instruction sets and they don't need CPU enhancements for multimedia encoding and decoding.
They want battery life. They want a snappy interface. They want to run Safari/Firefox/Chrome. They want to write in Word/Excel/PowerPoint. They want to use full fledged iLife.
And ARM + GPU + good hardware acceleration can offer this better than x86-64.
The AppStore creates a great way to cover ther architectural differences. Buy it once, and automatically buy it for both.
I am all for keeping the MacBook Pro a Pro-machine with Intel processors and all bells an whistles. But ARM could be big for the consumer space.
Bingo. An ARM-based MacBook Air is unlikely in 2012 but could debut in 2013. Boasting great all-round performance, compatibility with iOS and OSX apps, with unbelievable battery life. It could be a hybrid touch/keyboard device.
On the point of "CPU enhancements for multimedia encoding and decoding", there's a lot of that already in the iPhone and iPad. Hence the Garageband and iMovie apps for iOS. Not to mention all the hardware decoding of audio and video that iOS devices have been doing for years.
I'll be buying a new laptop for Logic Pro and some graphic design work. Is there any difference in processing power between the current and/or upcoming MBA and current MBP? About the same? I'm hoping the new MBA has a backlit keyboard at least.
You'll definitely need a MBP, IMO. Even if the new MBA has Sandy Bridge it will be half the processing power of the MBPs. It may have a backlit keyboard.
I'm not sure about you but Logic Pro and the way it's used now with tons of plugins means you need a beefy processor. Are you going to connect any Firewire devices or drives?
I'm a little old-skool and err on the side of caution.
By the way If you do get a MBP, be sure to get a SSD or at least a 7200rpm hard disk. Makes a distinct difference.
This is probably true, but it does not mean much. Apple has kept its OS multi-architecture, even during the days they only sold PPC. The fact they they have built this does not mean at all they are going to produce it. They might have just gone further than usual in their testing (e.g. test including all the normal behaviour that comes with it, not just the lab and for that they need a believable housing).
But, except for stuff like Parallels to use Windows, it would run just fine. And what is more,if they do this, instead of A Parallels x86 VM you could have iOS running on an ARM VM. You could run your iOS apps on your Mac. But that would require a touch interface. Maybe this is some sort of Mac OS X / iOS future OS test project.
I never really understood why people would want to use a notebook/laptop to do video work. Personally I need a big desk for my camera, decks, external drive, breakout, playback monitor, a nice big mouse pad, tons of cables, a big ergonomic chair, custom keyboard, and at least one 30" cinema display. I look at laptops as portable devices that come with a lot of compromises. Video editing is one of those disciplines, like cad or desktop publishing, that requires an office, in my opinion.
I disagree that CAD requires an office. When we started using CAD the desktop systems had 15"CRTs, then 17", then 19", then 21" ... Now we have 22" dual LCDs, but: today's portable computers outperform yesterday's desktop systems many times. For a while software was throttled by hardware performance, but today I can use CAD quite productively on a laptop. The difference has become negligible, but the advantages of having my entire office with me wherever I go are immense. BIM 3d modeling software still requires desktop class hardware, but I am confident it won't be long before portable hardware catches up to software demands (I'm talking low single digit years).
There are much more Developers that developed for either iOS or iOS AND MacOSX. Porting is a lot easier then compare to PowerPC times.
When the timing is right, that is Software, API, Library, LLVM Compilers, and Powerful enough ARM Core, Apple will switch. Not to mention save average 200+ USD per Mac when using ARM Core.
It still does depend on Microsoft also compiling Office:Mac for ARM for it to be feasible. Now Microsoft has announced (yeah, I know, an *announcement* by Microsoft is worth next to nothing) to get Windows (and supposedly Office) on ARM. But the big question would remain: would Microsoft deliver Office:Mac for ARM? Or would Apple need to deliver a Rosetta-like solution?
I'd be the first to say it is less than ideal for big projects where big monitors do make a difference. The thing is CAD users have a wide array of interests or requirements and for many a modern laptop is fine.
