I guess you guys don't realize that the Tegra 2 is last year's chip. You are comparing the A5 to a chip from 2010. The next chip will be out in products in June. Code-named Kal-El, it is the first quad-core processor on the market. Five times faster than the Tegra 2.
That is THIS YEAR's chip. Cannot wait to see how the A5 stacks up.
Apple would be better served buying the chips from Nvidia. By 2014, Nvidia will have chips 100 times faster than Tegra 2. This is an arena Apple will not be able to keep up in.
Interesting how when it was the A4 vs Tegra and Snapdragon it was, hey you compare what you are shipping, the future is just so much vapor. But when the A5 comes out it flip flops to "I gues you guys don't realize that the Tegra 2 is last year's chip."
Opportunistic sour grapes is all your post is about.
And Nvidia has had a horrible time with power management. I guess you can take the engineers out of SGI, but you seemingly can't take the SGI out of the Nvidia lead engineers. I have no confidence they can maintain prominence in the marketplace, they have been showing dangerous engineering and business parallels to SGI for some time, and Huang's ego also cost them at least two to three years in the power management regime.
And down the road, it's not out of the realm of possibility for Apple to unify its Mac and iOS device hardware to all use some quad-or 6-core ARM variant. Especially for portables like the MacBook Air. Of course, that would require porting Mac OS X to the AX chip of the future. But don't forget that Mac OS X ran on RISC chips from the start. Been there, done that transition.
So far, I am skeptical of Mac OS X AppKit and iOS UIKit being merged completely. However, with more fanciful speculation ? if Apple ever builds a 4 or 6 core ARM SOC, I assume that GCD would make a significant impact within IOS. Another factor would be the increasing capabilities of LLVM to target different architectures, thus smoothing transitions between IA_64 and ARM. As mentioned above, the MACH-O file format allowed the switch between RISC and CISC CPUs in the past.
It is my belief that Apple keeps its options robust with regard to computer architectures.
Excellent points, but could you explain this one in more detail. I?m wondering why less metal layers would create a larger chip.
Very simplified analogy: Instead of building high density multi-store apartment buildings connected by skyways, you build suburban townhouses and single family homes. Same number of people, more streets and acreage necessary.
Domino, I agree with you. It's probably going to be a transition with more than a few bumps. As as you correctly hinted at, "normal" consumers don't care a twit about open, closed, chipsets or even OS for that matter. The bumps and bruises will be with the handset/tablet manufacturers. But apparently Google feels some control needs to happen if Android is going to continue it's somewhat surprising success. I think they're right. I don't know anyone who predicted they'd have this kind of mobile market presence this soon. Android was severely underestimated as a competitor. Still is. Google is a bit better developer than they've been given credit for going by market success. And evil? I still don't see any real evidence of it.
Apple bought PA Semi just for their chip design talent back in 2008. It cost Apple $278 million, and now their investment is finally starting to pay off. The A5 chip is the fastest mobile SoC on the market, it runs cool, it is economical with battery power, and it costs Apple far less than an off-the-shelf chip from, say, Intel.
We really dont know how much PA Semi tech made it into A5. Fact is we don't know much at all about the A5. It is however an obvious advantage for Apple.
Quote:
And down the road, it's not out of the realm of possibility for Apple to unify its Mac and iOS device hardware to all use some quad-or 6-core ARM variant.
I'd say that is extremely remote. What people seem to mis here are a few very important things that have made Mac acceptable. One is X86 compatibility. Like it or not being able to run Windows on a VM is a valuable thing. Second many of your Mac users are professionals and thus need fast machines. ARM is pathetically slow when put up against an Intel implementation. Third ARM is currently a 32 bit platform. For many common Mac apps, 32 bit is becoming useless.
However we should not dismiss consumer machines using ARM chips. However I can't see Apple calling these Macs.
Quote:
Especially for portables like the MacBook Air. Of course, that would require porting Mac OS X to the AX chip of the future. But don't forget that Mac OS X ran on RISC chips from the start. Been there, done that transition.
