How Intel’s Medfield will dismantle ARM | Hacker News Reader

How Intel’s Medfield will dismantle ARM

extremetech.com

63 pointsevo_967 comments

The article overstates the case against ARM in the mobile marketplace. Counterpoints not mentioned are: 1. The single company Intel can never match the wide variety of SoC (System on a Chip) components that the diverse ARM fabricators will offer. 2. The ARM fabricators will remain close behind Intel with regard to lithography generations. 3. Intel's Medfield has not been independently benchmarked on shipping mobile devices. 4. Industry partnerships established for the previous Intel mobile part did not create any notable results. 5. In order to dismantle ARM, Intel would have to significantly underprice commodity ARM fabricators - which would destroy Intel's historical margins. Note that Intel sold off a former ARM-compatible division - XScale - for this reason. 6. Furthermore some ARM partners plan to invade the x86 server market with low power, but capable alternatives to Intel servers, thus enlarging the potential ARM market and possibly counter-disrupting Intel's highest margin business.

Furthermore, the major proprietary player (Apple) will likely never give up on their ARM chip to favor Intel. Unlike in the 80s where IBM used Intel chips and Microsoft software and created two heirs to their monopoly, this time will be wildly different.

ARM (and the mobile market) will crush Intel. Intel is seriously late to the game here, and never took performace/watt seriously for decades, and ARM is going to disrupt them hard.

Interesting that you mention Apple/ARM -- don't forget that, eventually, Apple had no choice but to shift from PowerPC chips in its laptops and desktops to Intel. The performance difference was just too vast.

I am fairly sure that the same thing will now happen with x86 and ARM -- it just depends on the timeline.

That was rather different, as Apple was entirely dependent on IBM and Motorola to produce new chip designs (and Motorola had effectively quit the desktop market). The G5 was actually reasonably competitive with the P4 at the time the move was made, and beat it on some SIMD work. The real trouble was that IBM had repeatedly slipped on its deadlines to produce a low-power laptop-appropriate G5, and the aging G4 architecture just wasn't good enough.

Today, the story is very different. Apple can produce its own SoC designs, as with the A5, or even produce its own ARM _core_ designs, like Qualcomm does, or it can buy from one of a number of healthy vendors who show no signs of quitting ARM anytime soon. If this Intel core had a genuine dramatic performance advantage over the A15 (which it will be competing with, given the timescales), at the same power usage, that would be one thing, but that's far from clear right now.

Apple had to move to x86 because IBM wouldn't invest in development of a fast and low power processor for just one client (Apple) and Intel had a product (Core Duo) ready for them. One client, even Apple, is a very small market. Unit sales of ARM processors combined dwarf x86 sales and there is a huge market for higher performance ARM devices, something that didn't exist for PPC.

That's correct. If you remember, Tim Cook described this limits on PPC based products as a "thermal problem" which is what drove them towards Intel. Presumably, ARM's licenses don't have this problem, in part, as @rbanffy implies, because the ARM architecture doesn't rely on a single source as the PPC did.

It wasn't just that Intel chips were faster. They also consumed less power. Just before the switch, Apple shipped a G5 desktop Mac with liquid cooling built in, which shows just how desperate the situation was for them.

Things are much different now. Intel chips are faster than ARM, of course, but they are also vastly more power hungry. Power consumption matters more than speed in most devices these days (otherwise Intel would already be in everybody's phone) and will likely continue to do so for a long time. Intel shows no sign of being able to reduce their power consumption to the level needed while still retaining decent speed, whereas ARM shows every sign of being able to keep increasing their speeds while keeping power consumption where it needs to be.

Is that 'They also consumed less power' correct?

http://datasheet.digchip.com/205/205-00367-0-970.pdf gives typical power dissipation as 42W @1.8 GHz, 1.3v

For a later design, http://datasheets.chipdb.org/IBM/PowerPC/970/PowerPC-970FX.p..., section 3.1.5 gives 60W for the 2.2 GHz version at 105 degrees Celsius.

http://en.wikipedia.org/wiki/List_of_Intel_Pentium_4_micropr... gives comparable values for the TDP for 2002-era Pentiums.

