Slashdot Mirror
ARM, Intel Battle Heats Up
An anonymous reader writes "Low-power processor maker ARM Holdings is stepping up rhetoric against chip rival Intel, saying it expects to take more of Intel's market share than Intel can take from them. With Intel being the No. 1 supplier of notebook PC processors, and ARM technology almost ubiquitously powering smartphones, the two companies are facing off as they both push into the other's market space. 'It's going to be quite hard for Intel to be much more than just one of several players,' ARMs CEO said of Intel."
I've been hearing about how ARM is going to destroy Intel for the last 5 years at least and I haven't seen the products yet despite the promises thrown about with the Cortex A9. It looks like the cortex A15 willl be able to beat Medfield... but you aren't getting those A15s in large quantities until next year when Intel will have the next iteration of Atom ready anyway. Oh and 64 bit? That's gone from an insanely important feature when Intel didn't have it to being useless bloat when Intel does have it and ARM doesn't, but it's OK because in 2015 you might be able to get an ARM chip with 64 bit support....
Since 2008 when the much derided Atom debuted, Intel has gone from not having anything that could remotely run a smartphone or tablet to having Medfield, which is competitive although not industry leading in the smartphone and table space. I have yet to see ARM come out with anything that even threatens a run of the mill Core 2 yet... so why is ARM talking so much trash?
It might be that ARM is a little more nervous that there is finally some real competition in the mobile space, which is a boon to consumers. I'd like to see AMD get an x86 solution down into this power envelope too so that there would be multiple competitors on the x86 side as well.
AntiFA: An abbreviation for Anti First Amendment.
Low-power processor maker ARM Holdings
ARM Holdings do not "make" processors, low powered or otherwise. They design, they develop, and they certainly license. But they don't make.
Interestingly from a Slashdot point of view they're probably the most high profile example of an "IP" company with a positive image.
Intel gets my respect for being one of the few companies to invest heavily into research. Seriously, they do a lot of "fundamental research" work, and so far, it's worked well for them. They develop products all the time that never get released because they're too "experimental" - Larrabee is the example that comes to mind first - and justify the expense because the information learned is worth the $100M they spent on an unreleasable product.
Intel, you can hedge your bets. Take one of your teams - rumor has it the Itanium team won't be working on that much longer - and tell them to make a desktop-quality ARM processor. You've already got the ARM license, do something with it. Figure out how to bump up the clockspeed (if *Apple* can do it, so can you), throw cache at it, bring the core count up to eight or so. Target your own Core i3 chips both in speed and in cost.
You do that, Intel, and you basically can't lose (barring sudden inexplicable incompetence). If the ARM desktop project completely fails, well, you just proved that x86 chips are unbeatable on the desktop market (which will never completely disappear). If the project succeeds, you'll win no matter which architecture comes out on top, and you'll have the advantage of having an experienced ARM team to help you take the best features of ARM and put them in your mobile x86 chips.
22nm lithography. Intel is, as usual, a node ahead of basically everyone else. Other fabs just got their 28nm half node online not long ago, late last year. So we are seeing products based on that start appearing on the market. The current nVidia and AMD GPUs would be some notable ones, but there are (or at least will be) ARM chips too.
Intel though, they didn't do the 28nm half node (they haven't done half nodes so far), they went straight to the 22nm node and it is online and running full swing. Ivy Bridge chips using it have shipped in large quantities.
What that means is Intel can pack more transistors in to a given die size, and have them use less power per transistor. For mobile, that is a big advantage. That means even in the event their shit does less per transistor, they can make it up with more transistors. Also means things like 64-bit are less problematic to implement (64-bit requires more transistors).
Now I've no idea if Intel what arenas Intel will choose to compete in, but if I were ARM I wouldn't be looking forward to direct competition. I'd hope it remains largely how it is: Intel focusing on the high end (from netbooks all the way up) ARM focusing on the low end (from tablets all the way down). No competition, no problem. I wouldn't be enthused about the prospect of having to compete with someone in the low power market who has a better process.