There is really little to port now. In any event I would imagine Apple jeeps Mac OS/X running on several architectures as a normal part of keeping the code base solid. Not to mention is the fact that the kernel and many of the libraries used on iOS are the same as those built for Mac OS/X.
Sadly people seem to think there is a huge gulf between the two OS's. That has really never been the case, each builds upon what is learned in the other. The differences there are to support entirely different usage patterns, security and marketing.
Quote:
If I were to make a silly wild-ass guess, I'd say that Apple will be quietly developing the RISC-based AX version of Mac OS quietly. They'll sit back and watch Microsoft botch their attempt to port Windows 8 to Tegra. It'll be buggy, the Tegra chip will run hot, and laptops running the combination will get poor battery life. And as we've seen, Microsoft has a poor track record of providing apps with backward compatibility. Two words: "XP Mode."
While I suspect that Mac OS/X already runs on ARM I see zero incentive for Apple to try to market such devices. It is far easier to grow iOS in a different direction and keep marketing simple. Right now the consumer has a clear distinction between iOS devices and Macs, apple really shouldn't muddy the waters.
Quote:
Apple could watch the Windows 8 + Tegra dumpster fire to get out of control, then drop the bomb. Mac OS 11 running on 4- and 6-core AX chips, Grand Central Dispatch balancing the load perfectly, MacBooks running as cool as iPads with enormously long battery life, apps running fat binaries just like they did in the 68k to PowerPC and the PowerPC to Intel transitions.
And running pathetically slow with little capability to tackle the bigger jobs. I just don't think you have an understanding of just how far behind ARM is performance wise, such machines would simply not meet the performance needs of many Mac users.
Again though I'm not dismissing that future iOS devices might have capability beyond today's iPad and target the consummer with simpler needs. Rather the point is why would Apple muddy up the marketing between the Macs and iOS devices. If or when Apple comes out with more iOS devices I can't see them going after the Mac market with them.
Look at it this way, the iPad is a massive success, however it hasn't impacted Mac sales at all yet. The reason in my mind is that they service different needs and markets.
Quote:
Should be fun to watch.
Well if it ever happens. The netbook market was a flash in the pan due to poor performance measured in a number of ways. I don't see ARM bringing anything special to this market.
And running pathetically slow with little capability to tackle the bigger jobs. I just don't think you have an understanding of just how far behind ARM is performance wise, such machines would simply not meet the performance needs of many Mac users.
Again though I'm not dismissing that future iOS devices might have capability beyond today's iPad and target the consummer with simpler needs. Rather the point is why would Apple muddy up the marketing between the Macs and iOS devices. If or when Apple comes out with more iOS devices I can't see them going after the Mac market with them.
Look at it this way, the iPad is a massive success, however it hasn't impacted Mac sales at all yet. The reason in my mind is that they service different needs and markets.
Good points all, well made, but consider this: computer processing power has been outstripping software needs for a while now. Sure, there are edge cases where more grunt is always better (video rendering, heavy duty image processing, 3D modeling, high end visualization) but, increasingly, the average Intel desktop/laptop now has vastly more power than the majority of the users of such machine will ever want or need.
That, in fact, is the basis of the success of the iPhone and the market it spawned, and the accelerating success of the iPad-- these type of devices have reached the point where they can do most of what most people want to do with computers.
The iPad is really the poster child for this reality: with software optimizations it can do the work of a Mac from about five years ago, and about five years ago was when desktop/laptop machines started to be as powerful as most users actually needed (outside of the kinds of games that even then were moving largely to consoles).
So it appears to me that most of the processing advantages of desktop/laptop processors are being piled on top of what is already overkill for most users, and that the new mobile chipsets are now moving comfortably into the sweet spot of being able to provide plenty of performance for most of what folks are doing with their computer.
So if Apple were to start running OS X machines on the Ax platform, they wouldn't be crippled machines for most people because most people aren't really using the power of what they have now in their MacBook Pros or Airs or Minis.