I get the impression that IBM's data is real max (but 'Maximum power at 85 C is the ONLY value that is guaranteed by Manufacturing Test.'), while TDP may be lower than max (Reading http://en.wikipedia.org/wiki/Thermal_Design_Power)

Or am I overlooking a factor for #CPUs?

I always thought the water cooling was either an effort to decrease noise or over engineering or preparation for more cores and/or more speed (either of which were a necessity to keep up with Intel chips).

I think you're comparing things a little too early. In 2002, things were still very much up in the air. In hindsight, the PowerPC first started showing signs of real trouble when Apple had to de-rate the entire Power Mac G4 line by 50MHz between the initial announcement and the first ship, but it still did fairly well for quite a while. It wasn't until close to the switch when the promised 3.0GHz G5 failed to materialize that the differences really got big. The Intel switch wasn't announced until 2005 and didn't happen until 2006.

I don't have hard data, but as far as I know, once Macs went Intel, they all got significantly faster and cooler.

This was most visible at the top of the line. The last Power Mac G5 had the liquid cooling system and the top end ran four cores at 2.5GHz. I don't know about the liquid cooled model in particular, but the G5s in general were notorious for being noisy due to fans.

The first Mac Pro which replaced it ran 8 cores at 3.0GHz and was known to be astonishingly silent. I personally used one as my main work computer for years and it was impossible to get the fans to run fast enough to be audible unless you disabled the smart controls by e.g. leaving a kernel panic on the screen for a while. Obviously, this could be due to more advanced thermal design as well, but even allowing for that, you have twice as many cores, running at a higher clock rate, vastly faster in most real-world use, and with a much simpler cooling system.

The laptops had a similar story. The PowerBook G4s would get astonishingly hot, while the MacBooks were much less bad (although still not great initially).

Granted that the graphs are coming from Intel, but assuming they're accurate I'd say it's a sign Intel can reduce power consumption and retain decent speed.

http://www.anandtech.com/show/5365/intels-medfield-atom-z246...

Maybe they can, but I'll believe it when they see it. Going by past accomplishments, ARM has done very well in making their chips faster, and Intel has not done so well in making their chips more economical. Certainly there could be a change in the game, but precedent heavily favors ARM in this one.

I think that #1 is the more important "pro" for ARM, while #4 is the most important "con" for Intel.

How many times have we seen Intel announce a hardware partnership with a company (especially in mobile) that ultimately went nowhere? I think this comes down to, I'll believe it when I see the devices shipping.

Two other things:

* ARM's big.LITTLE promises to drastically reduce power consumption under low CPU load. It seems unlikely Intel will be able to offer something similar anytime soon.

* If you compare Atom Z2460 with SoCs based on Cortex-A15 (both are supposed to start selling by the end of the year), it doesn't look that well.

Yeah, due to the fact they're out-of-order, the A15s will should be back on top. When the Pentium Pro was introduced with OOO pipeline, it got a 20% speed increase over similarly clocked Pentiums. The issue might be can the A15s keep their pipeline full with such a small cache and memory bandwidth. And how good (if it even has one) is the pre-fetcher? If the float units can't be given data fast enough, it's not going to matter how fast it is in theory.

Note that OOO isn't a boolean checklist feature. It's a design choice that opens up a ton of optimization possibilities (the obvious big one being the ability to issue instructions around a DRAM load). But you need to dedicate silicon to actually make those work. How big is the register renaming pool? What are the edge cases where OOO doesn't work (flags registers are a big one here). Do the SIMD registers get renamed too, or are they serialized? Lots of questions, with AFAICT no public answers.

