Apple, ARM, and Intel

Hugh Pickens writes "Jean-Louis Gassée says Apple and Samsung are engaged in a knives-out smartphone war. But when it comes to chips, the two companies must pretend to be civil because Samsung is the sole supplier of ARM-based processors for the iPhone. So why hasn't Intel jumped at the chance to become Apple's ARM source? 'The first explanation is architectural disdain,' writes Gassée. 'Intel sees "no future for ARM," it's a culture of x86 true believers. And they have a right to their conviction: With each iteration of its manufacturing technology, Intel has full control over how to improve its processors.' Next is pride. Intel would have to accept Apple's design and 'pour' it into silicon — it would become a lowlymerchant foundry.' Intel knows how to design and manufacture standard parts, but it has little experience manufacturing other people's custom designs or pricing them. But the most likely answer to the Why-Not-Intel question is money. Intel meticulously tunes the price points for its processors to generate the revenue that will fund development. Intel's published prices range from a 'low' $117 for a Core i3 processor to $999 for a top-of-the-line Core i7 device. Compare this to iSuppli's estimate for the cost of the A6 processor: $17.50. Even if more A6 chips could be produced per wafer — an unproven assumption — Intel's revenue per A6 wafer start would be much lower than with their x86 microprocessors. In Intel's perception of reality, this would destroy the business model. 'For all of Intel's semiconductor design and manufacturing feats, its processors suffer from a genetic handicap: They have to support the legacy x86 instruction set, and thus they're inherently more complicated than legacy-free ARM devices, they require more transistors, more silicon. Intel will argue, rightly, that they'll always be one technological step ahead of the competition, but is one step enough for x86 chips to beat ARM microprocessors?'"

Comments

Long term

[nurb432]

ARM will win the battle, as long as we dont see some sort of bloody patent war.

Its a better archetecture from the start.

Re:Long term

[Anonymous Coward]

Sure it is. That's why the best ARM CPU is hundreds of times slower than the best x86 CPU.

Re:Long term

[Anonymous Coward]

Care to explain why it is better?

Re:Long term

[Marillion]

If you measure operations per second, the x86 chip will win. If you measure operations per second per watt, the ARM chip will win.

Re:Long term

[rahvin112]

What a bold prediction, you understand of course that Intel has buried every single competing architecture from the past? Intel has a process advantage, even if they have to spend 10% of their die on decoding/rearranging they still have a significant transistor lead by remaining a process ahead AND still use lower power. Not only that but because x86 is nothing more than an abstraction layer at this point the internal architecture of their chips is free to move with the winds of computing in the best direction for the balance of power use, processing capacity and weight. They've had almost 2 decades to improve this abstraction layer to the point of perfection.

People like you forget how long it takes to design and build a microprocessor. From design to hard silicon is almost 5 years. So the designs Intel releases this year were planned out in 2007. Given the ARM didn't start to make an impact (on Markets Intel considers themselves part of) until 2006-7 we are JUST starting to see an Intel design philosophy that emphasizes power as a critical function. Haswell is probably the first chip that Intel hasn't tried to tack power efficiency on add-on at tape out. I fully expect Intel to demonstrate that x86 under their lead has the ability to compete directly with ARM on their best footing, power consumption.

So watch and learn young padawan. Intel has the best process engineers in the business and if things in the foundry business keep going like they are (TSMC and Global Foundaries have both been very very late moving forward on process while Intel hasn't missed a stride) they are going to be two steps ahead on process in the next year or two and that would be an advantage not even the best ARM design could beat even if Intel bungles their design. I fully expect that if Intel wants it they could take the whole ARM chip market. The only reason they haven't up till now is it would destroy their margins. So we will watch them balance their designs to retain the high margin products and forgo the cheap. This could ultimately be their undoing but once power efficiency becomes a priority of their designs which begin with Haswell, Intel will be in a position to take the ARM chip market any time they want.

Don't ever discount the power of the foundry.

Re:Long term

[nurb432]

As a former EE i can attest that the design of ARM is far more elegant than the abomination of the x86 line. The only reason they are 'faster' is they throw more transistors ( electrical power ) to work around fundamental design flaws.

ARM could easily scaled up to surpass x86 performance if power was no longer a factor, but it is so you wont see that happening.

Backwards compatibility with existing 'enterprise' apps, well you do have me there.

Re:Long term

[Jeremiah Cornelius]

Long Term?

Apple should acquire AMD, and shift them to being primary supplier and ARMs dealer.

They will go for a song, while Apple had the highest market cap in history.

Re:Long term

[petermgreen]

I disagree.

In general as CPU cores and their associated cache structure get bigger and more powerful the proportion of transitors devoted to instruction decoding goes down. So the complexity of decoding becomes less important and the density of the code becomes more important (because if means you can fit more code in cache). Afaict x86 does pretty well at instruction density. 32-bit x86 is register starved but 64-bit x86 doubles the register count (making it the same as 32-bit arm).

Re:Long term

[Anonymous Coward]

ARM will win the battle, as long as we dont see some sort of bloody patent war.

Its a better archetecture from the start.

So was the 68000 and the PowerPC chip.

Re:Long term

[Anonymous Coward]

Nope. ARM's pipeline is too short and would have to be redesigned.

Re:Long term

[marcosdumay]

ARM has great instruction density... But it doesn't really matter, because you can power an ARM core and an instruction decompressor with less power than you needed for a x86 core, and zipped instructions have a much bigger density than x86, whatever architecture it is.

Anyway, the transistors on cache use much less energy thant he ones on the core. As cache becomes bigger, the core becomes less important, but not as fast as you imply.

Re:Long term

[__aaltlg1547]

What a bold prediction, you understand of course that Intel has buried every single competing architecture from the past? Intel has a process advantage, even if they have to spend 10% of their die on decoding/rearranging they still have a significant transistor lead by remaining a process ahead AND still use lower power.

We'll see if that really will happen. But it's very true about the process advantage. If Intel chose to build and ARM CPU, it would blow the doors off all the other ARMs because they can build faster/smaller/lower-leakage transistors. But it might be more expensive, and the ARM market is very price sensitive. I imagine there are Intel people who have considered this carefully and concluded that they can't build an ARM processor that's cheaper AND better than the competitors.

Re:Long term

[PlusFiveTroll]

You are forgetting 2 non technical matters.

A: Why let Intel 'win' so they can turn around and stuff higher prices clean up your backside? One supplier markers are best for the supplier.

B: Anti-trust. Once you start taking over entire markets all of a sudden you have even more government flashlights up your ass. Only a dumb person wants total control unless he has TOTAL CONTROL, otherwise powers stronger then you take too much interest in your daily work.

Re:Long term

[postbigbang]

You get into RISC vs CISC arguments here. There are only sample quantities of 64-bit ARM at this juncture. There's really no good reference model when you start going that direction with ARM.

Intel isn't paranoid enough to survive, methinks, and will be left behind.

And the scale you'll see is already here; SeaMicro is doing it, and HP will do it if they can get Project Moonshot off the ground.

Transistors are one measure, an important one, but design ease and getting work done is another. Ultimately, power consumption will become an even larger quotient than it is now. Intel fixes stuff differently than ARM designers do, but if the market does walk away from x86, they're a one-trick-pony.

Re:Long term

[fm6]

What a bold prediction, you understand of course that Intel has buried every single competing architecture from the past?

Except for ARM. That architecture has been dominating the mobile device space for some time now, despite serious efforts by Intel to displace it.

Re:Long term

[nurb432]

Lets do this again in 10 years and see where the chips fall and who was right.

Re:Long term

[nurb432]

I also think that microsoft had as much or even more to do with this than Intel. Intel had better processor families than the x86 in the past ( like the 432 ) but due to microsoft's needs they never really had a chance. ( yes i know Microsoft dabbled in MIPS and Alpha support for a short time but they were never really serious about it as it fragmented their code base too much. )

The world is different now and microsoft does not dictate like they could before.

Re:Long term

[nurb432]

What a bold prediction, you understand of course that Intel has buried every single competing architecture from the past?

I forgot to mention that ARM hasn't been 'buried' or we wouldn't be having this discussion.

SPARC isn't gone either. ( basically a modernized MIPS, which was another wonderful architecture, from an engineering standpoint )

Re:Long term

[Anonymous Coward]

Sure, and ARM CPUs use 100 times less power and are a hundred times cheaper.

Re:Long term

[jmauro]

Intel isn't paranoid enough to survive, methinks, and will be left behind.

This is Intel. They are by far and away the most paranoid of all Silicon Valley companies. It's in their DNA. Their unofficial motto is "Only the Paranoid Survive".

Re:Long term

[gr8_phk]

ARM has great instruction density... But it doesn't really matter, because you can power an ARM core and an instruction decompressor with less power than you needed for a x86 core, and zipped instructions have a much bigger density than x86, whatever architecture it is.

...with less power... given the same process. Intel is now making 22nm tri-gate or whatever they're calling it. TSMC is not at 22nm yet and plans to use planar transistors when they do get there. So Intel probably has about a 2 year lead in power per transistor.

Meanwhile I wonder which ARMs even have instructions like divide or reciprocal square root.

Re:Long term

[Sable Drakon]

I guess that's why AMD is dying.

Re:Long term

[postbigbang]

I read the book. I see them not-dominating, not-getting original, making and squandering acquisitions, and generally living off the oil well in the basement called x86.

Tablets, smartphones, device control, embedded systems, graphics subsystems, all go some place called: not Intel. Innovation? Snore.

One. Trick. Pony.

Re:Long term

[Anonymous Coward]

By serious effort do you mean the medfield atom soc because that's only the first iteration. Wait for the 14nm airmont quadcore ht soc before you claim they are serious.

Re:Long term

[Anonymous Coward]

ARM has great instruction density... But it doesn't really matter, because you can power an ARM core and an instruction decompressor with less power than you needed for a x86 core, and zipped instructions have a much bigger density than x86, whatever architecture it is.

Anyway, the transistors on cache use much less energy thant he ones on the core. As cache becomes bigger, the core becomes less important, but not as fast as you imply.

i386 indeed has better(if not the best) code density according to
code density comparisons

Re:Long term

[Anonymous Coward]

i386 indeed has better(if not the best) code density according to
code density comparisons

http://www.eece.maine.edu/~vweaver/papers/iccd09/iccd09_density.pdf

Re:Long term

[Anonymous Coward]

Sure it is. That's why the best ARM CPU is hundreds of times slower than the best x86 CPU.

You must not be very good at math.

Re:Long term

[Anonymous Coward]

Anon Intel Employee here.

Very. Fucking. Paranoid

Re-look at some of the acquisitions from a "If we buy it, you can't" perspective and see if they make more sense

Re:Long term

[CAIMLAS]

... until you compare a suitably designed Intel chip to the task, in which you measure operations per second per watt per dollar, in which the Intel chip would win again.

The low end Atom SoCs Intel is coming out with now very much compete with ARM on price per performance watt.
(Unfortunately.)

Re:Long term

[Anonymous Coward]

I have both Intel RS-NDA and AMD NDAs signed and Intel is far more paranoid.

Re:Long term

[Vintermann]

SPARC isn't gone? Where is it, then?

Re:Long term

[Alioth]

But until Intel sells just the IP for the Atom, it won't gain much traction. Apart from low power, the attraction of ARM is you can buy IP cores and make your own SoC to your own design. Intel won't let you make your own SoC, you have to use theirs.

Re:Long term

[Guy Harris]

SPARC isn't gone either. ( basically a modernized MIPS

Does "modernize" include "added register windows"? The MIPS architecture, which lacks register windows, derives from the Stanford MIPS work, which also had no register windows. SPARC picked up the register windows notion from processors such as the Berkeley RISC. The commercial MIPS and SPARC architectures originated at about the same time, so I'm not sure SPARC could be considered a "modernized MIPS".

Re:Long term

[Guy Harris]

SPARC isn't gone? Where is it, then?

Living in Larry's house. x86 did displace it from the "engineering workstation" market.

Re:Long term

[TheRaven64]

Actually, Intel has been doing this for about two years now. Not only will they license the Atom core to you, they'll also fab the resulting SoC for you at a relatively low rate. So far, I'm not aware of any companies that have taken advantage of this, because they'd effectively be in the position of competing with people both up and down the supply chain from them. This is why ARM has made a point of not producing chips: it would make their licensees very nervous. This is part of the reason for ARM's strategy with ARMv8: they're intentionally not releasing a core for licensing until after some of their partners have done so, to encourage a more diverse ecosystem (the success of XScale and Snapdragon in the past have shown them that this is useful). They win either way: they get a small amount for every ARM-compliant CPU designed by third parties and a larger amount for every ARM-designed CPU, but the more ARM chips in the world the more money they get.

Re:Long term

[slashping]

Except that for ARM to scale up to surpass x86 performance, it would need out-of-order scheduling, multiscalar execution, instruction pre-decoding, hardware loop unrolling, speculative execution, memory/register renaming, multiprocessor cache coherency and all that fun stuff that Intel has. After that has been implemented, let's see if the design is still far more elegant. I'll bet it's going to look pretty similar, with x86 instruction decoding becoming an almost irrelevant issue compared to all the other things. And, by the way, have you taken a look at the Cortex architecture ? It's getting less and less elegant. You are right about the compatibility, though. ARM is a mess, with a dozen different, incompatible, architectures in the core alone.