So I could see a bifurcation of the Mac line into Mac Pros retaining the Intel architecture and being sold to actual professionals with an actual need for as much processing power as possible, with the Mac line (including the Mini, MacBook and Air and possibly new models) moving to the Ax architecture and becoming extremely cost competitive with vanilla Wintel boxes.
Plus, Apple is obviously going to keep working on merging the iOS and OS X platforms, bringing ever more cross compatibility along the way. It's not hard to imagine a $300 Ax Miini (being even smaller) which runs OS X out of the box but has an iOS compatibility mode that includes drag and drop between apps. Or even more likely an Ax based Air like machine for $600 which satisfies the "I need a real OS" crowd while providing iPad size and weight and battery life.
A 40nm version of the A5 would be around 86mm^2 (122/sqrt(2)) - although different foundries will have different transistor densities anyway, so you can't compare directly. A 32nm chip would be 61mm^2, a 28nm chip (most likely for A6) 43mm^2 (although scaling isn't perfect either!).
Tegra 2 doesn't include Neon vector floating point. This is quite large, apparently. Tegra 3 will include it.
A high yielding mature process can allow larger dies to be made than those on a lower yielding process, for the same price.
Tegra 2 also doesn't have on-package memory, unlike the A5.
Forgive me if this has been addressed already, but is the same nanometer process required for the entire package or could Apple be saving total package size in other areas that are simply not cost effective for their competition due to economics of scale?
Good points all, well made, but consider this: computer processing power has been outstripping software needs for a while now. Sure, there are edge cases where more grunt is always better (video rendering, heavy duty image processing, 3D modeling, high end visualization) but, increasingly, the average Intel desktop/laptop now has vastly more power than the majority of the users of such machine will ever want or need.
That, in fact, is the basis of the success of the iPhone and the market it spawned, and the accelerating success of the iPad-- these type of devices have reached the point where they can do most of what most people want to do with computers.
The iPad is really the poster child for this reality: with software optimizations it can do the work of a Mac from about five years ago, and about five years ago was when desktop/laptop machines started to be as powerful as most users actually needed (outside of the kinds of games that even then were moving largely to consoles).
So it appears to me that most of the processing advantages of desktop/laptop processors are being piled on top of what is already overkill for most users, and that the new mobile chipsets are now moving comfortably into the sweet spot of being able to provide plenty of performance for most of what folks are doing with their computer.
So if Apple were to start running OS X machines on the Ax platform, they wouldn't be crippled machines for most people because most people aren't really using the power of what they have now in their MacBook Pros or Airs or Minis.
So I could see a bifurcation of the Mac line into Mac Pros retaining the Intel architecture and being sold to actual professionals with an actual need for as much processing power as possible, with the Mac line (including the Mini, MacBook and Air and possibly new models) moving to the Ax architecture and becoming extremely cost competitive with vanilla Wintel boxes.
Plus, Apple is obviously going to keep working on merging the iOS and OS X platforms, bringing ever more cross compatibility along the way. It's not hard to imagine a $300 Ax Miini (being even smaller) which runs OS X out of the box but has an iOS compatibility mode that includes drag and drop between apps. Or even more likely an Ax based Air like machine for $600 which satisfies the "I need a real OS" crowd while providing iPad size and weight and battery life.
I'd be really surprised to see anything like that. Any gap between the current iPad and the 11" MacBook air will be filled with more powerful iPads, not less powerful Macs that require all new universal binaries.
You do realize that this isn't 1995 and the integration of hardware and software needs to be taken as a whole to show real world performance?
Right?
Quote:
Apple would be better served buying the chips from Nvidia.
HAHAHAHA! That's rich... I have a sneaky suspicion Apple won't be switching their strategy any time soon.
And as an owner of a Late 2008 MBP that had a crappy Nvidia GPU that had to be replaced not just once, but twice, I won't miss them.