Quote:
Originally Posted by toes
I disagree that CAD requires an office. When we started using CAD the desktop systems had 15"CRTs, then 17", then 19", then 21" ... Now we have 22" dual LCDs, but: today's portable computers outperform yesterday's desktop systems many times. For a while software was throttled by hardware performance, but today I can use CAD quite productively on a laptop. The difference has become negligible, but the advantages of having my entire office with me wherever I go are immense. BIM 3d modeling software still requires desktop class hardware, but I am confident it won't be long before portable hardware catches up to software demands (I'm talking low single digit years).
The flip side here is that high end software will continue to increase in capacity and thus demand bleeding edge hardware for sometime. Conversly more and more technical users will fine laptops to be good enough.
Especially with people realizing they end up on a MicroSoft upgrade treadmill for no good reason.
Quote:
Originally Posted by gctwnl
It still does depend on Microsoft also compiling Office:Mac for ARM for it to be feasible. Now Microsoft has announced (yeah, I know, an *announcement* by Microsoft is worth next to nothing) to get Windows (and supposedly Office) on ARM. But the big question would remain: would Microsoft deliver Office:Mac for ARM? Or would Apple need to deliver a Rosetta-like solution?
Rosetta isn't going tomcat it! ARM simply doesn't have the power to do emulation ins a user acceptable manner. At least no at this point in time.
I'd be very interested in finding out just how manynMac owners actually us Office on the Mac. I doubt that many actually use Office all that much, even the people that have it installed. I don't see office as a big deal compared to being able to run Windows itself in a VM. Being able to run Windows native apps in a VM is a huge Mac selling point. It gives you access to software that will never run on a Mac, ARM or not.
But it will be interesting if Apple strips out all the unnecessary fluff in the AppleTV - removes the WiFi capability, HDMI capability, etc., replaces the A4 with the A5, increases RAM to 4GB, increases Flash to 64GB, and loads it with Lion Server. And possibly add Thunderbolt port for interfacing with external storage, etc.
Mac Mini has proven itself as a Server solution - this could be an even better option for some uses. At a single stroke, Apple can create the cheapest server solution for enterprise. And at the server level, Apple is not hostage to 3rd party developers as much as it is in the desktop space. Because of Linux, most server software is open source and can simply be recompiled to run on this server.
Mac Micro anyone? Would you buy it for $399?
Apple already has the technology in Grand Central Dispatch to string together several of these Mac Micro servers, so that it becomes an easily scaleable, low cost, low power, low footprint server solution.
I can easily imagine render farms, web hosting providers, etc. jumping to this product eagerly.
This would also be an easy way for Apple to test the waters, before jumping in with ARM based desktops. 2-3 years later, the performance and availability of apps would be good enough that Apple can introduce ARM based laptop and desktop Macs without serious disruption.
I disagree that CAD requires an office. When we started using CAD the desktop systems had 15"CRTs, then 17", then 19", then 21" ... Now we have 22" dual LCDs, but: today's portable computers outperform yesterday's desktop systems many times. For a while software was throttled by hardware performance, but today I can use CAD quite productively on a laptop. The difference has become negligible, but the advantages of having my entire office with me wherever I go are immense. BIM 3d modeling software still requires desktop class hardware, but I am confident it won't be long before portable hardware catches up to software demands (I'm talking low single digit years).
To each his own. For me CAD requires a full keyboard without having to use the fn key. Also I need the numeric key pad as well as a mouse. I'm not referring to power of the cpu when I stated that CAD is best done in an office, it was more the environment. Laptops have terrible ergonomics and the concentration required to to work on a CAD project is not something you normally associate with in a temporary, transitory, or remote location. Like I said, I can't understand why anyone would choose to use a laptop for that type of work. Not saying it is impossible. Furthermore is it really worth spending a CAD license on a compromised laptop?
I think that right there is enough to discredit the rumor. Thunderbolt is welded to Intel's x86 chipset.
I can see how Intel's ThunderBolt chip might be welded to its chipset because it's Intel, but technically ThunderBolt just melds DisplayPort and PCIe x4 lanes onto the DisplayPort cable, and thus should work with any PCIe x4 lanes and DisplayPort signal.
And here we see the issue - since when did Apple's A5 processor include a PCIe interface?