Re:Long term

[TheRaven64]

Intel killed Alpha and PA-RISC by promising Itanium. They won there through political tactics, not through any technical advantages. Sun and Oracle between them tried very hard to kill SPARC using their awesome mismanagement superpowers, but haven't quite failed yet. PowerPC still dominates in automotive and a few other embedded industries, MIPS is in pretty much every router and a lot of other embedded systems, and ARM basically owns the mobile markets and has a large part of the embedded market. You could possibly claim that Intel beat ARM on the desktop, but it would be more accurate to say that Microsoft beat Acorn on the desktop and Intel went along for the ride.

That 10% becomes a lot more important going forward. The current buzzword in the semiconductor industry is dark silicon. To keep within the same thermal envelope (power dissipation per unit area), you need to have more transistors idle and in a low-power state in every subsequent generation. If you add complex vector instructions, for example, they're great because they give a big speedup when they're in use and draw almost no power when they aren't. The same with things like AES encryption. The instruction decoder, however, is something that you can't ever turn off. Xeons try to: they cache decoded micro-ops in tight loops, but this means that they have some extra SRAM for the micro-op cache and a micro-op decoder that must always be powered, and these between them take more power than an ARM decoder, and a big fat decoder that must be active all of the time.

Intel had an advantage over other RISC architectures (and Itanium) in terms of instruction density, which meant that they needed to waste a lot more die space with instruction cache than x86 to get the same fetch performance, but ARM is already about as dense as x86 and Thumb-2 is typically 5-10% denser, so Intel is on the losing side of this comparison for the first time.

The process advantage is something that Intel has had over AMD, but it's not something that they have to the same degree over some of the foundries that produce ARM chips. They're on 22nm, and the faster ARM SoCs are made on a 25nm process: that's nowhere near the kind of process advantage Intel is accustomed to. In terms of fab R&D, the industry is almost split into two camps, Intel on one side and everyone else pooling resources on the other side.

Re:Long term

[gtall]

Also, getting your SoCs from Intel still means you have a single supplier that could squeeze you in the next contract, especially if they find they really like your SoCs. With ARM, you can shop around for suppliers and keep several if your runs are large enough.

Re:Long term

[gtall]

Yep, that's why Intel got rid of their ARM business several years ago, it would have produced stuff that would piss on their x86 business. And other companies wouldn't buy their ARM chips because they wanted to keep Chipzilla ARMslength away.

Re:Long term

[Anonymous Coward]

That is long outdated, thumb has been replaced with thumb2, ARM can do unaligned load-/stores since ages (if configured for it), hardware division has also been a configuration option since ages.
Maybe it's still correct, but that seems very unlikely.

Re:Long term

[RaceProUK]

yes i know Microsoft dabbled in MIPS and Alpha support for a short time but they were never really serious about it as it fragmented their code base too much

Except for CE (runs on x86, MIPS and ARM, and previously SH4), but that's a totally separate codebase from mainline Windows anyway.

Re:Long term

[hattig]

That's going a bit far, especially when you consider performance per dollar per watt.

And the fact that a $20 chip is great for a $300 device, but a $100 chip isn't. And Intel won't want to destroy their lovely 60% profit margin.

As for the story "Even if more A6 chips could be produced per wafer — an unproven assumption" - the A6 is a known size, and that size is smaller than the i3, so more chips can be made per wafer. Working chips - that's another matter.

Re:Long term

[hattig]

Haswell will be available for $20 in its underclocked ultra-low-power configuration?

Re:Long term

[hattig]

You are aware that different ARM designs have different pipeline lengths depending on their market position? Or is all your knowledge of ARM based upon an ancient university lecture about simple RISC Fetch/Decode/Execute pipelines that the first ARM employed?

Re:Long term

[Anonymous Coward]

I don't think you've taken a look at the Cortex architecture at all, judging from what you have written there.

Re:Long term

[Anonymous Coward]

Condescending much?

Intel doesn't want ARM to win, nor does it want to join the ARM ecosystem (been there, done that), but has no answer that won't kill Intel's 60% margins, except trying to stop ARM reaching upwards into Intel's core markets by marketing, backhanders and other incentives. Intel has to convince the world that they need the extra performance that Intel's chips provides - yet the world is quite happy with their ARM phones and tablets, and raw CPU performance is nowhere near as important as it was five or ten years ago.

What has Intel got that can compete with the full ARM ecosystem? ARM Cortex M, ARM Cortex R, as well as ARM Cortex A. Never mind the custom ARM implementations - Qualcomm Krait, Apple Swift, etc?

Right now, x86 has 64-bit and a really good performance in the laptop and above range. ARM IMO won't compete there for quite a while, but what if the market moves further to tablets and other ARM-based light devices? That kills off the "high margin non-cheap" area of the market that is Intel's primary hope.

Re:Long term

[hattig]

You've also got to consider China with their MIPS64 based Longsoon SoCs. Might not be a threat anytime soon, but with government backing and a desire to do things themselves, it probably will grow and grow, and inevitably start growing outside China as well.

Re:Long term

[jmauro]

They are a one trick pony, but that pony isn't x86. It's never been their processors or their designs, but their manufacturing capacity. The fixed costs are so high to build a fab, that if anyone seriously tries to challenge them they basically cut them off at the knees in terms of price before their production capacity is even built. And since the marginal costs of each processor is very low, they have a huge margin to do it in.

They'll let AMD and others squeak by for anti-trust reasons, but no one can challenge them on manufacturing costs, not even the ARM producers.

Re:Long term

[postbigbang]

It's really been confined to their own stuff. Imagine all of the money they could have made evolving their fabs towards building *anyone's* chips. But they didn't do that. They didn't evolve and market great semiconductor fab/architectural software. They didn't make a zillion dollars from memory.

AMD is a quirk, just like Cyrix, IBM, and others were. ARM is killing them. They should have considered what a 6502-based design scheme could do: undercut them both in costs in ideological development imagination.

Re:Long term

[Anonymous Coward]

Actually, this isn't entirely true, and it depends heavily on the entire platform and on the workload. Ivy Bridge Xeons have amazing performance per watt for server workloads. They may cost 10 watts per core (average load) vs 1W for an A9, but this buys you 3GHz quad-issue 64-bit instructions, or triple-issue 256-bit AVX. The A9 is dual-issuing 32-bit instructions or 64/128-bit NEON vector instructions at (probably) less than half the clock speed. The Xeon has a huge cache and massive memory bandwidth, so it might retain its order-of-magnitude advantage for memory-bound tasks as well. A Cortex-A15 can triple-issue and can scale to higher frequencies, but it also uses considerably more power, which is why big.LITTLE is necessary. Sure, Intel's instruction set sucks and they pay for it in the decoders, but they can amortize that over a huge core.

What's more, you can buy a Xeon server which is balanced for your favorite server workload, i.e. it has the right ratios of CPU, memory, disk and IO. That way you get good performance per watt (dollar, U, ...) for the whole system. This is harder to come by with an ARM core, though Calxeda is making progress. Calxeda ran a "benchmark" where they destroyed a Xeon server in performance/watt, but it's of the "lies and damned lies" variety, and the real results would be much closer.

Where ARM shines is in performance at smartphone power levels. Intel is encroaching on that too with the new Atom platforms, but they aren't there yet.

Re:Long term

[atlasdropperofworlds]

Yes, it's a good thing they are on-part with PC chips 10 years ago in terms of instructions-per-clock performance.... though even 10 years ago the P4's were clocked 3 times higher.

All that has happened here is that the gap between mobile (ARM) and desktop (x86) is just as big as it ever was, it's just that computational power has progressed to the point where the mobile side now has a decent amount of usable computation power. x86 processors are massive overkill for your average user there days, but enthusiasts, gamers, developers, artists, etc. can never really have enough computing power - at least not until every operation is instantaneous.

Re:Long term

[RocketRabbit]

ARM doesn't need better performance in an age when most devices are only using a fraction of their capabilities anyway. It is sufficient to beat Intel on power consumption alone.

Re:Long term

[petermgreen]

Meanwhile I wonder which ARMs even have instructions like divide or reciprocal square root.

The VFP floating point unit has divide and square root operations. It doesn't seem to have a decidated instruction for reciprocal square root though.

There doesn't seem to be an integer divide though, so you either have to do that in software or convert to floating point and back.

Re:Long term

[Anonymous Coward]

Intel killed Alpha and PA-RISC by promising Itanium. They won there through political tactics, not through any technical advantages. Sun and Oracle between them tried very hard to kill SPARC using their awesome mismanagement superpowers, but haven't quite failed yet. PowerPC still dominates in automotive and a few other embedded industries, MIPS is in pretty much every router and a lot of other embedded systems, and ARM basically owns the mobile markets and has a large part of the embedded market. You could possibly claim that Intel beat ARM on the desktop, but it would be more accurate to say that Microsoft beat Acorn on the desktop and Intel went along for the ride.

That 10% becomes a lot more important going forward. The current buzzword in the semiconductor industry is dark silicon. To keep within the same thermal envelope (power dissipation per unit area), you need to have more transistors idle and in a low-power state in every subsequent generation. If you add complex vector instructions, for example, they're great because they give a big speedup when they're in use and draw almost no power when they aren't. The same with things like AES encryption. The instruction decoder, however, is something that you can't ever turn off. Xeons try to: they cache decoded micro-ops in tight loops, but this means that they have some extra SRAM for the micro-op cache and a micro-op decoder that must always be powered, and these between them take more power than an ARM decoder, and a big fat decoder that must be active all of the time.

Intel had an advantage over other RISC architectures (and Itanium) in terms of instruction density, which meant that they needed to waste a lot more die space with instruction cache than x86 to get the same fetch performance, but ARM is already about as dense as x86 and Thumb-2 is typically 5-10% denser, so Intel is on the losing side of this comparison for the first time.

The process advantage is something that Intel has had over AMD, but it's not something that they have to the same degree over some of the foundries that produce ARM chips. They're on 22nm, and the faster ARM SoCs are made on a 25nm process: that's nowhere near the kind of process advantage Intel is accustomed to. In terms of fab R&D, the industry is almost split into two camps, Intel on one side and everyone else pooling resources on the other side.

Is it fair to consider Apple and Samsung as the parallel to Intel and AMD?

Complicated Story

[TaoPhoenix]

I can't find the angle here.

"Legacy Free" vs "Costs".

"Legacy Free" is a nice sounding term for "won't run $hit". So much for your 1,000 app and app-lets you rely on, Business.

So I give up on this story and will let the rest of y'all thrash it out.

Re:Complicated Story

[Sir_Sri]

Won't run shit is interesting. With Windows forking into an ARM and x86 (or AMD64/IA64 whatever want to call it) versions, the writing may be on the wall for Intel. If one of the ARM guys can produce chips that will do the 150-200 dollar price bracket as well as Intel chips can on windows this becomes a whole other ball game.

I'm not sure where anywhere near there yet. But with Qualcomm feasting on the remains of AMD, Samsung producing millions of parts a year and a few others with them it's entirely possible that within the next 10 years ARM will be a major competitor to x86. Which is why MS is forking - it's going to confuse the hell out of consumers and is, from an end user perspective a terrible idea to go out and buy a Windows RT anything on friday (windows 8 launch day) but MS plans to support their ugly bastard for a long time, so who knows. And in 3 or 4 years when we see Windows 9 roll around we may have enough software that has been compiled and for and runs on both that your 'won't run shit' assertion would no longer apply.

Re:Complicated Story

[cheesybagel]

Microsoft is doing half-assed ARM support. They could have added fat binaries to Windows 8 but they didn't. They could have announced ARM ports of MS Office. Didn't hear anything about it.

Re:Complicated Story

[Jeremi]

"Legacy Free" is a nice sounding term for "won't run $hit". So much for your 1,000 app and app-lets you rely on, Business.

I think that's less of a problem in the cell-phone market than in the desktop market.

In the cell-phone market, increasingly there is an App Store type service that automatically upgrades people's installed applications as necessary, so the onus is no longer on the user to do the work.

Re:Complicated Story

[WolfgangPG]

Seriously? Windows RT (Windows 8 on ARM) comes with Office 2013 -- Excel, PowerPoint and Word. You apparently didn't hear about it because you haven't been paying attention -- but Office for Windows RT is rather old news.

Re:Complicated Story

[KiloByte]

With Windows forking into an ARM and x86 (or AMD64/IA64 whatever want to call it) versions

Windows dropped IA64 support, like it did PPC, Alpha and MIPS before.

Re:Complicated Story

[Desler]

With Windows forking into an ARM and x86 (or AMD64/IA64 whatever want to call it)

AMD64 and IA64 are not even remotely the same. Itanium is vastly different than x86 and AMD64 hence why it failed to catch on.