Quote:
By 2014, Nvidia will have chips 100 times faster than Tegra 2.
Ok - but how much will they cost? How much power will they use? How will they really perform? Since Apple owns the design of their chips, the more they make, the lower their per-chip cost gets since the cost of the intellectual property/design for them is a fixed cost, not a per-unit cost if they were buying from a provider like Nvidia.
They can also customize their design in ways that someone who is buying a simple commodity off-the-shelf chip simply can't.
Do you really think it's an accident that Apple is handily beating all of their tablet "competitors" on pricing?
Quote:
This is an arena Apple will not be able to keep up in.
From your keyboard to Jobs' screen. I guess time will tell. Something tells me Apple isn't worried.
I suppose some will honestly feel that the slightly higher resolution display and 16:9 aspect ratio on a TN panels are more important than a more universal 4:3 IPS panel on a tablet, or that their cameras are more awesomer despite being cameras on a tablet, or that it?s not to locked to iTunes and you don?t have to buy all your stuff from iTunes Store, or that Android is open and free? I said ?feel?, I didn?t say they would ?think?.
A couple weeks ago we were wondering if this larger chip could fit into the iPhone 4/G4 iPod Touch casing. I recall that it?s only slightly larger on the short access making it possible if they maneuvered some other chips on the board.
sheff?s remark that it could be for better cooling makes sense to me. Do we know how thick the A4 and A5 PoPs are compared to other chips? No one has yet made a logic board as dense as Apple?s iPhone 4 board. They are stacking chips on either side so perhaps thickness is more important than area in this case.
Memory size becomes a moot point with AirShare, I can access my entire iTunes library from a 16GB iPad and only need to carry around the essentials for $499.
I managed to pick up an ex demo iPad 16GB (first gen with a dock and a case) for $249 from a store which didn't know it's real value.
There's also 16GB iPad plus 20GB MobileMe storage at the same pricepoint you can have 36GB.
Tegra 2 doesn't include Neon vector floating point. This is quite large, apparently. Tegra 3 will include it.
This is a big deal. ARM VFP floating point SUCKS. Neon is quite good, as long as you stick with single precision, and for the demanding graphics apps where it counts, double precision is not generally used. I spent summer 2010 developing embedded image processing algorithms for an OMAP plaform. NEON is about 3x faster than VFP even if you don't know what you're doing or have a rubbish compiler. If you do know what you're doing, it's more like 8x.
I'd be really surprised to see anything like that. Any gap between the current iPad and the 11" MacBook air will be filled with more powerful iPads, not less powerful Macs that require all new universal binaries.
What you say is certainly true for the very near term (e.g., 2 years or so). However, what if we look further out. Will ARM be 64bit? 6 cores? 2 GHz clock? I think the point that is being made is that ARM is accelerating fast and the improvements in Intel chips are becoming less important. The issue isn't that intel can't keep up (they can). Intel's problem is that their increases in horsepower are becoming increasingly irrelevant to people's computing needs. At some point we may cross a line where ARM provides sufficient horsepower and Intel's advantage is irrelevant. Keep in mind that despite Intel's prowess, it has struggled to reduce power consumption. It also doesn't have a terribly good track record with SoC. Thus, there could very well be a time (maybe in the next 5 years) when ARM starts to replace chips in the mobile PC realm and potentially some day in consumer desktops.
I think the failure of ARM in Netbooks had a lot to do with 1. premature use of ARM (it still isn't there), 2. Lack of an OS optimized to take advantage of ARM 3. Microsoft using its marketing power to push WinXP. With MS out of the picture, an OS optimized for ARM, and more advanced ARM chips, we may see a different result.
Also, I think the major difference between iOS and OS X is touch vs. mouse optimization. You will never see OS X on a touch device and you will never see iOS on a Mac. However, there is no inherent reason why Apple will prefer one CPU architecture over another. Of course, switching isn't trivial, but Apple has shown that it can carry out a switch without a hitch. I think the real issue will always be a balance between computing power, power consumption, and cost. Right now Intel has such an advantage in the computing power (needed computing power) that it outweighs power consumption and cost in Macs. But things are changing fast........