I think point 2 is the most important -- and I think you're wrong. Intel is already way, way ahead of TSMC, GloFo, and Samsung -- and the gap, if anything, shows signs of extending.

Price will definitely be important, but judging by how hard Intel is coming into this segment, I can only assume that it's willing to match or get very close to ARM component costs.

Intel has just announced a 32nm Atom SoC. Meanwhile, the first products based on TSMC's 28nm process have started to ship.

Yes, Intel is usually the first to bring a new process online (and their 22nm Ivy Bridge is due to ship in a few months), but for the entire history of the Atom product line, Intel has failed to make use of their fab advantage. They probably never will, so long as it is more profitable to put the desktop/laptop/server product lines on a new process first.

Currently there are no notable shipping devices with Intel CPUs. Thus, up to now, Intel's fabrication process advantage has not been enough. I speculate that the rest of the world will stay close enough to Intel to make the advantage non-disruptive to ARM. There is however a technological chasm that ARM must successfully cross - namely the transition to high performance systems enabled by the ever higher transistor budgets permitted by Moore's Law.

One can imagine that big L2 caches, high memory bandwidth and eventually 64-bit addressing will be important for mobile devices. On one hand Intel can attempt to drive existing high performance x86 solutions to low power, and on the other hand ARM can attempt to extend their architecture for high performance - given more transistors and better power characteristics available to both sides, the rest of the world lagging Intel.

It looks like a worse ISA will win because it happens to be owned by the owner of the better lithography process. In a really efficient market, those would not be tied. Microsoft using Windows to push Office was bad for society. Why is Intel good for society?

The old RISC vs CISC is completely dead now. Intel have now shown they can match ARM's equivalent powered CPUs in terms of energy efficiency, and that was the last argument the RISC proponents had in the argument.

Intel's chips are internally RISC, with the only overhead of decoding and fusing the ops and the transistors to do this. And despite this overhead, they're still miles ahead.

The only argument left then is the actual instruction sets at asm level, and here, CISC has an advantage, because they can do so much more in one instruction (take memory addressing modes for example), so they're more compact from an executable point-of-view and instruction cache point-of-view.

Intel is beating ARM because Intel are a process ahead. If an Intel fab made ARM chips on the same process they use for i7s, that ARM chip would beat the pants off the Intel chip made on the same process.

Intel does waste lots of chip real estate on caching big instructions (http://www.anandtech.com/show/3922/intels-sandy-bridge-archi...) and on wide decoders that fault and go to microcode when they encounter some instructions. Think of how inefficient the SSE and AVX encodings are, because the need to escape 8086 code. Or the extra byte needed to access the extra 64bit registers. New CPUs are seldom limited by decoder width on regular code but it does cost watts and it costs space that could have been used for cache or for making the chip cheaper.

I'm not even sure ARM is the best ISA. It doesn't matter which alternative is the best: I just predict that the best won't win because Intel will use their superiority in manufacturing to artificially boost the success of their lackluster CPU design guys.

What are you talking about?

Intel's 65nm (Pentium M) can still beat similarly clocked (1.5 Ghz) ARMs at 28nm - that's not due to process, that's due to a much more powerful architecture designed for speed over power consumption (Pentium M, then Core, Core 2).

The instructions that fault are the very rarely used ones like weird string processing ones. Those are not going to be used in the middle of number-crunching loops.

Escaping float ops hasn't been needed since the 486 days, so I'm not sure what on earth you're talking about.

Intel's branch predicting and memory / cache pre-fetching are the best in the business. To say their CPUs are "lackluster CPU design", is pretty ridiculous, despite Atom being pretty crap.

> Intel's 65nm (Pentium M) can still beat similarly clocked (1.5 Ghz) ARMs at 28nm

You can only compare chips that have the same power consumption. A 1w chip against a 45w (or whatever) chip is not a comparison you can make.