Re:Complicated Story

[cheesybagel]

Hmmm... According to this article it doesn't come with macros or VBA so businesses can't run their crapola apps on it. That mostly defeats the purpose. Still it's less bad than I expected.

Re:Complicated Story

[jtownatpunk.net]

Oh, I see. So I can run my Win 8 Pro apps on my Win 8 RT tablet? No. Can I run my Win 8 Phone apps on my Win 8 RT tablet? No. In RT, Microsoft has created an environment completely distinct from both their desktop and phone platforms.

Ya lost me, Microsoft. Why would I want three separate platforms between my desktop, mobile, and semi-mobile devices? This doesn't make sense. With either Android or iOS, I can have the same apps on my phone and my tablet. With "real" Windows 8, I can have the same apps on my desktop and tablet. I can see solid cases for either of those scenarios. What I can't see is an advantage to isolating the tablet from both platforms.

What does RT provide me as a consumer? As far as I can tell, it just adds complexity.

Re:Complicated Story

[WolfgangPG]

Correct. Windows RT is targeted at consumers. if you want business LoB apps, Business Macros, etc... get a Windows 8 Tablet.

Re:Complicated Story

[Anonymous Coward]

"Legacy Free" is a nice sounding term for "won't run $hit". So much for your 1,000 app and app-lets you rely on, Business.

I think that's less of a problem in the cell-phone market than in the desktop market.

In the cell-phone market, increasingly there is an App Store type service that automatically upgrades people's installed applications as necessary, so the onus is no longer on the user to do the work.

Yeah pretty soon it will be as good as the linux package managers i've been using since the early 1990s. Yay innovation!

Re:Complicated Story

[WolfgangPG]

As a consumer it provides you with a secure and easier to maintain Windows. You just get your apps from the app store, it comes pre-loaded with office and you shouldn't have to worry about the x86/x64 legacy issues.

Of course that might not be what you want, but that is what it offers the consumer. You get a truly productive tablet, that is cheaper than Windows 8, ligther, and more secure (since it is more locked down, hopefully hacks would occur less and malware shouldn't be as frequent in the Windows App Store). However it is more restricted and doesn't run legacy applications.

Re:Complicated Story

[jader3rd]

What does RT provide me as a consumer? As far as I can tell, it just adds complexity.

Well, iOS doesn't run legacy software and yet people seem to enjoy running it. So on day one RT will be dumb. But let's say 3 years from now when there may exist non crappy apps, you'll be able to run the same non crappy RT apps on a desktop as well as a portable device.

Re:Complicated Story

[Anonymous Coward]

I realize it's mere anecdote, but I haven't seen anyone use macros or VBA in Office for the better part of 10 years. I think the, "this is a bad idea in every way" mantra finally sunk in.

Re:Complicated Story

[Archangel Michael]

Windows Tablet PRO-- key word

Windows Tablet is gonna be both ARM and X86/X64. Which is gonna cause more confusion when son says "Buy Windows Tablet Pro" and Dad buys "Windows Tablet" (not pro) and it won't do what he wants.

Re:Complicated Story

[PopeRatzo]

If one of the ARM guys can produce chips that will do the 150-200 dollar price bracket as well as Intel chips can on windows

If the queen had nuts she'd be king.

Re:Complicated Story

[cbhacking]

Microsoft's branding around Windows sucks abysmally, but that's no excuse for just getting it wrong. The correct information is easy to find. Please stop muddying the waters with brand names that outright do not exist.

There is no such thing as Windows 8 RT. Windows RT and Windows 8 are very nearly identical, but Windows RT is expressly *not* marketed as Windows 8.

There's no such thing as Windows 8 Phone. Just like there was no such thing as Windows 7 Phone. Windows Phone is the name of the OS family. Windows Phone 8 is the new version. As with Windows 8 and Windows RT, it runs on the NT kernel, but that doesn't make it the same product any more than RHEL, Android, and Chrome OS are the same product (even though they all run Linux kernels).

Windows RT is intended for those who want an iPad-class device that also supports multiple users, filesystem access, sideloading, Windows networking, USB peripherals, scripting (powershell, CMD, WSH), Office, and Xbox integration, plus regular updates from the OS developer but many different hardware options to choose from. They want the low cost and low power usage of ARM, in a device that is cheaper, lighter, and gets battery life than an ultrabook at the cost of computing power and backward complexity.

It's not intended to replace people's PCs, or their smartphones, or their consoles. At least, not any more than the iPad or Nexus 7 are intended to replace such things. It's a decent alternative to a Kindle Fire (there's already an official Kindle app, and unlike the Fire you don't have to do any jailbreaking to control how the FS is allocated). It supports game controllers but is more multi-purpose than a console (and has less mastery of gaming). It's much more accurate to say that Windows 8 can run WinRT apps than it is to say that Windows RT runs Win8 apps.

Re:Complicated Story

[jtownatpunk.net]

That wooshing sound is my point going over your head. Nothing I said had anything to do with the number of apps available.

iOS = Phone and tablet.
Android = Phone and tablet.

Full Windows 8 = Desktop and tablet.

Windows 8 RT = Tablet only.

With three of those platforms, you get two devices sharing a common base of applications. With one of those platforms, there is no crossover. Windows 8 RT does not share an application base with the phone. It does not share an application base with the desktop. It compliments nothing. It doesn't give you a bigger phone. It doesn't give you a smaller desktop. You have an app for your phone, a different app for your desktop, and yet another app for your tablet. Windows Phone and Windows RT are orphans. Do you see what I'm saying now?

If RT could run Phone apps, I wouldn't be typing right now. It would be a bold and smart move by Microsoft. They'd be the only company offering consumers a choice between pairing the tablet with the phone or with the desktop. But that's not what they've done.

Re:Complicated Story

[Sir_Sri]

Right, I confused the Itanium instruction set with whatever intel brands its 64 bit ISA.

Re:Complicated Story

[Sir_Sri]

Right, I confused the Itanium instruction set with whatever intel brands its 64 bit ISA.

Duplicate comment because multiple people corrected the same thing in my post...

Re:Complicated Story

[drinkypoo]

Right, I confused the Itanium instruction set with whatever intel brands its 64 bit ISA.

Just remember, ia64 == iTanic == shit sandwich, amd64 is where it's at. Which is why it's so heartbreaking to see AMD so far to the rear in terms of performance today.

Re:Complicated Story

[Anonymous Coward]

You seem to be confused already. Don't worry, it's quite understandable. 'Windows Surface RT' is the name of the ARM version, and 'Windows Surface PRO' is the name for the x86 version. It won't be easy. Stay strong.

Re:Complicated Story

[jon3k]

It can run a web browser, so as early as today, and as late as 5-10 years, it'll run almost everything most people need (see: Live365). Other than the fraction of a percent doing CAD or 3D modeling, etc. And that's a long time for ARM to come a long way in terms of power.

Re:Complicated Story

[gr8_phk]

I'm not sure where anywhere near there yet. But with Qualcomm feasting on the remains of AMD, Samsung producing millions of parts a year and a few others with them it's entirely possible that within the next 10 years ARM will be a major competitor to x86.

ARM is competing with itself. With all those companies making ARM chips they have significant price competition which will lead to reduced R&D budgets. Meanwhile if you want top performance there is only one game in town and they get to charge top dollar. If Intel ever does view ARM as a threat, they can just license the designs like everyone else and throw their own GPU and radio hardware into the mix on a more advanced process for less money. Or perhaps they'll license ARM instruction set only and slap both ARM and x86 decoders on their existing cores, along with a GPU, radio, and better process.

Re:Complicated Story

[KiloByte]

AMD is leading, when it comes to bang for the buck. So unless you need single core performance but nothing else, you'd better pick AMD.

Re:Complicated Story

[Anonymous Coward]

Intel actually refers to their implementation of x86-64 (it differs ever so slightly from AMD64) as "Intel 64", having called it "EM64T" in the past.

Re:Complicated Story

[Pulzar]

ARM is competing with itself. With all those companies making ARM chips they have significant price competition which will lead to reduced R&D budgets.

On the other hand, with so many companies trying to build their own high-end ARM cores, the total ARM R&D budget is enormous. It guarantees that there never will be a bad ARM generation, as at least a couple of those companies will have a winner in every round.

Or perhaps they'll license ARM instruction set only

ARM makes money whoever licenses their stuff. If Intel pays for it, they are only better off.

Re:Complicated Story

[Sir_Sri]

ARM makes money whoever licenses their stuff. If Intel pays for it, they are only better off.

AFAIK intel has had a dormant ARM licence for years.

Re:Complicated Story

[ozmanjusri]

You missed one:

Linux = Phone and tablet and desktop and clouds and embedded and....

In fact the key undiscussed event that's happening in computing is the decoupling of data and applications from devices. For most of the world, it doesn't matter if they have an iPhone, Android tablet, Windows laptop, Google Chromebook or Linux desktop in front of them. The things they value and interact with will be available and will work.

There are still a few straggling pockets of vendor lockin, (like office document formats), but the world is beginning to route around that bit of damage.

Re:Complicated Story

[Dahamma]

Actually, *8* is the key word. He said "Windows 8 Tablet", which is correct. The OSes for tablets are "Windows RT" and "Windows 8 (Pro)". Windows RT is the ARM port. Windows 8 is the x86 port. "Pro" is just a different version of Windows 8. And both Windows 8 and Windows 8 Pro can run (full) Office for x86.

Re:Complicated Story

[Anonymous Coward]

that's no excuse for just getting it wrong.

So?

Who cares?

Seriously, Microsoft knows with absolute certainty people are going to get the wrong "Windows" and suffer distress, be out of pocket etc etc. There can be no doubt at all that they are expecting that result.

So why whine at a Slashdot poster about intentional confusion when it is clearly part of Microsoft's business plan?

Re:Complicated Story

[Dahamma]

In fact, it is all about the available apps. If the same apps are ported to the different platforms, consumers really won't care. And porting/cross compiling Windows Phone 8, Windows RT, and Windows 8 (metro) apps really isn't that hard (Windows 8 desktop apps are a different story, but that's not really relevant here). Hell, even the Xbox 360 media apps can use a lot of the same tools and code as those other platforms.

Now, whether people will write apps for these platforms is an interesting question. But given the user and developer bases Microsoft has, I'm guessing they will.

Re:Complicated Story

[jader3rd]

It doesn't give you a smaller desktop. You have an app for your phone, a different app for your desktop, and yet another app for your tablet. Windows Phone and Windows RT are orphans. Do you see what I'm saying now?

Yes, I think I get what you're saying. Thank you for your clarification; perhaps I can make some of my own.
When iOS was first released (2007) it didn't run any sort of legacy programs from existing touch screen smartphones, and yet people really seemed to enjoy using those devices. When Android was first released, and was still just a phone OS, people seemed to enjoy using those devices. When the iPad first came out it could run existing iPhone apps; but even Apple said that the experience for many apps wasn't good and that app authors needed to (and still do) optimize their apps for the larger screen. There's hardly an article on the web about Android tablets that doesn't mention how there's a pithy of tablet optimized apps, and that while the tablets can run all Android apps, most of them really suck on the tablet. So I don't think that not having the ability to run legacy programs is a nail in the coffin for any new device/platform.

So right now a Windows RT computer does have limited appeal, due to not having much of an ecosystem. That was kind of my point by saying "3 years from now", the ecosystem may come. The ecosystem may come easily because if developers write apps for the Windows 8 Store (targeting desktop) the apps will also light up on Windows RT; and they will light up without any "tablet optimization" step that iOS and Android apps suffer from. The screen sizes are the same. So I think that there will be crossover soon, at least from the point of view of the end user; because they'll be able to find the same apps on a Windows RT computer as they do on their Win8 laptop/desktop.

If RT could run Phone apps, I wouldn't be typing right now.

Yes, Microsoft doesn't have a runtime that runs across Win32, .Net, WinRT, Phone and Xbox. But they do have portable assemblies which do allow for an assembly to run within .Net, WinRT, Phone and Xbox. So someone would still need a separate app/program for the different runtimes, but if the business logic is the same for all of the apps there only needs to be one portable assembly.

Re:Complicated Story

[Anonymous Coward]

So in other words, for the market that Microsoft really cares about (enterprise) get the tablet that will contain a Intel chip? Not an ARM processor. Sounds to me like half arsed ARM implementation.

Re:Complicated Story

[MrEdofCourse]

I feel bad that I spent all my mod points yesterday. This is pretty insightful.

Re:Complicated Story

[MrEdofCourse]

Well that's a bit like saying that since iOS and OS X use the same Xcode tools that there's an advantage. And yes, that is one advantage Apple and its developers can leverage over Android which just has tablets and phones (in volume, along with a handful of Chromebooks, TV boxes, etc...). But it does little for the consumer who still needs to buy apps for both iOS and OS X and learn how to use each (often the apps have very little in common in terms of UI).

But with Microsoft, it's confusing and frustrating to the user when the purchased app for their phone won't run on their tablet or vice versa. One of the reasons the iPad took off was because on day one it could run a bazillion iPod/iPhone apps (in 2X) mode, and universal binaries quickly brought about optimized apps that were "buy once".