So this prototype would have required custom silicon - an A5 processor, presumably one with off-package DRAM (to get 2, 4 GB capacities) as well as PCIe support. Given the multi-million cost of masks, that's an awful lot to spend on a prototype.
I'd far rather believe that the A6 includes PCIe for this type of role, or that Apple will be making two types of A6 - one for phones (no PCIe, dual-core, on-package DRAM) and one for tablets and Airs (PCIe, external DRAM bus, PCIe bus, SATA, more USB, etc). Apple would have test chips back by now, so these could be used in prototypes already.
Comments
I think this transition will be done for their consumer products only. Most consumers nowadays just buy MacBook Pros as their laptop even though they don't need the raw speed it offers. They don't really need those special intel instruction sets and they don't need CPU enhancements for multimedia encoding and decoding.
They want battery life. They want a snappy interface. They want to run Safari/Firefox/Chrome. They want to write in Word/Excel/PowerPoint. They want to use full fledged iLife.
And ARM + GPU + good hardware acceleration can offer this better than x86-64.
The AppStore creates a great way to cover ther architectural differences. Buy it once, and automatically buy it for both.
I am all for keeping the MacBook Pro a Pro-machine with Intel processors and all bells an whistles. But ARM could be big for the consumer space.
Bingo. An ARM-based MacBook Air is unlikely in 2012 but could debut in 2013. Boasting great all-round performance, compatibility with iOS and OSX apps, with unbelievable battery life. It could be a hybrid touch/keyboard device.
On the point of "CPU enhancements for multimedia encoding and decoding", there's a lot of that already in the iPhone and iPad. Hence the Garageband and iMovie apps for iOS. Not to mention all the hardware decoding of audio and video that iOS devices have been doing for years.
I'll be buying a new laptop for Logic Pro and some graphic design work. Is there any difference in processing power between the current and/or upcoming MBA and current MBP? About the same? I'm hoping the new MBA has a backlit keyboard at least.
You'll definitely need a MBP, IMO. Even if the new MBA has Sandy Bridge it will be half the processing power of the MBPs. It may have a backlit keyboard.
I'm not sure about you but Logic Pro and the way it's used now with tons of plugins means you need a beefy processor. Are you going to connect any Firewire devices or drives?
I'm a little old-skool and err on the side of caution.
By the way If you do get a MBP, be sure to get a SSD or at least a 7200rpm hard disk. Makes a distinct difference.
But, except for stuff like Parallels to use Windows, it would run just fine. And what is more,if they do this, instead of A Parallels x86 VM you could have iOS running on an ARM VM. You could run your iOS apps on your Mac. But that would require a touch interface. Maybe this is some sort of Mac OS X / iOS future OS test project.
I think this direction is inevitable, but I'm not holding my breath until I see iOS emulation running on OS-X (or vis a vis).
You've... missed this for the past few years, then?
Also, it's vice versa. Vis a vis means "face to face".
Also, not my screenshot. I don't have such nonsense in my Dock.
I never really understood why people would want to use a notebook/laptop to do video work. Personally I need a big desk for my camera, decks, external drive, breakout, playback monitor, a nice big mouse pad, tons of cables, a big ergonomic chair, custom keyboard, and at least one 30" cinema display. I look at laptops as portable devices that come with a lot of compromises. Video editing is one of those disciplines, like cad or desktop publishing, that requires an office, in my opinion.
I disagree that CAD requires an office. When we started using CAD the desktop systems had 15"CRTs, then 17", then 19", then 21" ... Now we have 22" dual LCDs, but: today's portable computers outperform yesterday's desktop systems many times. For a while software was throttled by hardware performance, but today I can use CAD quite productively on a laptop. The difference has become negligible, but the advantages of having my entire office with me wherever I go are immense. BIM 3d modeling software still requires desktop class hardware, but I am confident it won't be long before portable hardware catches up to software demands (I'm talking low single digit years).
When the timing is right, that is Software, API, Library, LLVM Compilers, and Powerful enough ARM Core, Apple will switch. Not to mention save average 200+ USD per Mac when using ARM Core.