Specifically, Intel now demonstrates a Penwell SoC with a Saltwell (Atom) core made on Intel's 32nm LP process and it benchmarks well against Cortex-A9 chips made on TSMC's 40nm process. I said that I guess that if Intel manufactured an ARM chip on Intel's 32nm LP process with the same die area and power consumption as their x86 chip, the ARM chip would be preferable. Like Anand said, We'll see how this SoC stacks up against multi-core Cortex-A15 chips on TSMC's 28nm process.

If Intel changes the priority and makes their Low Power process transform to the next node together with (instead of after) the high-performance process (currently one step ahead at 22nm), then they will be seriously ahead of whoever manufactures ARM chips (TSMC and others).

> The instructions that fault are the very rarely used ones

They complicate the decoders and take up space and I think take up latency (clocks) during decoding. This costs money even when these instructions are not used.

> Escaping ... so I'm not sure what on earth you're talking about

REX prefix: http://www.x86-64.org/documentation/assembly.html

> To say their CPUs are "lackluster CPU design", is pretty ridiculous

Their power efficient design is in-order and pretty crap. You should read what Linus Torvalds had to say about it. Yes, Conroe was awesome and Sandy Bridge is awesome, but that is not what we're talking about here.

CISC isn't necessarily more compact. Since there are more instructions, each instruction takes up more space. For example in ARM's thumb encoding, each instruction takes up 16 bits. A typical x86 instruction takes much more space.

I wonder how much of an x86 chip is dedicated to handling legacy instructions or largely unnecessary instructions. How much saving could be achieved by a radically simpler instruction set (for example no neg/call/ret/push/pop/bcd instructions). If you look at the http://en.wikipedia.org/wiki/X86_instruction_listings then most of the instructions seem unnecessary. Does anybody have some information about this?

> I wonder how much of an x86 chip is dedicated to handling legacy instructions or largely unnecessary instructions. How much saving could be achieved by a radically simpler instruction set (for example no neg/call/ret/push/pop/bcd instructions).

It's been a long time since I was in a closely related biz, but back then, the answer was "not much if you want to go fast".

Instruction decode isn't a huge deal. Micro-op sequencing isn't a big deal if you're going out of order (and many risc instructions tend to get broken into micro-ops).

The number of architectural registers used to be a big thing, but X86_64 is good enough. (I don't know if anyone got around to renaming memory.)

Okay, I probably should have been more specific - they might take up more memory, but you generally need less of them to do the same thing in RISC, so that usually more than compensates.

Instruction decode isn't really difficult - 4-way decode could arguably be called difficult, but there's not much of that.

They're certainly ahead on process. It remains to be seen if that will be enough.

1. Intel have a deal with TSMC to allow TSMC to fab Atom silicon for custom SoC solutions. 2. Intel is way ahead of other fabricators - die size alone isn't the only metric, there's things like powergating which Intel are way ahead on. They're about 1 year ahead. You also can't compared process sizes between fabricators, because they're not the same. 6. Only relevant for non-CPU / memory bound stuff - ARM's still very poor on float throughput and memory bandwidth (bus size is crap and still only 32-bit access to memory), so it'll only be things like VMs and webservers that use them for power consumption. It'll probably hurt AMD more. They still can't get their 64-bit stuff going, they're trying to do PAE (Physical Address Extension), and we know how well that worked on Intel's P4...

I'm guessing that Intel cut Motorola a really sweet deal to get them to sign an exclusive contract. The decision probably had little to do with Intel being a better platform, and more to do with cost per chip.

Let's face it - Windows 8 running on ARM was a pretty big wake up call to Intel. They knew that they had to have a major win, or risk getting disrupted in the near future. So they called up Sanjay Jha, and figured out what it was going to take to get an exclusive contract.

When you look at it, Medfield is 3 year-old desktop chip technology, which itself wasn't exactly new (the Atom had some improvements, but wasn't a radical departure from what preceded it). Intel isn't bringing anything new to the table - they just keep shrinking the same old chips, and reducing power requirements. At the core, they are still a PC chip company - they just figure out how to shoehorn those chips into smaller and smaller platforms.