Comments
Apple didn't build the A5 chip, Samsung built the A5.
Apple builds the A5, Samsung is just one of the contracted fabs. Contracted manufactring labor, nothing more.
I guess you guys don't realize that the Tegra 2 is last year's chip. You are comparing the A5 to a chip from 2010. The next chip will be out in products in June. Code-named Kal-El, it is the first quad-core processor on the market. Five times faster than the Tegra 2.
That is THIS YEAR's chip. Cannot wait to see how the A5 stacks up.
Apple would be better served buying the chips from Nvidia. By 2014, Nvidia will have chips 100 times faster than Tegra 2. This is an arena Apple will not be able to keep up in.
Interesting how when it was the A4 vs Tegra and Snapdragon it was, hey you compare what you are shipping, the future is just so much vapor. But when the A5 comes out it flip flops to "I gues you guys don't realize that the Tegra 2 is last year's chip."
Opportunistic sour grapes is all your post is about.
And Nvidia has had a horrible time with power management. I guess you can take the engineers out of SGI, but you seemingly can't take the SGI out of the Nvidia lead engineers. I have no confidence they can maintain prominence in the marketplace, they have been showing dangerous engineering and business parallels to SGI for some time, and Huang's ego also cost them at least two to three years in the power management regime.
And down the road, it's not out of the realm of possibility for Apple to unify its Mac and iOS device hardware to all use some quad-or 6-core ARM variant. Especially for portables like the MacBook Air. Of course, that would require porting Mac OS X to the AX chip of the future. But don't forget that Mac OS X ran on RISC chips from the start. Been there, done that transition.
So far, I am skeptical of Mac OS X AppKit and iOS UIKit being merged completely. However, with more fanciful speculation ? if Apple ever builds a 4 or 6 core ARM SOC, I assume that GCD would make a significant impact within IOS. Another factor would be the increasing capabilities of LLVM to target different architectures, thus smoothing transitions between IA_64 and ARM. As mentioned above, the MACH-O file format allowed the switch between RISC and CISC CPUs in the past.
It is my belief that Apple keeps its options robust with regard to computer architectures.
LOL I?d love to see a water-cooled smartphone.
Here you go:
Excellent points, but could you explain this one in more detail. I?m wondering why less metal layers would create a larger chip.
Very simplified analogy: Instead of building high density multi-store apartment buildings connected by skyways, you build suburban townhouses and single family homes. Same number of people, more streets and acreage necessary.
Apple bought PA Semi just for their chip design talent back in 2008. It cost Apple $278 million, and now their investment is finally starting to pay off. The A5 chip is the fastest mobile SoC on the market, it runs cool, it is economical with battery power, and it costs Apple far less than an off-the-shelf chip from, say, Intel.
We really dont know how much PA Semi tech made it into A5. Fact is we don't know much at all about the A5. It is however an obvious advantage for Apple.
And down the road, it's not out of the realm of possibility for Apple to unify its Mac and iOS device hardware to all use some quad-or 6-core ARM variant.
I'd say that is extremely remote. What people seem to mis here are a few very important things that have made Mac acceptable. One is X86 compatibility. Like it or not being able to run Windows on a VM is a valuable thing. Second many of your Mac users are professionals and thus need fast machines. ARM is pathetically slow when put up against an Intel implementation. Third ARM is currently a 32 bit platform. For many common Mac apps, 32 bit is becoming useless.
However we should not dismiss consumer machines using ARM chips. However I can't see Apple calling these Macs.
Especially for portables like the MacBook Air. Of course, that would require porting Mac OS X to the AX chip of the future. But don't forget that Mac OS X ran on RISC chips from the start. Been there, done that transition.
There is really little to port now. In any event I would imagine Apple jeeps Mac OS/X running on several architectures as a normal part of keeping the code base solid. Not to mention is the fact that the kernel and many of the libraries used on iOS are the same as those built for Mac OS/X.