Re:Complicated Story

[Guy Harris]

Right, I confused the Itanium instruction set with whatever intel brands its 64 bit ISA.

Which one? They have Itanium, formerly called "IA-64" (and perhaps also called "this was originally HP's design", but they probably worked with HP on it), and Intel 64, called "we licensed this extension of x86 to 64 bits from AMD but there are a few minor differences from AMD64".

Re:Complicated Story

[drinkypoo]

Intel actually refers to their implementation of x86-64 (it differs ever so slightly from AMD64) as "Intel 64", having called it "EM64T" in the past.

I don't give one tenth of one fuck what intel calls it. Remember how all the AMD processors were "i386" compatible? Now the intel processors are "amd64" compatible. Fuck intel and their lies.

Re:Complicated Story

[Anonymous Coward]

They may have drop support for NT4 on 32-bit PowerPC, yet they continuously build NT6 for 64-bit PowerPC for the Xbox-360 and Xbox720. Microsoft has licensed 64-bit Power from IBM.

Re:Complicated Story

[Anonymous Coward]

The last thing ARM needs is an Intel GPU. Seriously.

Show me a 22nm Atom. That's right, they're not out until 2013. You can get 28nm ARM SoCs today.

Re:Complicated Story

[Compaqt]

If there are so many companies, how do they sync the command set? Or does each ARM processor line (from Samsung, Qualcomm, TI or whoever) differ slightly?

Re:Complicated Story

[Dahamma]

Well that's a bit like saying that since iOS and OS X use the same Xcode tools that there's an advantage.

No, it's not. Have you developed Metro-style apps for Windows Phone, Windows RT, and Xbox 360? There is a ton of code (not just tools) that can be shared. While there are UI layout and API differences to deal with, that's not much different from the situation Apple is in now with a bunch of different screen resolutions & aspect ratios, hardware capabilities, etc. iOS devs have to deal with all of that, they just combine all of it into one binary, big deal (and sometimes they don't - there are plenty of separate iPhone and iPad versions of apps out there).

But it does little for the consumer who still needs to buy apps for both iOS and OS X and learn how to use each (often the apps have very little in common in terms of UI).

Umm, now I'm confused about your point. That's something Microsoft *has* addressed that Apple hasn't... Metro-style apps are very similar in terms of UI, and can be developed on pretty much every platform they have. If anything, the common look and feel is their whole *point*. The implementation of that is the thing that will be more effort.

But with Microsoft, it's confusing and frustrating to the user when the purchased app for their phone won't run on their tablet or vice versa.

Why is that? Metro apps are installed through an app store, anyway. There is no reason Microsoft can't just install the correct version for each platform via that store. I don't know if they plan to do that, but it's NOT an inherent technical issue. As I said before, it is somewhat more work for app developers to build separate binaries, but not *that* much more. It is a somewhat different strategy from Apple, and it remains to be seen if it will succeed, but there is no reason it *has* to be difficult for the customer.

And, while I agree the "2x mode" iPhone apps did help the iPad initially, it was never a particularly good experience. And in no way are they "universal binaries" - that term means binaries running on more than one CPU architecture. iOS apps are about as much universal binaries as a Windows app that runs on a laptop and a desktop.

Heh, now I bet the slashdot trolls will start calling me a Microsoft apologist/fanboy/paid shill because I defended their products. Well, I prefer to use a Macbook Pro (with both Linux and Windows VMs as needed), iPhone, and iPad at home, but I develop software running on Linux, Android, Windows, XBox, and iOS. Use the platforms you like, develop for the platforms your customers want.

Re:Complicated Story

[atlasdropperofworlds]

Any idea how long it takes for an ARM chip to compile the latest linux kernel? How about to encode 120 minutes of 4k video? How much folding can it do? Sure it's great for people who don't already use computers, but it isn't great for a huge segment of the population who has actual, real, demanding work to do. x86 is way far ahead of arm in terms of instructions per cycle, and CISC will always be ahead of RISC in this regard.

Re:Complicated Story

[RocketRabbit]

Explaining that there's no Windows 8 Phone, but that it's really Windows Phone 8, and that Windows 8 tablet is different from Windows RT 8 or whatever, to an average consumer, would probably be a more effective way to get their head to literally explode than anything you saw in the classic movie Scanners.

This is ultimately what's being discussed here.

Re:Complicated Story

[Anonymous Coward]

With either Android or iOS, I can have the same apps on my phone and my tablet. With "real" Windows 8, I can have the same apps on my desktop and tablet. I can see solid cases for either of those scenarios. What I can't see is an advantage to isolating the tablet from both platforms.

Unless you actually own one of their phones, the issue of those being another platform is probably a moot point. And it is likely that someone has figured out how to load Android on them by now. Hopefully that fix will shortly be available for the RT netbooks also. Sure it's simpler to just buy something else, but some people will be gifted these things.

Re:Complicated Story

[jon3k]

Most humans don't perform any of those activities and don't need that kind of horsepower, certainly not a "huge segment of the population" - that's just absurd. They login to facebook and twitter, create spreadsheets and word documents, use gmail and hotmail and watch cats fall off of couches on youtubes. Those tasks you listed represent a tiny fraction of the populace.

Re:Complicated Story

[cbhacking]

Microosft's branding, at least around Windows*, really and truly sucks. I try to keep people informed (it's "Windows RT", not "Windows RT 8") when discussing the technology, in the same way that I correct people who say "Droid" when they mean "Android" or "computer" when they mean "monitor". Microsoft doesn't make it easy though. I'd think it was some conspiracy to confuse, except that the handful of MS employees I've met who defend that branding appear to honestly think that people *love* the Windows brand, so incompetence seems at least as likely. Considering they at one point officially announced "Windows Phone 7 Series", claiming their branding people are thoroughly incompetent seems like a fairly defensible claim.

Re:Complicated Story

[RocketRabbit]

See you can't blame people for not getting the branding straight. Personally I can't be assed to even care enough to worry about what the names really are, despite my ability to learn these things if I really wanted. The excruciating minutia of Microsoft's marketing terms of art are just not worth my time.

So, ultimately, I would counter your claim and say that people have every excuse for getting it wrong.

What I understand is that there's an ARM Windows, on the tablet side, an Intel Windows on the tablet side, a PC Windows, and a phone Windows all of which have some sort of 8 in the name, excepting the tablet one which you say has no "8" in it, but I have read that it is a Windows 8 product before. Maybe you can chalk it up to a journalist being confused, or whatnot, but ultimately it's just a gigantic clusterfuck any way you look at it.

Re:Complicated Story

[MrEdofCourse]

I think you're missing the larger point due to the details.

I won't argue that developing across Windows all the way to the Xbox is easier than developing across iOS and OS X, but to your point, there is also code that can be shared going iOSOS X, developing across both is easy and an advantage. But it still doesn't mean much from the consumer who has to *purchase* per platform, and learn to use each with the different platform differences.

"there are plenty of separate iPhone and iPad versions of apps out there"

Yes, but most aren't, and it's very clear when they're separate. Again, having separate is a disadvantage. iPhone/iPad doesn't score a perfect 100 in this regard, but it's a hell of a lot better than 0.

" If anything, the common look and feel is their whole *point*. "

While there is more look and feel mobile to desktop with Metro, it's still not 100%. It depends on who you ask of course, but some feel it's a bastardization of both wherein having to use drop down menus on a phone or tablet sucks, while using something designed for touch on a desktop sucks. But even with this, it's still different enough to add to the confusion that Surface RT is not going to give someone the same experience.

"There is no reason Microsoft can't just install the correct version for each platform via that store."

Right, but the problem comes from a customer *buying* one and then going back to install on the another. They'll have to buy it again. Again, Apple isn't 100% perfect here, but much better than 0.

"And, while I agree the "2x mode" iPhone apps did help the iPad initially, it was never a particularly good experience. And in no way are they "universal binaries" - that term means binaries running on more than one CPU architecture"

Call them whatever you want, but the point is the app package contents contain resources and code which are designed to run optimized for different devices. I've never heard a different term for this. 2X on the other hand obviously wasn't optimized or ideal, but good enough to allow the iPad to be used for critical uses on day 1 until the optimized apps came out.

Nice Markov Chain generator.

[lsommerer]

Wow, that last article looks like a really good Makov Chain generator (or whatever the kids these days are using).

The War Between Intel Core and ARM

[preaction]

The war between CORE and ARM raged across thousands of worlds, ravaging the galaxy. Neither would waver in their belief in their own supremacy. For each side, the only acceptable outcome is the complete elimination of the other.

Re:The War Between Intel Core and ARM

[Synerg1y]

And off to the side all but forgotten, APUs bid their time...

Re:The War Between Intel Core and ARM

[e065c8515d206cb0e190]

Thanks dude, you made my day. I remember the first time I watched this intro... and the 10 following years spent playing the most awesome game ever.

(now you got me thinking, I had 16MB of RAM back then, now I have 16GB... time to see if building a 50,000 unit army would work!)

Re:The War Between Intel Core and ARM

[Anonymous Coward]

There are patches to get you too 5000, but the app can't handle any more.

Re:The War Between Intel Core and ARM

[Anonymous Coward]

Could someone please explain the joke? I don't think I ever played the game being quoted, but I'm interested.

Re:The War Between Intel Core and ARM

[preaction]

The two factions of Total Annihilation were the Core and the Arm: http://en.wikipedia.org/wiki/Total_Annihilation

nobody else has fab left?

[Anonymous Coward]

how about Texas Instruments?

Re:nobody else has fab left?

[MichaelSmith]

For TTL?

Re:nobody else has fab left?

[YetAnotherForumAcc]

TSMC, Global Foundries ....

Apple is rumoured to be shifting production to TMSC to take advantage of their upcoming 20nm process for their quad-core chips.

http://cens.com/cens/html/en/news/news_inner_41728.html

Re:nobody else has fab left?

[Sir_Sri]

And Qualcomm, and Globalfoundries. And I think a few more beyond that too.

Re:nobody else has fab left?

[Anonymous Coward]

Qualcomm is fabless, but I'm sure they'd love to replace the A6 with their own proc.

Re:nobody else has fab left?

[Wesley Felter]

TI has no modern fabs; they decided to get out of the fab business a few years ago.

Re:nobody else has fab left?

[cheesybagel]

AFAIK Qualcomm is a fabless company using TSMC to manufacture products.

Re:nobody else has fab left?

[Anonymous Coward]

On the "getting out of fab business", parent is incorrect. In fact TI had been buying up yard sales during the downturn.

http://www.eetimes.com/electronics-news/4204587/TI-buys-two-fabs-from-Spansion-Japan That's from 2010
>The move is the most recent in a series of analog manufacturing expansions announced by TI over the past 24 months. In total, the fabs will add capacity for more than $3.5 billion of additional analog revenue per year when fully operational.

>By year’s end, TI plans to ramp the 200-mm fab in Japan, which is capable of boosting its analog sales by some $1 billion per year. The existing tool set in the fab can be tuned for TI’s analog process. That fab is slated to make TI’s precision analog products, based on its HPA07 process.

Given the lower complexity and small die sizes, TI seems to be making money doing analog part from their not so bleeding edge fabs.

Re:nobody else has fab left?

[Anonymous Coward]

I have no idea where you're getting your information but you're flat out wrong. TI has millions of sq. ft. of wafer fab capacity spread throughout the world. TI opened a new 300mm wafer fab in the US about 5 years ago and has spend the last couple of years buying up fabs in the far east at fire sale prices.

"Genetic Handicap"

[Anonymous Coward]

"For all of Intel's semiconductor design and manufacturing feats, its processors suffer from a genetic handicap: They have to support the legacy x86 instruction set, and thus they're inherently more complicated than legacy-free ARM devices"

Oh shut up. This argument comes up every time there's an ARM vs Intel debate. And you know what? Intel is pushing hard and successfully into ARM's territory and ARM has yet to hit back with any chip that can compete with Intel in servers or high end laptops or etc. And that's WITH Intel's huge profit margins. ARM certainly doesn't have the profit margin's to spare in any price war. Intel is a huge monster to defeat, and its supposed handicap means far less worry for programmers, unlike trying to support the million and growing ARM SOCs out there and the nightmare that is.

Re:"Genetic Handicap"

[NatasRevol]

Are you really so ignorant as to not recognize that ARM isn't in the 'server or high end laptop' world?

They're cheap and low power. Perfect for small mobile devices.

However, I really don't understand why Intel won't play both sides of the fence. Why not build an AMD line/factory to offer both types of chips. Take away business from competitors. Get the past to pay for the future.

Re:"Genetic Handicap"

[Anonymous Coward]

probably because most people are penny pinchers and do not want to pay for their mobile phone additional $200 (intel low end price) or additional $1000 (intel high end price) until people start paying a lot more for mobile phones PC business brings much more money

Re:"Genetic Handicap"

[foniksonik]

I do believe that ARM is not a chip or even a product. It's an architecture that is licensed by others. This means that the company behind ARM makes money on every chip regardless of price. They don't care if it costs $17 or $170 to manufacture and distribute, they have little overhead so it's almost all profit at this point.