The new ARM-based devices (such as Tegra 3) are actually much more innovative and better-suited to mobile devices than Medfield. I think that we will see a lot more innovation in chip design from ARM licensees going forward (such as the low-power 5th core on the Tegra 3), and this will eventually disrupt Intel from the bottom up.

"So they called up Sanjay Jha, and figured out what it was going to take to get an exclusive contract."

Not to take away from the overall point of your comment, but isn't Google in the process of finalizing[1] their purchase of Motorola Mobility, so wouldn't Google have been deeply involved in the decision. In essence a phone to Jha, Rubin, and Page.

I think it's too early to tell weather Medfield is a hit or miss. Anandtech seems to be impressed with the Medfield Platform[2], so far based on what they have seen

"Intel finally did it. After almost five years of talking about getting into mobile phone form factors, Intel went out and built a reference platform that proved what they've been saying was possible all along. Furthermore, Intel also finally landed a couple of partners who are willing to show their support by incorporating Medfield into their product portfolio. The releases are still a few months away at the earliest (possibly even longer for Motorola) but it's much better news than Intel has ever reported before in this space.

The partnerships aren't out of pity either: Medfield is fast. I firmly believe had it been released a year ago it would have dominated the Android smartphone market from the very start. Even today it appears to deliver better CPU performance than anything on the market, despite only having a single core. GPU performance is still not as fast as what's in the A5 but it's competitive with much of the competition today, and I fully expect the dual-core version of Medfield to rectify this problem.

Based on the data Intel shared with us as well, the x86 power problem appears to be a myth - at least when it comes to Medfield. I'm still not fully convinced until we're able to test a Medfield based phone ourselves, but power efficiency at the chip level doesn't seem to be a problem.

Medfield and the Atom Z2460 are a solid starting point. Intel finally has a chip that they can deliver to the market and partners to carry it in. Intel also built a very impressive reference platform that could lead to some very interesting disruptions in the market.

While I'd like to say that Intel's Medfield team can now breathe a sigh of relief, their work is far from over - especially with more competitive ARM based SoCs showing up later this year. I'm really interested to see where this goes in the next 12 months..."

[1] http://www.theverge.com/2011/12/12/2630134/googles-acquisiti... [2]http://www.anandtech.com/show/5365/intels-medfield-atom-z246...

> Medfield is fast. I firmly believe had it been released a year ago it would have dominated the Android smartphone market from the very start.

Unfortunately, ARM is fast enough. The iPad can run an internet browser, and Angry Birds, which is all tablets are really good for. Gamers want better 3D, and maybe a physics chip, where ARM may have the edge. Number crunchers SSH into a real computer. The thin client may have had a few false starts, but it seems to be finally emerging for real.

There's nothing exclusive about the Moto deal; I'm not sure what you're saying there.

IMO Atom is radically different from Intel's previous designs — radically worse, that is. But it sure is cheap.

Apologies. I misread "extensive" as "exclusive" in the article. It doesn't matter - I'm sure that Intel gave Motorola a sweet deal to get an "extensive" commitment.

MMI has been shifting away from proprietary app and comms processors for a number of years and this is a a continuation of Jha's policies. This news is more of a loss for QCOM as they were certainly hoping to have an "in" with Jha to possibly displace TXN parts on Motorola devices.

INTC isn't done with Medfield, of course, you should expect more cores & and aggressive push on process nodes. I would consider it is a validation for their new mobile strategy: http://www.intc.com/releasedetail.cfm?ReleaseID=559064

And as to your point about ARM-based devices being more "innovative," to some extent all the ARM-based devices are basic shared-memory architecture with the same mix of IMG GPU's as Medfield. You could argue, conversely, that Intel is at least trying to innovate on their pipeline and memory architecture (an increase the L2 cache for example) which the ARM-based SOC's are not doing.