Intel is likely to keep the advantage too. Everyone else is hard at work setting up their 22nm fabs, but they are probably at least a year away, maybe more. Intel? They've been hard at work building Fab 42 inc Chandler which is to be their first 14nm facility. They say they'll have it online in 2013 (it'll be some time after it goes online until chips are shipping to consumers though), and they are pretty good about hitting their marks on that.
It is one of the things that has given them an edge is their massive R&D in to fabs that keeps them a node ahead of everyone. ARM can't do that, they are just a design company, not a fab, and none of the other companies that do fab work seem to be willing to plow in the R&D money that Intel is.
I have Cortex-A15s in the lab right now, finished simulation and verification, and samples are being boxed up this week, and we're shipping them this year. And for quantity we can do as many as TSMC can make for us. Then it boils down to if any devices makers are willing to pay the premium we have to charge for the bigger faster chip. You will see about half of Jellybean devices running a Cortex-A15, and nearly all Windows 8 devices when it first ships.
Your assertion is dissociated from reality. It completely ignores the netbook phenomenon, not to mention the inception of smartphones and tablet computers.
People don't buy these devices because they "want more CPU". After a certain level, the "CPU" amount is irrelevant and its practical effects are completely unnoticeable. There is a good reason why hardware companies rely on artificial benchmarks designed to push the hardware in completely unrealistic, useless and impractical scenarios to be able to compare their hardware against the competitor's offering, and therefore justify a higher asking price.
To drive the point home, I can tell you my personal case. My last two hardware purchases were a netbook and a smartphone, which, by today's standards, are considerably lacking o the "CPU" department. Yet, they are by far the two pieces of hardware which I use the most. I also have a desktop and a laptop which I've purchased a few years ago, and I actually use them for serious stuff which actually require real CPUs to crunch real numbers. I'm talking about structural analysis and CAD work. In spite of actually having to use a computer to actually do some serious number crunching to actually get a meaningful result, unlike calculating pi to the nth digit after the decimal point, the fact is that both my archaic desktop and laptop are more than capable of handling heavy workloads required for practical engineering work.
And this without even relying on OpenCL to take advantage of the hardware which is already present in the system and basically never leaves the idle state.
So, in short, contraty to what you said, people actually "don't want more cpu". People actaully know that they can't notice it after a certain point, which was actually passed about half a dozen years ago, and people are also aware that the inflated price tag associated with having "more cpu" actually doesn't justify the diminishing returns they get with that purchase. What they want is cheaper stuff that is actually good enough to get the job done, and if the job in mind is checking email, facebook and any other mundane tasks then people do know that the price tag of a supercomputer is completely unjustified, when they can easily get away with it by purchasing a glorified cellphone, with or without an embedded keyboard.
Slashdot, fix your code or at least hire someone who is competent at it to do it for you.
See the source.
Today one of the most significant features of the ARM family is its low power consumption. But that hadn't been an initial goal, according to Furber. “We designed the ARM for an Acorn desktop product, where power isn't of primary importance. But it had to be cheap. Cheap meant it had to go in a plastic package, plastic packages have a fairly high thermal resistance, so we had to bring it in under 1W.”
The power test tools they were using were unreliable and approximate, but good enough to ensure this rule of thumb power requirement. When the first test chips came back from the lab on the 26 April 1985, Furber plugged one into a development board, and was happy to see it working perfectly first time.
Deeply puzzling, though, was the reading on the multimeter connected in series with the power supply. The needle was at zero: the processor seemed to be consuming no power whatsoever.
As Wilson tells it: “The development board [we] plugged the chip into had a fault: there was no current being sent down the power supply lines at all. The processor was actually running on leakage from the logic circuits. So the low-power big thing that the ARM is most valued for today, the reason that it's on all your mobile phones, was a complete accident."