Sadly people seem to think there is a huge gulf between the two OS's. That has really never been the case, each builds upon what is learned in the other. The differences there are to support entirely different usage patterns, security and marketing.
If I were to make a silly wild-ass guess, I'd say that Apple will be quietly developing the RISC-based AX version of Mac OS quietly. They'll sit back and watch Microsoft botch their attempt to port Windows 8 to Tegra. It'll be buggy, the Tegra chip will run hot, and laptops running the combination will get poor battery life. And as we've seen, Microsoft has a poor track record of providing apps with backward compatibility. Two words: "XP Mode."
While I suspect that Mac OS/X already runs on ARM I see zero incentive for Apple to try to market such devices. It is far easier to grow iOS in a different direction and keep marketing simple. Right now the consumer has a clear distinction between iOS devices and Macs, apple really shouldn't muddy the waters.
Apple could watch the Windows 8 + Tegra dumpster fire to get out of control, then drop the bomb. Mac OS 11 running on 4- and 6-core AX chips, Grand Central Dispatch balancing the load perfectly, MacBooks running as cool as iPads with enormously long battery life, apps running fat binaries just like they did in the 68k to PowerPC and the PowerPC to Intel transitions.
And running pathetically slow with little capability to tackle the bigger jobs. I just don't think you have an understanding of just how far behind ARM is performance wise, such machines would simply not meet the performance needs of many Mac users.
Again though I'm not dismissing that future iOS devices might have capability beyond today's iPad and target the consummer with simpler needs. Rather the point is why would Apple muddy up the marketing between the Macs and iOS devices. If or when Apple comes out with more iOS devices I can't see them going after the Mac market with them.
Look at it this way, the iPad is a massive success, however it hasn't impacted Mac sales at all yet. The reason in my mind is that they service different needs and markets.
Should be fun to watch.
Well if it ever happens. The netbook market was a flash in the pan due to poor performance measured in a number of ways. I don't see ARM bringing anything special to this market.
And running pathetically slow with little capability to tackle the bigger jobs. I just don't think you have an understanding of just how far behind ARM is performance wise, such machines would simply not meet the performance needs of many Mac users.
Again though I'm not dismissing that future iOS devices might have capability beyond today's iPad and target the consummer with simpler needs. Rather the point is why would Apple muddy up the marketing between the Macs and iOS devices. If or when Apple comes out with more iOS devices I can't see them going after the Mac market with them.
Look at it this way, the iPad is a massive success, however it hasn't impacted Mac sales at all yet. The reason in my mind is that they service different needs and markets.
Good points all, well made, but consider this: computer processing power has been outstripping software needs for a while now. Sure, there are edge cases where more grunt is always better (video rendering, heavy duty image processing, 3D modeling, high end visualization) but, increasingly, the average Intel desktop/laptop now has vastly more power than the majority of the users of such machine will ever want or need.
That, in fact, is the basis of the success of the iPhone and the market it spawned, and the accelerating success of the iPad-- these type of devices have reached the point where they can do most of what most people want to do with computers.
The iPad is really the poster child for this reality: with software optimizations it can do the work of a Mac from about five years ago, and about five years ago was when desktop/laptop machines started to be as powerful as most users actually needed (outside of the kinds of games that even then were moving largely to consoles).
So it appears to me that most of the processing advantages of desktop/laptop processors are being piled on top of what is already overkill for most users, and that the new mobile chipsets are now moving comfortably into the sweet spot of being able to provide plenty of performance for most of what folks are doing with their computer.
So if Apple were to start running OS X machines on the Ax platform, they wouldn't be crippled machines for most people because most people aren't really using the power of what they have now in their MacBook Pros or Airs or Minis.
So I could see a bifurcation of the Mac line into Mac Pros retaining the Intel architecture and being sold to actual professionals with an actual need for as much processing power as possible, with the Mac line (including the Mini, MacBook and Air and possibly new models) moving to the Ax architecture and becoming extremely cost competitive with vanilla Wintel boxes.