Of course I guess this just means you'll see ARM A15 designs coming to market sooner in an attempt to separate from Medfield?

> Windows 8 running on ARM was a pretty big wake up call to Intel.

Windows 8 stil hasn't shipped, won't ship for quite some time and won't run most current Windows PC software. Don't forget non-x86 Windows NT ran on MIPS, PPC, Alpha and Itanic (and i960, Clipper and SPARC, IIRC) processors and it mostly failed because it couldn't run software compiled for x86's.

It also failed because, to within a small margin of error, nobody owned MIPS, PPC, Alpha, or Itanium systems. ARM systems, on the other hand, are selling tens or maybe even hundreds of millions a year right now.

And for the foreseeable future none will be able to boot Windows 8.

Unless Microsoft wants to ship an installer that jailbreaks and installs W8 on iPads.

I wouldn't be surprised at all if they come out with windows phone 8 based on windows 8 and not windows mobile. That way they could have one os for phones/tablets/pc's/ect. They could use their market/repository as a way to force people to modify uis. Since all of the 3rd party apps are written in a version of silverlight(they haven't dropped this restriction yet right?) it should have few incompatibilities.

Sure, but there's still an established concept of "ship an ARM tablet and get people to develop for it". In other words, there's no Wintel monopoly to contend with as there was before. I think that's largely what NT-on-not-Intel fell victim to, despite coming from one of the Wintel companies.

I haven't seen anything to indicate this is an exclusive deal.

This feels rather overblown to me.

Intel is entering a market. Agreed.

This will mean that ARM has to compete with them. Agreed.

This will be good for the consumer. Agreed.

And Intel will completely wipe the floor with the current market monopolist. Nope, sorry, you lost me there.

Sure, they _might_ be able to outflank ARM, but they haven't managed to do that so far, and ARM are the experts in this space. Intel have tried to outflank experts before, and their graphics processors don't seem to be the market leaders.

This is the nature of the beast in today's tech journalism. Every story is cast as some sort of Manichean struggle. It's more than a little bit silly. We can follow the competition between tech companies without imagining that one or the other is sure to go out of business soon.

Spot on! ARM should worry, but really Intel doesn't have a great track record here. And ARM does. Also, ARM has been scaling up pretty well, and the graphics cores on those chips are arguably better than what Intel has ever been able to turn out.

Summary: We'll see, and it will be fun to watch.

anandtech provided a more level-headed analysis. Intel's new design performs well and beats ARM designs - that have already been in market for a year already. This years ARM designs based on A15 are quite likely to be as good or better. A1,5 the next generation arm, is wider (more parallel decode and execution units than Atom) and is out-of-order further increasing instructions-per-cycle (at the expense of some complexity and die size). This is like the architectural difference between 486 and pentium.

At least on paper, the A15 should be notably faster than Atom if at similar clock speeds (MHz).

I agree with you.

This article sounds like Intel PR to me.

Intel has the largest share in the PC graphics market, over 50% of the total.

They have the largest share in not providing decent graphics.

ARM produces a design people incorporate into their chips. I'm not entirely certain you cannot take an ARM core and enhance it if you pay the right fees (when I worked on them, I don't think we did, but we just need a core to do management for our hardware-based video decoder).

Not sure if IBM fabricates any chips for people right now, but we used UMC and TSMC, so it's really not about ARM for the manufacturing process, it's about those two companies. I don't know what UMC and TSMC are advertising as their current process, it might be a little behind Intel, but I'm sure they'll continue to shrink processes.

The question is why would an x86 processor need to be put on your phone? I guess because the history of Intel so far is all x86, their other processor lines have all been phased out (I hesitate to say failed, but that might be the right word). Intel really doesn't want to compete on low-power, low-margin chips, they like making money on the desktop and server.