Intel OTOH sells chips. They a much higher amount of manufacturing and sales costs.

Re:"Genetic Handicap"

[Anonymous Coward]

Intel is pushing hard and successfully into ARM's territory and ARM has yet to hit back with any chip that can compete with Intel in servers or high end laptops or etc.

I remember ARM2 & ARM3 running circles around 80268 & 80386. Damn, I'm getting old!

Re:"Genetic Handicap"

[gr8_phk]

However, I really don't understand why Intel won't play both sides of the fence. Why not build an AMD line/factory to offer both types of chips. Take away business from competitors. Get the past to pay for the future.

ARM is killing all the other companies in a blood bath of competition. Meanwhile Intel is absolutely dominating in fabrication. The foundries don't feel a need to compete with Intel because x86 isn't their market so they don't feel the need to worry.

Re:"Genetic Handicap"

[Anonymous Coward]

Actually, simpler designs tend to be better for programmers. Instead of the ridiculously screwy rules for amd64, you have something that can be readily called an architecture. Most programmers aren't going to be implementing drivers or writing assembly language. The reason the Intel Atom processors are doing as well as they are is Intel is able to crank them out in vast quantities using processes more advanced than any ARM manufacturer has, and throwing large numbers at transistors to give performance. Manufacture ARM chips on those lines and they'll trivially beat Intel's Atom processors.

Re:"Genetic Handicap"

[Pulzar]

ARM certainly doesn't have the profit margin's

Intel has operating/profit margins of 30/22%, while ARM Holdings has 37/27%.

On top of that, ARM wouldn't engage in a price war with Intel directly, as they don't sell any processors themselves. Intel would be fighting other giants like Samsung, Qualcomm, Apple, nVidia, etc. Samsung itself has almost 50% higher market cap than Intel.

Re:"Genetic Handicap"

[Anonymous Coward]

The multi-billion dollar high volume manufacturing company has a 30% gross margin, while the small DESIGN company has a 37% margin.

Apples to Oranges. Samsung and TSMC would LOVE intel's operating margin.

Re:"Genetic Handicap"

[jkflying]

Samsung isn't just in the computer chip industry. They make screens, phones, tablets, ships (yes, the big cargo hauling ones) etc

Re:"Genetic Handicap"

[Aluvus]

In reality, the implementation of a modern x86 CPU does not (and has not for years now) look anything like a 386. The chips have a chunk of decoder logic that translates x86 instructions into the processor's own internal instructions ("micro-ops"). I believe it was AMD that made the point a few years ago: that decoder logic doesn't get much more complex over time, and therefore (by the magic of process shrinks and increasing transistor counts) actually gets cheaper over time. At this point it's basically a trivial cost, other than the design work to make sure the decoder works. IIRC Intel has made similar statements.

So basically this statement makes me wonder how much the author actually knows about the CPU business.

Re:"Genetic Handicap"

[Guy Harris]

Actually, simpler designs tend to be better for programmers. Instead of the ridiculously screwy rules for amd64, you have something that can be readily called an architecture.

The programmers affected by that are mainly the compiler, assembler, linker, and debugger developers. Even most parts of OS kernel code are turned into machine code by a compiler (and may have to run on multiple instruction set architectures, so the developers aren't targeting a particular instruction set).

Re:"Genetic Handicap"

[Anonymous Coward]

If the legacy aspects of x86 were not a recognized issue, then why did Intel attempt to abandon it in the transition to 64-bits? Yes, the attempt was unsuccessful; however, there must have been reasons for the attempt, especially when a transition to a new architecture opened up the possibility of customers migrating to non-Intel architectures.

Re:"Genetic Handicap"

[Anonymous Coward]

the truth is that the arm instruction set is also far from perfect.
it hasn't got enough registers (same problem intel has), its
code density sucks, and the cache-coherency model forces
the processor to flush cache lines manually, tying up the cpu
unnecessarly. (okay, that's not isa, but platform. you get the
point.)

Re:"Genetic Handicap"

[VGPowerlord]

it hasn't got enough registers (same problem intel has)

The largest gain when switching from x86 applications to x86-64 applications is that x86-64 has twice as many registers as x86 does.

Re:"Genetic Handicap"

[RocketRabbit]

I think the issue here is that there are no Intel phones, no serious Intel tablets, and that these are the largest, and fastest-growing, processor markets in the world.

You can only sell so many Xeons destined for the datacenter. Maybe, what, one per ten thousand phones out there?

The other point, which you have deliberately missed, is that the legacy x86 instruction set cause every Intel processor to require much more silicon, which means much more power consumption, which precludes any smart manufacturer from including an Intel processor in their product. They will never be able to compete with ARM as long as the x86 instruction set is in the Intel processors. At this point, even a lowly ARM is plenty fast for 95% of the users out there.

x86 vs ARM, what about atom

[Anonymous Coward]

Intel's published prices range from a 'low' $117 for a Core i3 processor.

What about atom? You know, the processor produced by Intel, specifically for the same markets that ARM are dominating now.

Re:x86 vs ARM, what about atom

[viperidaenz]

But it wouldn't look as good in the article if they compared the $17.50 ARM's to the $19 Atoms (although the only $19 atom is the E620, which is pretty shit, event for an atom, the new N2760 is supposed to be a similar price to directly compete with ARM SoC's)

Re:x86 vs ARM, what about atom

[Anonymous Coward]

Going by the DMIPS figures (not the best of metrics, but the only directly comparable one I can find), the E620 has significantly lower performance than Apple's A6, per core, which the E620 only has one of.
It would be better compared to a single-core Cortex A8.

I wasn't expecting it to be nearly this bad when I started searching for numbers. As you say, that's pretty shit even for an Atom.

Re:x86 vs ARM, what about atom

[Anonymous Coward]

Never mind the price, a Core i3 draws roughly 10x the power. Presumably Apple spends enough time on user studies to realize that people want their battery to last at least 15 minutes...

Well, sort of, but not really.

[Elbereth]

Intel has made ARM processors in the past (xScale), and, apparently, still retains an ARM license. Intel has manufactured RISC chips, as well (i960, for example). There is absolutely no reason why Intel wouldn't/couldn't produce an ARM chip, if they wanted to. There's just no reason to do so.

Also, using the Core i3 as an example of Intel's "low-end" is not very fair. Intel's low-end chips are the Pentium and Celeron, not the i3. The Atom is the closest thing to a competitor to the ARM chips. Pricing for Atom chips varies extensively, from $20 to $100, depending on features,

Re:Well, sort of, but not really.

[amorsen]

Intel has made ARM processors in the past (xScale [wikipedia.org]), and, apparently, still retains an ARM license.

They were crap though. I have an XScale-based PDA lying around somewhere. They were truly the Netbursts of the ARM world: high clock speed and power consumption but low performance.

Re:Well, sort of, but not really.

[Amouth]

Compared to the Samsung arm chips at the same time the xScale blew the doors off them in performance clock for clock, and at that time no one did well with power consumption except when asleep.

Re:Well, sort of, but not really.

[Anonymous Coward]

No-one did well with power consumption so it's ok that Intel were by far the worst

Did that contortion make your back hurt?

Re:Well, sort of, but not really.

[tlhIngan]

Intel has made ARM processors in the past (xScale [wikipedia.org]), and, apparently, still retains an ARM license.

They were crap though. I have an XScale-based PDA lying around somewhere. They were truly the Netbursts of the ARM world: high clock speed and power consumption but low performance

Intel sold the ARM license to Marvell who owns the architectural license to it. Intel does re-license back the Xscale core for some of their networking processors though.

As for Xscale being crap - back in the day, StrongARM and Xscale were the top of the line - the PXA255 being one of the fastest ARM chips around. The next-generation chip was supposed to be even faster, but Intel sold it to Marvell who doesn't seem to have done anything with it.

While StrongARM was pushing 200MHz, other ARMs were barely breaking 133MHz and not very fast at it. When the PXA255 upped it to 400, it was no competition. Then ARM decided they had enough of being outclassed by Intel and designed some decent ARM11 cores and continued onward with the Cortex series.

Re:Well, sort of, but not really.

[Anonymous Coward]

I had a Dell Axim x51v PDA back in the day. Its Xscale processor ran at a whopping 624 Mhz and used a PowerVR-derived graphics co-processor.

I can assure you it was not underpowered. The damn thing could run Quake 3 Arena.. though the controls left something to be desired! :)

Intel retains the short term process advantage

[Anonymous Coward]

For the time being, they have a big advantage in that they not only are producing working 22nm designs, but are using 3D gate transistors.

Most other fabs are going to stick with planar transistors until they move to a smaller process node. Intel has already proven that they can do finfets, whereas no other fab has had a serious production run of finfet parts. If they run into problems transistioning to finfets, Intel will likely end up another 6 months further ahead of them than they already were.

Ironic

[fm6]

If Gassée is right about "architectural disdain" then it's kind of ironic. Intel itself exhibited the same disdain for x86 architecture when they initially refused to make their first 64-bit chip, the Itanium, backward compatible with it. It was only after AMD demonstrated that the architecture still had legs that they brought it to the 64-bit world — after wasting billions on Itanium development.

Those that forget history, yada yada.

Re:Ironic

[DNS-and-BIND]

It's ironic that you posted that ironic comment, as it's ironic that Gassée would be right after being so spectacularly wrong about a similar topic.

"I once preached peaceful coexistence with Windows. You may laugh at my expense - I deserve it."
-- Jean-Louis Gassée, CEO Be, Inc.

Re:Ironic

[fm6]

It's ironic that you posted that ironic comment

Huh?

Re:Ironic

[SpazmodeusG]

You have that completely backwards. The first Itaniums WERE backwards compatible with IA-32 (x86) at the hardware level. It was later Itaniums that ditched backwards compatibility in favour of the software based IA-32 Execution Layer.

Re:Ironic

[fm6]

I stand corrected. But it sounds to me like the performance of the x86 instruction set was truely pitiful — so much so that it was claimed that replacing it with software emulation actually improved performance. So I remain convinced that Itanium design did not make backward compatibiity a priority.

what am I missing?

[ThorGod]

Apple's the one currently manufacturing their A6 chips for $17, while the comparable Intel chip retails for much more?

Isn't this more a statement of how well Apple's vertical integration of chip manufacturing went?

Re:what am I missing?

[ebunga]

Well for starters, speed, flexibility, speed, speed, flexibility and speed.

Intel not the only or even likely source

[Anonymous Coward]

Given that Qualcomm, Samsung, TI for now (Apple could buy TI's mobile chip business they are divesting themselves from), and pretty much anyone who wants to can be an ARM vendor, I don't see a lot of particular interest in going Intel for this sort of business.

I'll say also that it's disingenuous to compare i3 pricing to snapdragon part, intel also sells Atom processors which would be a bit more comparable.

They have to support the legacy x86 instruction set, and thus they're inherently more complicated than legacy-free ARM devices, they require more transistors, more silicon.

I've been told and am very willing to believe this has always been a somewhat dubious argument and what realities that feed into that belief diminish with every generation of processor.

Intel and MS are currently in the same codependent boat. They also have similar behavior, where both make moves ostensibly to compete with prominent Tablet and Phone components, but putting premium pricing on such efforts that make them non-starters because they only want the business if the revenue model is *exactly* what they have grown accustomed to.

DESTROY ALL LEGACY

[ebunga]

Wow, ARM people are just like Java cultists... calling everything else legacy.

Re:DESTROY ALL LEGACY

[Microlith]

Microsoft has long referred to anything not-Microsoft as legacy. It's nothing new to absolutists.

Dear redditors...

[Anonymous Coward]

There are no cat pictures here
There are no rage comics here
There are no advice animals here
All of our memes have been beaten down and we don't care for new ones.
We don't have English majors writing fake stories to have slashdotters ask them anything, except for our submitters.
There is no slashdot gone wild, thankfully.

So what are you doing here?

This comment is brought to you by shitslashdotsays

Re:Dear redditors...

[Anonymous Coward]

We're here. Get used to it. Welcome to the Monday that Never Ended.

Inflammatory story...

[ericloewe]

It's been pretty much proven that the "x86 legacy baggage" or however you want to put it does not seriously affect Intel's Atom for phones.

http://www.anandtech.com/show/6330/the-iphone-5-review/10

Razer i, which has an Atom processor, beats A6, the best performer in the ARM field, most of the time in non-GPU tasks (one area it is lacking is GPU power), while power consumption is average for a phone. Android adds additional overhead not present in iOS, too.

If anyone can work miracles and cram x86 into a phone, it's Intel. As ARM designs have to start dealing with greater complexity, Intel can apply their immense experience with x86 and improve performance without dramatically increasing power consumption.

With some more work, I can see Atom beating the hell out of any ARM design in the same power envelope. I'll give it one or two generations.

Re:Inflammatory story...

[thammoud]

That's some serious ass whipping. No power numbers though. At 7B per a new FAB, just wondering who has the muscle to compete with Intel.

Re:Inflammatory story...

[rsborg]

If anyone can work miracles and cram x86 into a phone, it's Intel.