Plus, Apple is obviously going to keep working on merging the iOS and OS X platforms, bringing ever more cross compatibility along the way. It's not hard to imagine a $300 Ax Miini (being even smaller) which runs OS X out of the box but has an iOS compatibility mode that includes drag and drop between apps. Or even more likely an Ax based Air like machine for $600 which satisfies the "I need a real OS" crowd while providing iPad size and weight and battery life.
Tegra 2 doesn't include Neon vector floating point. This is quite large, apparently. Tegra 3 will include it.
A high yielding mature process can allow larger dies to be made than those on a lower yielding process, for the same price.
Tegra 2 also doesn't have on-package memory, unlike the A5.
Sorry if these points have already been made.
Good points all, well made, but consider this: computer processing power has been outstripping software needs for a while now. Sure, there are edge cases where more grunt is always better (video rendering, heavy duty image processing, 3D modeling, high end visualization) but, increasingly, the average Intel desktop/laptop now has vastly more power than the majority of the users of such machine will ever want or need.
That, in fact, is the basis of the success of the iPhone and the market it spawned, and the accelerating success of the iPad-- these type of devices have reached the point where they can do most of what most people want to do with computers.
The iPad is really the poster child for this reality: with software optimizations it can do the work of a Mac from about five years ago, and about five years ago was when desktop/laptop machines started to be as powerful as most users actually needed (outside of the kinds of games that even then were moving largely to consoles).
So it appears to me that most of the processing advantages of desktop/laptop processors are being piled on top of what is already overkill for most users, and that the new mobile chipsets are now moving comfortably into the sweet spot of being able to provide plenty of performance for most of what folks are doing with their computer.
So if Apple were to start running OS X machines on the Ax platform, they wouldn't be crippled machines for most people because most people aren't really using the power of what they have now in their MacBook Pros or Airs or Minis.
So I could see a bifurcation of the Mac line into Mac Pros retaining the Intel architecture and being sold to actual professionals with an actual need for as much processing power as possible, with the Mac line (including the Mini, MacBook and Air and possibly new models) moving to the Ax architecture and becoming extremely cost competitive with vanilla Wintel boxes.
Plus, Apple is obviously going to keep working on merging the iOS and OS X platforms, bringing ever more cross compatibility along the way. It's not hard to imagine a $300 Ax Miini (being even smaller) which runs OS X out of the box but has an iOS compatibility mode that includes drag and drop between apps. Or even more likely an Ax based Air like machine for $600 which satisfies the "I need a real OS" crowd while providing iPad size and weight and battery life.
I'd be really surprised to see anything like that. Any gap between the current iPad and the 11" MacBook air will be filled with more powerful iPads, not less powerful Macs that require all new universal binaries.
Cannot wait to see how the A5 stacks up.
You do realize that this isn't 1995 and the integration of hardware and software needs to be taken as a whole to show real world performance?
Right?
Apple would be better served buying the chips from Nvidia.
HAHAHAHA! That's rich... I have a sneaky suspicion Apple won't be switching their strategy any time soon.
And as an owner of a Late 2008 MBP that had a crappy Nvidia GPU that had to be replaced not just once, but twice, I won't miss them.
By 2014, Nvidia will have chips 100 times faster than Tegra 2.
Ok - but how much will they cost? How much power will they use? How will they really perform? Since Apple owns the design of their chips, the more they make, the lower their per-chip cost gets since the cost of the intellectual property/design for them is a fixed cost, not a per-unit cost if they were buying from a provider like Nvidia.
They can also customize their design in ways that someone who is buying a simple commodity off-the-shelf chip simply can't.
Do you really think it's an accident that Apple is handily beating all of their tablet "competitors" on pricing?
This is an arena Apple will not be able to keep up in.
From your keyboard to Jobs' screen. I guess time will tell. Something tells me Apple isn't worried.