From my limited experience, ARM is vastly easier to work on than x86, as far as assembly programming is concerned. I personally hope this chip fails, because I'd prefer not to have x86 polluting the world anymore than it already has.

> I'm not entirely certain you cannot take an ARM core and enhance it if you pay the right fees

You can. There are at least two licenses, a core license and an architecture license. Qualcomm has an architecture license; the Scorpion (Snapdragon's ARM core) is an ARMv7 core somewhat similar to an A8, but is not actually an A8. Companies with core licenses have also modified them; the Samsung/Apple/Intrinsity Hummingbird/A4 used an A8 core modified by Intrinsity to work efficiently at 1GHz; TI later followed suit. Intrinsity's main business model, in fact, was optimization of other peoples' core designs.

Ok, cool, I should have looked it up, but the sentence in the article was "Except for Qualcomm, which makes its own ARM-compatible cores, every other ARM chip on the market uses reference CPU cores", which I didn't think was true (and I don't think 'compatible' is the right word for Qualcomm, probably 'extended' ?).

Also, he talks about this future 22nm process from intel for 2013, as though the other fabs are going to sit at 28nm forever (even though Intel is also currently at 28nm).

It just seems like a poorly written article, or someone who copied a little too much from Intel's PR.

You have to pay ARM extra to be able to tweak their cores or design your own. Qualcomm does this with their Snapdragon processors, and the Samsung/Intrinsity collaboration that spawned Apple's line of SoCs and a few for Samsung is also based on modifying the ARM cores. Most everybody else goes for the cheaper ARM licenses.

In short, if Motorola didn’t think that Medfield was competitive, it wouldn’t have agreed to make a bunch of Medfield-powered phones and tablets.

Either that, or Intel paid it a bunch of money to make it worth their while.

Exactly, or they're just making a mistake - like Logitech did with their Google TV boxes, which cost them $100m in operating profits.

Yep; I'd be particularly suspicious of this given that the two companies involved are Lenovo, which doesn't have a significant smartphone market anyway, and has a huge PC market, and Motorola, which is in reasonably dire financial straits. If HTC, or, for that matter, Apple, were a launch partner, that might be more promising.

> Consumers, for the first time ever, will actually have a choice — just like AMD and Intel on the desktop, you will be able to pick a smartphone or tablet with a CPU that best suits your needs.

Consumers don't particularly care what CPU is in their smartphone or tablet.

I'm sort of biased towards ARM, but I'm still very impressed by how well Intel is doing. The history of computers is littered with the corpses of the companies that brought down by competitors with cheaper, not as good solutions. Look at 3.5" hard drives, x86 vs the RISC machines, etc. And mostly they failed to do anything because competing when they still could would have eaten into their profit margins. But here Intel is devoting actual resources to a smaller, cheaper, less powerful offering.

Intel's SoC Atom part will not, in my opinion, be priced below ARM parts competing for a particular manufacturer's design. Intel is a high margin company and I think will sell high-priced Medfield on the basis of higher performance. Intel is the high-priced market incumbent in this case - and may well be disrupted soon by certain ARM low-power servers.

The spec's I don't think tell the story here. It's going to come down to price and marketing.

Intel still has some very strong brand recognition with their 'intel inside' campaign, and though most people don't actively look for intel chips when buying a computer, they do recognize when the computer they are buying has it, and it makes them feel comforted (much like McAfee or Norton etc. etc)

The hardware manufacturers will build on a competitive price point. I suspect Moto's dedication to Intel was driven more by price than by performance.

Not to say that I don't believe intel will perform, and I also disgree with the article, and I think javascript performance is hugely important to the end user.

14nm in two years? Whoa. That's going to be huge for the server market.

Imagine 24 cores per chip with the same power draw as they have today for six cores.