Then why haven't they yet? It's not like they haven't been trying for years. Why did Apple (and Google) have to create the market that MS and Intel now feel they need to invade?

Intel is the challenger here, and ARM has proven itself several billion times over.

Re:Inflammatory story...

[rsmith-mac]

"Trying" is probably an overstatement in this case. Intel has a well-devised plan to get there, but it's a plan that involves them taking one step at a time. First they needed the Atom CPU design, then they needed to get it integrated into a true SoC, then they need to integrate their own GPU, etc.

Intel Atom roadmap

Silvermont is where Intel makes their architectural leap over ARMv7 (Cortex) with the new Atom architecture coupled with Intel's own, higher performance GPUs. Then in 2014 Intel does Airmont, where Atom gets promoted to first-class status in Intel's fabs, jumping to new process nodes at the same time as Core. If all goes to plan, at this point Intel will be roughly a node ahead of the competition with an architecture as good as or better than any planned ARMv7 designs. This is the tick-tock strategy in full swing, the same strategy that is currently bludgeoning AMD to death.

So Intel may be the challenger here, but never underestimate them. Their fabs are unrivaled and they can afford to hire some of the best architects on Earth. If Intel does their homework and doesn't screw up, they're a very dangerous foe. The only place Intel can't (or won't) go is into low-margin products, and as bad as competition from Intel would be, the ARM partners don't want to sacrifice their margins too much just to scare off Intel. It would be a Pyrrhic victory.

Re:Inflammatory story...

[Anonymous Coward]

Using chrome javascript benchmarks to compare ARM vs Intel processors is about the *worst* possible comparison you could make! It's entirely possible that Chrome's ARM javascript engine doesn't even use JIT for example. Unlikely, sure, but the benchmark really shows where most effort has gone into V8 or Kraken.

Re:Inflammatory story...

[foniksonik]

But who is going to buy their chips?

Not Apple. Not Samsung.

Intel doesn't have a major buyer lined up. They'll have to license the architecture out for pennies or make a new market.

Could be a case of too little too late unless they are willing and able to do low margin, high volume.

Re:Inflammatory story...

[Anonymous Coward]

and that brings us back to Windows 8... and for that matter, Android (Motorola has already shown they will use Intel)

Re:Inflammatory story...

[ericloewe]

I can imagine Microsoft adopting Atom as one possible SoC for Windows Phone 8 (or whatever version). Motorola seems to like the idea of an x86 phone (a Motorola 68k phone would be interesting, though). I can see other Android vendors who don't happen to have a SoC division gladly moving to Atom if it's cheap enough and tangibly better.

Re:Inflammatory story...

[ericloewe]

Power numbers follow in page 13.

Razr i has average to above average battery life and has an average battery, if I recall correctly

Re:Inflammatory story...

[shutdown -p now]

Pretty much every manufacturer of Win8 tablets will buy those chips. I would be surprised if it didn't include Samsung.

Genetic disadvantage? Hardly

[scheme]

'For all of Intel's semiconductor design and manufacturing feats, its processors suffer from a genetic handicap: They have to support the legacy x86 instruction set, and thus they're inherently more complicated than legacy-free ARM devices, they require more transistors, more silicon.

Intel and AMD x86 processors moved on to using micro-ops and risc like operations internally years ago. The only disadvantage nowadays is a small translator that converts x86 machine code into micro-ops. Compared to the actual logic or cache on the cpu the number of transistors that the translation takes is minimal and not a big deal especially when you consider the size of cpus nowadays.

Re:Genetic disadvantage? Hardly

[Anonymous Coward]

Yeah. The only people parroting this CISC vs RISC legacy nonsense are people who haven't picked up an architecture book in the past decade.

RISC lost all the battles, but won the war!

Re:Genetic disadvantage? Hardly

[Chuckstar]

EXACTLY. ARM's architecture may provide a slight advantage for low-power use compared to x86. But it's very, very slight. Certainly, Intel's advantage in process technology would outweigh ARM's advantage in architecture. The only real reason x86 hasn't competed with ARM so far in very-low-power is that no one has tried hard enough. There's finally enough demand for higher-end low-power chips that Intel is taking notice. I think Intel is also taking notice because they don't like seeing an ARM-based software ecosystem developing that could rival the x86-based software ecosystem. So it's as much a defensive play as a profit play.

Re:Genetic disadvantage? Hardly

[Anonymous Coward]

Intel's reason for not entering the market with an Atom-like SOC, which started in the 90's with the cancellation of Tinma, and killing several pentium M iterations, is that it would erode demand for their midmarket celeron offerings. The current Atom is basically an emergency "oh shit" product pushed out the door prematurely to counter the Arm11 and Cortex series used in smartphones and low cost laptops.

Intel is now deadly serious about low power x86, because if they don't act the company will be seriously dead in the consumer space with Arm v8.

Re:Genetic disadvantage? Hardly

[Anonymous Coward]

From the compiler point of vew CISC won the war, and that's what matters.

Re:Genetic disadvantage? Hardly

[Kjella]

Yep, I checked a little and the instruction decoder was about 5% of the die space on the first Pentiums, which means it'll take up 0.01% on a Ivy Bridge quad core. Even if you adjust for Even if you look at an Atom and adjust for the instruction set being much larger with 64 extensions, SSE etc. it'll still take up zero point something percent on a modern CPU.

Re:Genetic disadvantage? Hardly

[lkcl]

' Compared to the actual logic or cache on the cpu the number of transistors that the translation takes is minimal and not a big deal especially when you consider the size of cpus nowadays.

it's not the number of transistors that's actually so important as it is the number of times you have to change them from 1s to 0s and back. i read somewhere, so don't take this as gospel, that the register bank of any CPU takes something like a whopping 30% of a general purpose processor's power budget.

so a memory cache, 1st or 2nd level, would not be so power-hungry as a register bank, because you only change a few bits of any one cache entry at a time. a register bank on a 64-bit architecture, however, on average 64 bits are going to change on any two-operand calculation.

the point is that the instruction "translation", which includes shuffling the data around, has to operate at the "rated" i.e. the external i.e. the x86 clock speed. this *will* consume lots of power - i can't say how much: we need an intel engineer to tell us, and that's not going to happen.

bottom line is that AMD and Intel, with x86 "translation", are onto a losing game. intel only keeps ahead - one step ahead - by having access to geometries 1.5 times smaller than the competition. with ARM processors in 28nm being better, power-wise, than x86 processors in 20nm, and yet still offering "good enough computing" performance, both AMD and Intel are *definitely* going to lose out. AMD already is. Intel's just had a wake-up call: http://bit.ly/Ra0RIH

Re:Genetic disadvantage? Hardly

[The Finn]

The only real reason x86 hasn't competed with ARM so far in very-low-power is that no one has tried hard enough.

I wonder what VIA would say about that. it seemed like they had the atom-level market carved out before the atom appeared.

Re:Genetic disadvantage? Hardly

[sproctor]

Anand has a really good write up of the Intel instruction decoder at http://www.anandtech.com/show/6355/intels-haswell-architecture/6. It's significant, but who's to say ARM wouldn't be similar? I also think the article author is overstating the point.

Re:Genetic disadvantage? Hardly

[Plumpaquatsch]

'For all of Intel's semiconductor design and manufacturing feats, its processors suffer from a genetic handicap: They have to support the legacy x86 instruction set, and thus they're inherently more complicated than legacy-free ARM devices, they require more transistors, more silicon.

Intel and AMD x86 processors moved on to using micro-ops and risc like operations internally years ago.

Jesus Fucking Rodriguez! Using "micro-ops and risc like operations internally" means you can't be any more CISC.

Re:Genetic disadvantage? Hardly

[Chuckstar]

Would VIA point me to the broad range of VIA-powered phones and tablets on the market?

article author doesn't know what they are talking

[Anonymous Coward]

ARM processors are significantly smaller than Core processors, so they take less space on each wafer. So there's no question that there will be far more processors per wafer.

In addition, since they are smaller, a flaw in the wafer will damage a smaller amount of the wafer than with a larger processor.

So the percentage of processors that are good from a given wafer will be larger as well.

Now, there's still the question of if they could make as much money from the larger number of ARM chips as they do for the smaller number of Core chips, but remember that he's also quoting the retail price of the Core chips vs the wholesale price of the ARM chips.

Why would Intel be an alternative to Samsung?

[mridley]

I skimmed the source articles and I'm confused as to the premise of this discussion that it makes any sense to move from Samsung to Intel for ARM fabrication. Intel may have some foundry business but I would think if Apple were looking at an alternate foundry they would be considering options like TSMC or UMC, not Intel. Although doing some Googling to check my facts on this comment since I've been out of the semiconductor world for a long time does reveal that apparently Apple did consider TSMC. But that still leaves plenty of third party foundries.

Wrong Question

[Anonymous Coward]

Why isn't Apple constructing their own chip fab?

They have phenomenal cash reserves & the claim has been made they designed their own chip. So they have the market, the cash & the know-how to design what they need. Why not leverage it?

Is chip fabrication really so far outside the core competency of Apple? Is the investment really that risky? Or is the service of chip fabrication that competitive that there's really no advantage to investing in a plant when the roadmap for Apple's chips requires flexibility that would be impinged upon by an allegiance to specific hardware?

My guess is that there's no security in the chip market for the short term, unless you are Intel... but with AMD laying off 15% of its workforce, who knows what's in the future for Apple chips.

Re:Wrong Question

[MarioMax]

Why isn't Apple constructing their own chip fab?

That's actually a damn good question.

My only guess is because it still makes more economic sense for Apple to outsource chip manufacturing to a third party. Brand new fabs are about the opposite of cheap as it gets: on the order of $30,000-$50,000 per square foot of cleanroom for the factory alone, including the actual construction cost and tooling the factory, but not counting labor costs and engineering support needed to run the factory, and not counting annual upkeep (which probably approaches 10% of the build cost per year alone).

Re:Wrong Question

[Anonymous Coward]

Because all of the really good engineers they would need to pull off such a move...

Saw what happened to woz and tend not to wanna play with apple.

In the end

[SuperMooCow]

And after centuries of devastating battles, both went extinct. However, amidst the ashes, some of the little AVR cores survived and replaced both. ...

New for 3012, the ATmega81928 running at 30MHz! The speed will blow you away!

Comment removed

[account_deleted]

Comment removed based on user account deletion

All right, let's compare more

[Chemisor]

If you measure miles per hour, the horse will win. If you measure miles per hour per calorie, the snail will win. Now, which one would you like to pull your next carriage?

Re:All right, let's compare more

[Fishchip]

It depends. Is my carriage playing Modern Warfare 3 or Angry Birds?

Re:All right, let's compare more

[bennomatic]

How many snails am I allowed to rig up? And can I put them on a geared set-up which will allow them to achieve horse-like speeds?

Re:All right, let's compare more

[runningduck]

Or which one would like to be on the menu of an up-scale restaurant?

Re:All right, let's compare more

[mjwx]

If you measure miles per hour, the horse will win. If you measure miles per hour per calorie, the snail will win. Now, which one would you like to pull your next carriage?

A Nissan 370GT uses 11 L/100KM, a Nissan Micra uses 6.5 L/100KM. Both will do the same job but the Micra wont got 0-100 in 6 seconds. So it depends if I'm racing or saving fuel.

Re:All right, let's compare more

[Anonymous Coward]

As many as you'd like. By the time you got up to horse-like speeds, you'd be burning more energy per watt. The failure of the analogy is that the horse starts off with a more efficient movement method, whereas in the case of ARM vs Intel, ARM has the theoretically more efficient architecture. It should compare horse vs. spider, but if I had a carriage pulled by a bunch of spiders I'd be running a lot faster in the opposite direction, screaming as I go. Hey, the analogy holds up.

Re:All right, let's compare more

[LordLucless]

It depends. If I only need to move a fraction of a centimetre, I'll go for the far more efficient, and better suited for the job, snail. Also, what's the relative price of horse vs snail food?

The reason ARM's getting a look-in now is that many devices are over-engineered for the uses people are putting them to. The cost of power in desktop computing is going up, and obviously, the power budget in mobile devices is becoming more and more crucial as sheer processing power becomes overkill.

Re:All right, let's compare more

[Anonymous Coward]

If you measure operations per second, the x86 chip will win. If you measure operations per second per watt, the ARM chip will win.

If you measure miles per hour, the horse will win. If you measure miles per hour per calorie, the snail will win.

Wrong units. "miles per second" is a valid substitute for "operations per second", but "calorie" is an invalid substitute for "watt". A watt is a measure of power, which is a rate of energy consumption, whereas a calorie is a direct measure of energy. You should use calories/hour instead; or just cancel the units, and evaluate the snail in terms of miles per calorie.

Re:All right, let's compare more

[BlueLightning]

In France, both.

Re:All right, let's compare more

[Anonymous Coward]

If you measure miles per hour, the horse will win. If you measure miles per hour per calorie, the snail will win. Now, which one would you like to pull your next carriage?

Whichever one will have me cleaning up less horseshit.

Re:All right, let's compare more

[bennomatic]

I love the image. I'm picturing a horror version of the Santa Claus myth: "On Creeper, on Crawler! On Bitey, on Spinny!"