From my own limited knowledge, it would appear the only differences between the chips are:
1) dual-core PowerVR GPU in the Apple chip (vs Geforce ULV)
2) Implementation of SIMD NEON engine (Tegra doesn't implement NEON)
3) Memory controller differences? Are they both single-channel LPDDR2?
What else would cause the huge size difference?
What else would cause the huge size difference?
Power management?
You can now buy the WiFi version of the 32GB Xoom for $599, matching Apple?s 32GB iPad price point.I suppose some will honestly feel that the slightly higher resolution display and 16:9 aspect ratio on a TN panels are more important than a more universal 4:3 IPS panel on a tablet, or that their cameras are more awesomer despite being cameras on a tablet, or that it?s not to locked to iTunes and you don?t have to buy all your stuff from iTunes Store, or that Android is open and free? I said ?feel?, I didn?t say they would ?think?.
A couple weeks ago we were wondering if this larger chip could fit into the iPhone 4/G4 iPod Touch casing. I recall that it?s only slightly larger on the short access making it possible if they maneuvered some other chips on the board.
sheff?s remark that it could be for better cooling makes sense to me. Do we know how thick the A4 and A5 PoPs are compared to other chips? No one has yet made a logic board as dense as Apple?s iPhone 4 board. They are stacking chips on either side so perhaps thickness is more important than area in this case.They?ve come a long way.
Memory size becomes a moot point with AirShare, I can access my entire iTunes library from a 16GB iPad and only need to carry around the essentials for $499.
I managed to pick up an ex demo iPad 16GB (first gen with a dock and a case) for $249 from a store which didn't know it's real value.
There's also 16GB iPad plus 20GB MobileMe storage at the same pricepoint you can have 36GB.
Tegra 2 doesn't include Neon vector floating point. This is quite large, apparently. Tegra 3 will include it.
This is a big deal. ARM VFP floating point SUCKS. Neon is quite good, as long as you stick with single precision, and for the demanding graphics apps where it counts, double precision is not generally used. I spent summer 2010 developing embedded image processing algorithms for an OMAP plaform. NEON is about 3x faster than VFP even if you don't know what you're doing or have a rubbish compiler. If you do know what you're doing, it's more like 8x.
I'd be really surprised to see anything like that. Any gap between the current iPad and the 11" MacBook air will be filled with more powerful iPads, not less powerful Macs that require all new universal binaries.
What you say is certainly true for the very near term (e.g., 2 years or so). However, what if we look further out. Will ARM be 64bit? 6 cores? 2 GHz clock? I think the point that is being made is that ARM is accelerating fast and the improvements in Intel chips are becoming less important. The issue isn't that intel can't keep up (they can). Intel's problem is that their increases in horsepower are becoming increasingly irrelevant to people's computing needs. At some point we may cross a line where ARM provides sufficient horsepower and Intel's advantage is irrelevant. Keep in mind that despite Intel's prowess, it has struggled to reduce power consumption. It also doesn't have a terribly good track record with SoC. Thus, there could very well be a time (maybe in the next 5 years) when ARM starts to replace chips in the mobile PC realm and potentially some day in consumer desktops.
I think the failure of ARM in Netbooks had a lot to do with 1. premature use of ARM (it still isn't there), 2. Lack of an OS optimized to take advantage of ARM 3. Microsoft using its marketing power to push WinXP. With MS out of the picture, an OS optimized for ARM, and more advanced ARM chips, we may see a different result.
Also, I think the major difference between iOS and OS X is touch vs. mouse optimization. You will never see OS X on a touch device and you will never see iOS on a Mac. However, there is no inherent reason why Apple will prefer one CPU architecture over another. Of course, switching isn't trivial, but Apple has shown that it can carry out a switch without a hitch. I think the real issue will always be a balance between computing power, power consumption, and cost. Right now Intel has such an advantage in the computing power (needed computing power) that it outweighs power consumption and cost in Macs. But things are changing fast........