Actually, I think the Moore's law benefits of continued chip feature shrinkage will be felt most strongly in personal devices. For example, to perform voice recognition, Siri must send the raw waveform to a cloud server for processing, because the local CPU and RAM are insufficient for acoustic and language modeling. With a couple more CPU generations, i.e. say four years, I expect mobile devices to perform these sort of calculations locally. Beyond speech recognition is machine vision - object/scene detection and recognition. Currently, wireless bandwidth limitations preclude a realtime cloud-based facility for machine vision, but more capable mobile device CPU/GPUs could enable local processing.

In general, Moore's law progress enables less expensive cloud servers to perform the same tasks, but that same progress enables mobile devices to perform new categories of tasks!

"Currently, wireless bandwidth limitations preclude a realtime cloud-based facility for machine vision"

I do not think that is all what stops it. Just as speech recognition, machine vision is quite different from computer graphics. We _know_ how to make graphics orders of magnitude better; just look at Pixar's latest movie. 'All' we have to do to bring that to your phone is do it more efficiently.

For speech recognition and machine vision, I am less optimistic. Reason? I never hear about, say, "almost perfect, but it takes a 1000 CPUs a day to process a second of sound" for speech recognition or object recognition.

I wonder why that is? Is it because there are problems we haven't solved yet, or because nobody approaches the problem from that side?

For clarity... I am an AI developer. Recently I added a feature to my dialog system so that if I have not typed anything into the dialog web app in a while, the system begins capturing frames from the webcam and performing face detection using JavaCV. If a face is detected, then face recognition is performed on the clipped greyscale face. If the face best matches mine from the training examples, then the system uses Google speech-to-text synthesis to say hello to me, "hello Stephen".

A smartphone cannot practically do this because JavaCV will not run fast enough on an ARM CPU and wireless bandwidth limitations preclude sending whole realtime captured phonecam frames to a server. But, when Moore's Law permits, JavaCV will run on the mobile CPU/GPU combination - and my phone could say hello to me when it recognizes me.

I know that Google is working on face-recognition phone-unlocks - but I wonder if their implementation involves a cloud server and one-time sending of a captured frame?

Hmm. Does Android have fat binaries for the NDK? I can see Motorola et al getting a lot of complaints about games not working...

Android doesn't use 'fat binaries' by the strict definition of the term, but the APK format that apps are shipped in can include multiple individual builds of both the main app and required libraries so you could pack x86 and ARM builds in a single APK.

FWIW, there are already Android devices out there on x86 -- all of the currently shipping Google TV devices are built around Atom. The NDK isn't officially supported on those GTV platforms but it isn't difficult to get it working in practice.

Also, any complaints for this would probably be directed at Google (though I guess Google and Moto will be the same company soon anyway) because arch filtering is generally handled by the Android Market, so for users with these devices they just won't see the games showing up at all on their local install of the Android Market (since the market will be filtering out anything NDK based that doesn't have an included x86 build), so it looks like a market issue (if they already know the game exists), not an app issue.

In my experience, the average user is only dimly aware, if at all, of Google's involvement.

Exaggerated title and claims attract impressions. News at 11.

lolololololol

To me, the articles main support for the rise of Intel as a mobile chip competitor are: 1). Motorola made a deal 2). Intel can leverage their own manufacturing process/plants

The first is stupid. Considering how much Motorola bent over for Microsoft, this is probably more Microsoft doing Intel a solid to get into the market rather than a purely "logical" financial long-term decision by Motorola.

The second I can't really argue about with any real knowledge. However, upgrading plants is fucking expensive. By ARM separating themselves from that portion of the chip business, they have basically enabled a secondary competitive market of improving ARM designs. Intel has locked themselves into the position of Overseer of all-best-designs, and whether or not this is true, it will lead to comparative stagnation in innovation.

I'm might be wrong, but I honestly think that Intel will lose simply because their mentality depends too much on an idea of limited competition. They are too used to AMD and legacy slaves; the switch to open-source and web-based systems will lead to insane competition and crunch their profit margins like a nutcracker.

Reply on news.ycombinator.com