Re:All right, let's compare more

[Da_Biz]

How many snails am I allowed to rig up?

I, for one, welcome the creation of a Beowulf cluster of these snail-pulled carriages.

Intel should preparel x86 replacement

[faragon]

This article shows the obvious: Excluding caches, performance per transistor in Intel x86 CPUs is very low. As example, current best performance per clock in the Intel CPUs is the AVX (Core i5/i7 -Sandy and Ivy Bridge-), delivering up to 8 FLOP per cycle with AVX SIMD opcodes (2 SIMD ALUs) while in previous generations was just 4 FLOPS per cycle with SSE2/3/4 (just 1 SIMD ALU). Thats miserable (back in 2000, the Playstation 2 was already capable of FMAC opcodes with 8 FLOPs/clock per SIMD ALU!!!). As example, similar performance with 4 FLOP per cycle with one SIMD ALU, at one fraction of waffer area.

Here is a 50$ ARM general purpose multicore-CPU example for matching 999$ performance of fastests Intel Core i7 (e.g. i7-3770K 3.9GHz (peak), 4 CPU, 8 threads, 2 SIMD ALU/CPU = 8 SIMD ALUs = 64 FLOPs/clock -> 3.9*10^9Hz * 64 FLOP/s = 249.6 GFLOPS:

• 4 x ARM OooE (e.g. Cortex A9-like) 2.0GHz with 2 SIMD FMAC-capable ALUs/CPU (ALU = 16 FLOPs/clock, i.e. 2 ALUs = 32 FLOPs/clock -> 4 * 2.0*10^9 * 2 * 16 = 256 GFLOPS
• 4 * 32KB + 4*32KB (256KB) L1 full-speed code and data cache
• 4 * 256KB (1MB) L2 half-speed cache
• 2 MB L3 half-speed cache
• 2 or 3 lane ring bus (cheaper interconnect).

For increasing integer and load/store performance, it could be achieved with pipeline and issue/execution modifications, using more functional units. The limit is to keep the OooE simple enough for avoiding wasting transistor in executing tons of instructions unnecesarily.

Re:Intel should preparel x86 replacement

[Anonymous Coward]

FLOPS alone seems like a poor metric. If one uses that as a basis for comparison, then all CPUs are blown out of the water by a very weak CPU paired with a strong GPU.

What sort of benchmark would be more appropriate for this sort of comparison? Some sort of combined server/desktop/tablet workload?

Re:Intel should preparel x86 replacement

[Anonymous Coward]

Yes and the Peak FLOPS/transistor is even higher in a GPU, but no one sticks those in their phones either. Most real performance gains come from figuring out how to use the peak FLOPS ability, not simply increasing calculation ability at the expense of everything else.

Re:Intel should preparel x86 replacement

[Anonymous Coward]

Get back to me when the first ARM PC, not tablet, is mainstream.

Re:Intel should preparel x86 replacement

[Anonymous Coward]

excluding caches? are you mad? the performance gap between the cpu
and memory is huge. cache performance is if anything more important
than cpu performance. performance per transistor is not relevant. performance
per dollar or per watt is, and neither is directly coupled to performance/transistor.

if a calculation happens on a cpu, and its results can't be pushed to memory
(in time) did it really happen?

Re:Intel should preparel x86 replacement

[Anonymous Coward]

In theory? Yes. In practice? Not at all.

Your theoretical high IPC depends on high quality code and high quality compilers. In practice, it's been proven that neither of these exist.
VLIW, RISC, EPIC - All of these fell to x86 for the same reason. Nobody could write code and compilers that would bring real world performance close to theoretical. It's why intel arch chips rule today.

Think of it this way:

Honestly, most code is crap. Bloated, abstract languages with many layers and many libraries.
Intel simply makes chips that are very good at running bad code.

Re:Intel should preparel x86 replacement

[RocketRabbit]

Talk to Microsoft. They're the ones who have threatened to withdraw Windows OEM volume pricing and force full retail prices on any manufacturer that offers an ARM-based PC maker that produces anything but a Netbook.

Apple will be the first PC maker to put ARM into a full fledged PC, because they don't have to fear Microsoft's threats.

Intel must embrace ARM

[gymbrown]

Long-Term Survival Depends On Intel Embracing ARM

There have multiple organizations that had the capability of competing in the next market place and chose to ignore it because there was more profit in the last market place. The organizations lost market share and shrank as the old market place shrunk and the new market place expanded.

The following examples come to mind:
The Swiss started electronic watches but mechanical watches were more profitable
Kodak started designing digital cameras but film made more money.
Remington made computers but their consumer division was their main business.

Someone will make the next generation of processer chips. Intel will make less money in the short term if they decide to join the club but will eventually shrink if they don’t.

Explain this one to me...

[Qubit]

Intel has some of the best (the best?) fabs in the world, and has chips that use a smaller process than what other companies are pushing out, right? So why can't they make a small, power-efficient chip that can at least meet (if not beat) the offerings from ARM and the licenses?

To put it another way, Wikipedia tells me that ARM Holdings has 2,000 employees and a revenue of about 490 million pounds (in 2011). Intel has 100,000 employees and a 2011 revenue of 54 billion dollars (about 34 billion pounds). How hard is Intel really trying to get a foothold in this market?

Easy question: How many smartphones are shipping right now with an Intel chip in them? Or to make it easier for you, can you name a currently-shipping smartphone that *doesn't* run on an ARM chip, without googling it?

Another metric: Can you think of any form factor of computer (laptop, netbook, desktop, server, game console, smartphone, dumbphone, tv, settop box, etc...) in which x86 marketshare has increased in the last 5 years? (I think apple's ppc -> intel move was 2006, so no dice there :P )

Re:Explain this one to me...

[MarioMax]

Intel has some of the best (the best?) fabs in the world, and has chips that use a smaller process than what other companies are pushing out, right? So why can't they make a small, power-efficient chip that can at least meet (if not beat) the offerings from ARM and the licenses?

From what I've read on AnandTech, low power Haswell chips might meet your criteria, which are due out the middle of next year. I'd be very surprised if Broadwell (the 14nm die shrink of 22nm Haswell) doesn't.

Re:Explain this one to me...

[marcosdumay]

SO, they promisse that the chip they'll release next year can compete with the chips ARM is selling* now? (And, yeah, that would be the first time Intel overpromissed on power consuption... Forget about last year, and the year before that, and...)

* Ok, ARM doesn't sell, I know. but they are getting done now, with inferior processes, bigger feature sizes and are still competitive with the offering Intel has for the next year.

Re:Explain this one to me...

[shutdown -p now]

They already have a chip that's competing and selling now. It's called Medfield.

that old canard about x86 complexity

[epine]

The old 80386 based on the "complex" x86 instruction set had 275,000 transistors. Intel is now making chips with 2.6 billion transistors and somehow what they once implemented as one functional unit within a budget of 0.000275 billion transistors is holding them back?

Certainly they would rather do a few things differently had they been worried about 2013 back in 1978. Transistor count is the least of the matter. What buggers up x86 is the number of active transistors handling the instruction stream at each instruction cycle. There's no way to align variable-length instructions without active transistors (regardless of whether the transistors involved amount to a wart on a small toe of a juvenile mosquito).

The x86 story bugs the hell out of me. Considered how well it actually held up for 45 years (and counting) it's one of the ugly duckling success stories of all time (hint: it wasn't so ugly after all).

It was also a founding member of the Steve Jobs reality distortion field. I'm concerned his posthumous aura will continue to glow with the uplift of falsehood. He should be credited more for what he accomplished than the lies he polished to get there.

It wasn't just Steve, it was the entire RISC consortium manufacturing an Achilles heel out of whole cloth. Far closer to the truth of the matter is that x86 has a much higher design cost than an orthogonal clean-sheet alternative. The design cost was a small multiple. Intel's resources were a large multiple. It didn't go well for RISC. The much vaunted DEC Alpha had a metal connect layer for single-cycle carry-add propagation that forever segregated it from the mass-consumer price point. It was the instruction set. No, it was the instruction set aided by a titanium stent.

Also, the RISC design advantage does not extend to the memory cache and system bus design. These are a bear to design well for any instruction set. The RISC people moaned about the exceptional Pentium Pro performance level on server workloads (it was the first memory bus from Intel that didn't totally suck). Well, Intel broke into the server market with their crappy old x86 instruction set by grafting it onto a titanium alloy cache hierarchy and bus controller (with multiple dies grafted into the same chip package at enormous expense). Cache latency and branch prediction absolutely dwarf instruction set as the big thing to worry about since around this time. If Steve hadn't grabbed onto the inferiority of CISC around this time, it might have died a timely death.

In low power applications, ARM has a real advantage, enough to win a huge market share at race-to-the-bottom price points. How much does the cost of a CPU influence a handset? How much everything else? I've put $300 Intel CPUs in $2000 boxes. I've put $250 Intel CPUs in $1000 boxes. I've put $60 CPUs in $500 boxes. A $16 CPU in a phone that retails for $600 for just a few months, before landing in the discount bin? I'm sure Intel wants a huge slice of that.

One reason Intel has held their ground is that the Cortex-A15 (out-of-order superscalar multiprocessor) is starting to look a lot like the old Pentium Pro. Sure the instruction set is modern and clean (though it took ARM surprisingly long to come up with the mixed 16/32 bit instruction encoding format due to misguided ideological purity; how many active transistors does it take to determine whether the next 32 bit chunk from the instruction stream is one lump or two? More or less than the number of active transistors in the icache devoted to storing common instructions bloated to 32 bits just because?). But all the rest of the issues are pretty much the same: branch prediction stalls, cache snooping, and memory path latency.

From Intel's perspective, an ugly instruction set is good for business. (Then they went on a jag thinking that if ugly is good, atrocious is better, and the Itanium was hatched with a jackhammer from a mastodon egg.)

After another three die shrinks, when half the processor implements on-demand power management, and most of the other half provides task-specialized execution units, is the instruction set going to matter a hill of beans for anything other than legacy lock-in?

Re:that old canard about x86 complexity

[Animats]

Far closer to the truth of the matter is that x86 has a much higher design cost than an orthogonal clean-sheet alternative.

True. Years ago I went to a talk where the head of the Pentium Pro design team showed a graph of the number of engineers working on the project. It peaked around 3,000. Nobody had ever had a CPU design team that big before.

The variable length instruction alignment problem of x86, although ugly, isn't a huge consumer of transistors. AMD dealt with it by expanding instructions to fixed length when loaded into cache. Intel dealt with it by sometimes starting ambiguous cases in parallel and discarding the bogus results later. The downside of fixed-length instructions, as in RISC machines, is code bloat - PowerPC code is about twice as big as x86 code, which impacts cache miss rate.

While one instruction per clock RISC CPUs (low-end MIPS and DEC Alpha parts, and the Atmel AVR series are examples) are simple, superscalar machines executing more than one instruction per clock are almost as complex as x86 CPUs. That's why RISC stopped being a win.

Harry Pyle was developing the instruction set for the Datapoint 2200 in his dorm room at Case Tech in Cleveland in the late 1960s. Same building I was in; different floor. That led to the 8008 and the 8080 and the 80286 and the 80386 and ...

Re:that old canard about x86 complexity

[Anonymous Coward]

Also, the RISC design advantage does not extend to the memory cache and system bus design.

Sure it does. ARM has a weakly consistent memory model, while x86 in most cases has a strong one. ARM mostly supports memory access on natural alignments, while most x86 instructions can load or store to any alignment.

Sure the instruction set is modern and clean

I think the switch to 64-bit computing can be something of an equalizer here. On one hand, the AMD64 instruction set added sorely needed registers and separated away a little bit of the cruft, and on the other hand ARM is adding yet another instruction set to support 64-bit in ARMv8.
Some people seem convinced that 64-bit ARM processors will herald a new age of low-power ARM servers. There's some momentum building that way, but I'll hold off judgement until I see the real-world numbers.

It's all about code parallelism

[__aailob1448]

For a long time, single-core applications were the rule so the CPU Mhz race was on. Once that ended around 3Ghz, the pressure was on for programmers to make computer code better at dividing the load between multiple cores.

It turns out that ARM does well with lower frequencies, and delivers the best performance per watt ratio. Also, it turns out that once all your code is written for 2, 4 or 8+ cores, it doesn't matter much if your cpus are clocked at 1.3Ghz (A6/Snapdragon) instead of 2.6Ghz (i7 in macbook pro 2012).

And if you're doing mobile, where battery life is a big factor, you need the ppw ratio more than anything, so you go ARM.

On mobile, Intel is in a similar situation now that they were against AMD back in the AMD64 days. Their current models (atom) are inferior but competitive. They are dominating servers and desktops which gives them a secure base to experiment from and I expect their mobile offerings in the next 5 years to bridge the gap with ARM.

Will they win? I have no clue. They might crush ARM or become the premier ARM licensee with the best ARM chips. Either way, Intel is going to lead.