I disagree that CAD requires an office. When we started using CAD the desktop systems had 15"CRTs, then 17", then 19", then 21" ... Now we have 22" dual LCDs, but: today's portable computers outperform yesterday's desktop systems many times. For a while software was throttled by hardware performance, but today I can use CAD quite productively on a laptop. The difference has become negligible, but the advantages of having my entire office with me wherever I go are immense. BIM 3d modeling software still requires desktop class hardware, but I am confident it won't be long before portable hardware catches up to software demands (I'm talking low single digit years).
The flip side here is that high end software will continue to increase in capacity and thus demand bleeding edge hardware for sometime. Conversly more and more technical users will fine laptops to be good enough.
It still does depend on Microsoft also compiling Office:Mac for ARM for it to be feasible. Now Microsoft has announced (yeah, I know, an *announcement* by Microsoft is worth next to nothing) to get Windows (and supposedly Office) on ARM. But the big question would remain: would Microsoft deliver Office:Mac for ARM? Or would Apple need to deliver a Rosetta-like solution?
Rosetta isn't going tomcat it! ARM simply doesn't have the power to do emulation ins a user acceptable manner. At least no at this point in time.
I'd be very interested in finding out just how manynMac owners actually us Office on the Mac. I doubt that many actually use Office all that much, even the people that have it installed. I don't see office as a big deal compared to being able to run Windows itself in a VM. Being able to run Windows native apps in a VM is a huge Mac selling point. It gives you access to software that will never run on a Mac, ARM or not.
Mac Mini has proven itself as a Server solution - this could be an even better option for some uses. At a single stroke, Apple can create the cheapest server solution for enterprise. And at the server level, Apple is not hostage to 3rd party developers as much as it is in the desktop space. Because of Linux, most server software is open source and can simply be recompiled to run on this server.
Mac Micro anyone? Would you buy it for $399?
Apple already has the technology in Grand Central Dispatch to string together several of these Mac Micro servers, so that it becomes an easily scaleable, low cost, low power, low footprint server solution.
I can easily imagine render farms, web hosting providers, etc. jumping to this product eagerly.
This would also be an easy way for Apple to test the waters, before jumping in with ARM based desktops. 2-3 years later, the performance and availability of apps would be good enough that Apple can introduce ARM based laptop and desktop Macs without serious disruption.
I disagree that CAD requires an office. When we started using CAD the desktop systems had 15"CRTs, then 17", then 19", then 21" ... Now we have 22" dual LCDs, but: today's portable computers outperform yesterday's desktop systems many times. For a while software was throttled by hardware performance, but today I can use CAD quite productively on a laptop. The difference has become negligible, but the advantages of having my entire office with me wherever I go are immense. BIM 3d modeling software still requires desktop class hardware, but I am confident it won't be long before portable hardware catches up to software demands (I'm talking low single digit years).
To each his own. For me CAD requires a full keyboard without having to use the fn key. Also I need the numeric key pad as well as a mouse. I'm not referring to power of the cpu when I stated that CAD is best done in an office, it was more the environment. Laptops have terrible ergonomics and the concentration required to to work on a CAD project is not something you normally associate with in a temporary, transitory, or remote location. Like I said, I can't understand why anyone would choose to use a laptop for that type of work. Not saying it is impossible. Furthermore is it really worth spending a CAD license on a compromised laptop?
I think that right there is enough to discredit the rumor. Thunderbolt is welded to Intel's x86 chipset.
I can see how Intel's ThunderBolt chip might be welded to its chipset because it's Intel, but technically ThunderBolt just melds DisplayPort and PCIe x4 lanes onto the DisplayPort cable, and thus should work with any PCIe x4 lanes and DisplayPort signal.
And here we see the issue - since when did Apple's A5 processor include a PCIe interface?
So this prototype would have required custom silicon - an A5 processor, presumably one with off-package DRAM (to get 2, 4 GB capacities) as well as PCIe support. Given the multi-million cost of masks, that's an awful lot to spend on a prototype.
I'd far rather believe that the A6 includes PCIe for this type of role, or that Apple will be making two types of A6 - one for phones (no PCIe, dual-core, on-package DRAM) and one for tablets and Airs (PCIe, external DRAM bus, PCIe bus, SATA, more USB, etc). Apple would have test chips back by now, so these could be used in prototypes already.