Apple doesn't want to be *more* dependent on Intel

[steveha]

Intel wants to be the only company that can meet your needs. That way, they can make you pay premium prices for their chips. This is perfectly understandable; that is what is best for Intel.

Apple wants to be vertically integrated. They want full control over everything they do. Partly this is so they can keep as much as possible of the money they collect; partly this is so that they can guarantee excellent quality and excellent availability. This is what is best for Apple, and it isn't bad for their customers either.

Intel does not want to become just another ARM source, competing on price with all the others. But Apple will never lock themselves in to depending on Intel for mobile chips, when ARM chips have been shown to be more than adequate. And Apple would not be investing in custom ARM chips if it was planning to adopt Intel mobile chips.

People keep pointing out that Intel's mobile x86 chips are competitive with ARM. That won't cut it. Intel's chips would have to be better, and so much better that the risk of depending on Intel is worth it.

That was the case for the PowerPC to x86 transition! Intel's chips were so much better than PowerPC for laptops that it was worth getting into an entangling relationship with Intel. AMD was not able to guarantee delivery of the massive quantities of chips Apple was planning to sell, and Intel was, so AMD wasn't really an option... but at least they served to keep Intel from trying to charge totally outrageous prices for their chips; there was always a credible threat of going to AMD.

Hmm. It's looking like AMD is going to crater in spectacular fashion soon. I wonder if Apple will make a serious attempt to buy what's left of the company. That would enable Apple to make its own x86 chips! Eh, probably not. AMD is behind Intel on process, so switching to AMD chips would mean taking a hit on performance, power use, or both.

The "SemiAccurate" web site thinks that Apple will transition to using ARM chips for laptops, not just for mobile devices, once ARM chips are good enough (which they will be soon). So, transitioning away from x86 and to, say, multi-core 64-bit ARM chips is another way Apple can untangle from Intel.

Apple may not be in a big hurry to actually complete the transition away from Intel chips; just a credible threat of switching to ARM chips might be enough to negotiate good prices on x86 chips. That would leave lower power consumption as the main reason to go to ARM, but a laptop's display is probably the worst power drain, especially with a Retina display.

steveha

How many petaflops does your tablet do?

[WillAffleckUW]

Mine flops and flips so much it burns out the battery in one minute.

Petaflop score: 1 Trillion

usefulness score: Zero.

Look, we buy these things to do stuff for us. A dead phone or tablet is pretty useless, no matter what it scores.

Tis a fool....

[Darth_brooks]

Tis a fool who looks for logic in the chambers of the human heart. Or from Cupertino. And that's not a dig, Apple fans, that's just the truth. Apple will dump Intel when they feel like it, for reasons that they alone decide.

Apple is a bit like the interrogator in 1984. They believe that can levitate off the ground and float around the room should they choose to, and what the outside world thinks makes no difference at all.

Re:Tis a fool....

[mjwx]

Tis a fool who looks for logic in the chambers of the human heart. Or from Cupertino. And that's not a dig, Apple fans, that's just the truth. Apple will dump Intel when they feel like it, for reasons that they alone decide.

Apple is a bit like the interrogator in 1984. They believe that can levitate off the ground and float around the room should they choose to, and what the outside world thinks makes no difference at all.

This.

It's Intel looking at the big picture. Samsung was one of Apple's biggest suppliers, look at what Apple tried to do to them (although it did backfire horribly for Apple, you cant count on that happening every time). Apple is turning out to be a riskier partner than Microsoft was.

Re:Apple doesn't want to be *more* dependent on In

[steveha]

Also note that Apple has people paying $2500 and up for the Mac Pro, and $1000 and up for laptops. But mobile devices are closer to $500, and the Android competition is hitting the $200 price point.

There just isn't as much room to pay top dollar prices for Intel parts in the mobile space.

So even if Intel mobile x86 parts are slightly faster than the ARM chips, will Intel be happy selling at prices competitive with ARM prices? History suggests "no". The cheapest Atom chips are around $20 but Intel makes those suck, just as much as Intel can get away with.

Intel is the master of segmenting markets. Different chips at different price points have different features enabled. Cheaper chips are as crippled as possible, to encourage you to buy a more expensive chip. For example, Intel doesn't support virtualization features on their less-expensive chips; and Intel mostly reserves support for ECC RAM to only the Xeon processors.

(In contrast, AMD puts full functionality in all their parts; they are #2 and they are trying harder to please the customer. That is how you can get an HP Proliant MicroServer with a 1.5 GHz dual-core AMD Turion processor for $320 at Newegg, with full support for virtualization and ECC RAM. I cannot imagine a MicroServer with equal or better Intel parts hitting that price point.)

Intel will try to balance the functionality it allows into the mobile chips against the price it can get. Apple just wants the best chips for the cheapest price. These two goals are not in alignment.

That's kind of stupid

[Anonymous Coward]

Apple should acquire AMD, and shift them to being primary supplier and ARMs dealer.

That's rather dumb considering AMD doesn't actually manufacture chips any more.

If AMD did make chips, Apple could get nice GPUs and license ARM cores. But then if AMD hadn't dropped the ball on manufacturing they might still be a viable company.

Re:Apple doesn't want to be *more* dependent on In

[inputdev]

Intel is the master of segmenting markets. Different chips at different price points have different features enabled. Cheaper chips are as crippled as possible, to encourage you to buy a more expensive chip.

This is enough for them to fail. It is a gross waste of resources to make crippled hardware. I think they have enjoyed a monopoly position and they are finally facing a disruptive technology.

Just realized...

[gr8_phk]

Intel has the best process engineers in the business and if things in the foundry business keep going like they are (TSMC and Global Foundaries have both been very very late moving forward on process while Intel hasn't missed a stride) they are going to be two steps ahead on process in the next year or two and that would be an advantage not even the best ARM design could beat even if Intel bungles their design.

Intel is not playing the same game. With everyone else (except AMD) making ARM devices on older process nodes, Intel should not make ARM chips because that would create the perception of competition and force TSMC and GF to advance their process. So long as all the foundry customers appear to be competing with each other it looks like a close race and there may be less pressure to advance. The further Intel stays away from their products, the less those guys will feel like they are competing with Intel and they will not worry about the process gap so much - they're still close to their "competitors" capability after all.

Everyone seems to have forgotten what business they're in. Those who can design have gone fabless while those who can fab now have more than enough customers to not care about process advancement so long as they can keep up with their perceived competition. In fact, all those customers probably slow them down with countless designs that each need scheduling and a design tweak or two. Meanwhile Intel turns that crank every 2 years. The longer people forget that they're all in the same business, the wider the gap is going to get.

If you can figure out ...

[PPH]

... the politics between Apple, Samsung, Microsoft, Intel, ARM and others, you should be working at the United Nations on a solution for eternal world peace.

There is so much sub rosa crap (not all of it ethical or legal) going on between the players we may never know the truth.

TSMC

[Roogna]

Don't know what Jean-Louis is talking about, as there were press releases and everything not long ago about Apple ramping up production at TSMC foundries. Don't think they feel they need Samsung or Intel for their ARM production.

Re:Performance wise - x86 still beat the hell

[Anonymous Coward]

And the i7 will last all of 30 seconds.

to be fair, arm has legacy as well

[Anonymous Coward]

a lot of cruft has cumulated into the current arm instruction set.

TSMC and Global Foundries

[tyrione]

Two reasons Apple doesn't give a shit about Intel for their embedded products.

Even Nokia was caught napping.

[Anonymous Coward]

Intel may go the Nokia way as falling dino

Re:article author doesn't know what they are talki

[Anonymous Coward]

Intel's 22nm and 14nm finfet processes will be the lowest power in the industry for years. Intel could charge Apple $50 per chip and they'd pay for it.

more research needs to be done

[Anonymous Coward]

Intel would have to accept Apple's design and 'pour' it into silicon — it would become a lowlymerchant foundry.'

if they did research, they would know intel already took steps to becoming a merchant foundry around 3 years ago

why?

intel knows the more they shrink the process the more chips they can make
unfortunately there's a ceiling to how much you can actually sell to a market
couple that with a move to 450mm wafers and intel will have a lot of spare capacity
becoming a 'part' merchant foundry will ensure they make use of their assets

INTEL CPUs arent even X86 any more Jean

[cheekyboy]

95% of the intel cpus are not even made for IA64, its all RISC style microcode.

And only a small front end decodes IA64/32 into microcode, kind of like a JIT bytecode engine, so in theory, intel could make a small frontend for ARM, and thus make all intel cpus naturally ARM compatible, or PPC if they wanted to be the next supplier of CPUs to XBOX and PS4 , they could make a special i7 with a PPC IS decoder to internal microcode. And if it could swap between two sets in runtime even better.

Replay of the mainframers

[Ancient_Hacker]

This could be a replay of the old days of mainframes. At more than one company, the engineers came up with mainframes on a desk, but the marketers could not see selling a desktop mainframe at the old 7-digit prices. So they just making the big boxes, til their eventual death. This happened to CDC, Data General, Digital, and Perkin-Elmer to name a few. Intel will undoubtedly survive, but it could be a long painful decline or change of direction. The "new architecture" fanatics there probably don't have much traction after the Itanium disaster.

intel will never sell 20$ processor. its not there

[Anonymous Coward]

market.

The Average Commuter Train Is RISC-Based

[Anonymous Coward]

When I rode the train back home every second fuck had a an ARM based device. So RISC won, CISC is the hot legacy shit that burns my hand at work.

Intel will go the way of IBM.

Tons of Register Aliases

[Anonymous Coward]

..only makes power consumption worse for Intel.

Like IBM

[Anonymous Coward]

..Intel is scared of a low-end, much cheaper upstart (ARM) and they want to artificially protect their fat margins. Look how it worked out for IBM and their mainframe business. All they protected is a tiny market by now.

That'l Be January 1st, 2013

[Anonymous Coward]

That's when they Big Tablet Keyboard Boom has reached your ears. Large Tablets+USB Keyboard can already perform all of the tasks of 90% of office workers. Without burning their hands and frying their balls

Only True If

[Anonymous Coward]

..you have a wall power socket. Intel-based computers are more heating machines than computers. That does not work at all with mobile devices and the crappy performance of laptops proves it.

Re:Only True If

[RocketRabbit]

It's true - since fall started, I kept the door in my office room closed and the heat off. The only thing running in here is a Mac Mini and it heats the room to a comfortable 66Â just doing household media center type duties.

To be fair, it's more cost effective per BTU than a space heater.

Did You Describe IBM ca 1987 ?

[Anonymous Coward]

Then they were the Undisputed King Of Computing. They had everything from semiconductors, multi-chip-modules, nobel prizes, sql databases to efficient machine code translators and still got crushed by the sleazy little Bill Gates and his horizontal and "little" model of computer making.

Compared to ARM, Intel is a vertical company, doing all sorts of expensive crapola (compilers, supercomputers, virus scanners) while ARM just designs efficient CPUS. Unlike Intel, they don't have the millstone of fab process development around their necks. Instead they use TSMC and Samsung fabs which they share with 1001 other companies. Great economics I would say.

Also, Intel is stuck in the "high performance monopolist" mindset, as much as IBM was stuck in the "high performance, expensive mainframe" mindset then.

Re:Apple doesn't want to be *more* dependent on In

[RocketRabbit]

"a laptop's display is probably the worst power drain, especially with a Retina display."

The majority of power used in a modern display is consumed in lighting the thing, not switching pixels.

power numbers

[Anonymous Coward]

Once you get below a certain point, it doesn't matter anymore. The rest of the system still needs power. If the rest of the system uses 1000 units of power, we don't care greatly if the CPU uses 2 or 10.

revenue isn't the right number

[Anonymous Coward]

ARM doesn't actually produce chips. Either add in the chip companies for ARM, or subtract the manufacturing costs from Intel. Also note that none of the chip companies, Intel included, is just about CPUs. ARM is just CPUs.

Re:Apple doesn't want to be *more* dependent on In

[steveha]

Check out this link: http://www.informationweek.com/byte/personal-tech/tablets/why-the-ipad-3-regressed-in-battery-life/232602960

To achieve the mind-boggling quad-XGA resolution of 2048 by 1536 pixels using the display technology available today, Apple had to more than double the power draw of the LED backlight that lights up the iPad screen. According to DisplayMate, Apple had to bump up the backlight power draw from 2.7 watts on the iPad 2 to 7 watts on the new iPad.

As DisplayMate explains it, that huge bump in power consumption in the LED backlight was caused by the use of amorphous silicon type LCD panels whose transistors block out more light when pixel density increases. The iPhone 4 in contrast has even higher pixel density--but its use of low temperature polysilicon (LTPS) technology makes the iPhone 4 more than twice as energy efficient per square inch to achieve the same brightness as the new iPad. The problem is that LTPS technology is expensive and would not have been practical for a screen as big as the new iPad's. A new display technology called indium gallium zinc oxide (IGZO) has the benefit of better energy efficiency at a competitive price, but IGZO isn't ready for mass production.

This means we are both correct. The majority of power use is just to light the display, and with a Retina, more power is needed.

But technology is on the horizon that should someday permit a Retina display that doesn't need as much brightness and thus saves power.