Over 50 people have run it on their PCs so far, and posted the results. Bless the reader, known only as ‘N’ in his post, who ran it on his Atom netbook and achieved a blistering 1,935 seconds’ render time. That’s a mere 26 times slower than the Core i7 980X.

But now Intel has weighed in to the fight, and not surprisingly, it’s winning, thanks to a monster PC overclocked to nearly 5GHz.

That mass of components and pipes and things in the photo is a PC system buried deep in the heart of Intel’s very own UK labs, and it’s the property of one Steve Anderson, an Intel virtualisation engineer. Intel’s Alistair Kemp fills us in:

“Well here at Intel we could not resist the challenge, particularly when in our IT department we have overclocking nut Steve ‘DaFridgie’ Anderson. We gave him the challenge of beating 60 secs but restricted him to only a single-socket setup. Last night he ran it on his rig. Result: 50 seconds to complete.”

Hmmm. 50 seconds. That’s quite fast.

The rig itself uses phase-change cooling: in other words it’s attached to a chuffing great freezer, which I believe is the big box on the right of the photo. That yellow meter with the readout is showing the temperature of its output: yes, that’s minus 40 degrees celcius.

I asked Steve to send over some more details of his rig (which Alistair says is his “pride and joy”). He responded with this:

“The system is built as follows:
Chassis – Dimastech test rig
Processor – Intel Core I7 980x retail CPU
Memory – Corsair Dominator GT memory ( 2250 CAS 8 ) running at 2046 at CAS 7
Motherboard – EVGA Classified X4 E762 motherbaord
Power supply – 2 x Corsair 950w Single Rail Power supply (1 for system, the other for graphics card) automatically switched
Graphics card – 1 x EVGA GTX480
CPU cooling – Asetek Lightspeed refrigerated cooling system with a running temperature range of -40 to -32oC at 18oC ambient temperature
Storage – Seagate 160GB Sata drive (16mb cache)

Construction – Custom built for performance, not for looks! “

Well, quite. He has some interesting thoughts on the finer points of the smallpt test and the best way to get maximum performance from it too:

“The smallpt benchmark is a nice one to run, simple and very effective at loading up processors and checking for stability at speed. For the run illustrated by the screenshot [see the screenshot above] I set pretty much the same settings as I use for processor-intensive 3D benchmarks. This benchmark is sensitive to both memory performance as well as QPI speed and these were both adjusted up to the limits of stability at this processor temperature. Intel i7-980x processors do like the cold but also need some extra voltage to drive to extreme frequencies, hence the 1.63v CPU core voltage.”

But it doesn’t, of course, end at a mere 5GHz. No, no, no. They want to cool it down just a tad more. Well over a hundred degrees more, actually:

“I would expect that as we cool the processor further down to extreme cold temperatures (-180C) with Liquid nitrogen cooling, we would see stable runs on this benchmark in the 5.7-6.0ghz range.
We’ll be doing some extreme overclocking around the end of august and I’ll add this benchmark to the list to see if we can provide an even more impressive result.”

So, if anyone else out there is thinking of beating 50 seconds, you’d better do it before the end of August. Unless you happen to have a handy liquid-nitrogen cooler knocking around in the back of your garage, of course.

Anyone?

Having been writing about photo-realistic 3D graphics rendering for issue 192 of the magazine, I’ve been getting myself back up to speed with the state of 3D graphics and looking into the absolute best techniques for achieving realistic lighting. And along the way I’ve got a new insight into the sheer speed of the latest CPUs.

Turns out the best 3D rendering algorithm is a hugely intensive method known as path tracing, which is sort of like ray tracing’s dad. The theory behind the method actually pre-dates ray tracing, but it’s only now that PCs are getting fast enough for experimental dabbling at home.

The good part is that, while it needs a heck of a lot of computing power to do, path tracing is actually a fairly simple technique to implement.

But where to get a path-tracing application to play with?

Well, Kevin Beason has written a beautiful example of minimalist programming with his path-tracing renderer, smallpt. It’s a complete functioning renderer, with a 3D scene (based on the research-standard Cornell box scene) embedded into the program.

Smallpt generates and saves to disk the fully rendered, near-photorealistic image you can see above. And it’s written in a ridiculously compact 99 lines of C++ code. That’s the entire renderer, including the scene itself.

Kevin provides only the source code on his site, but I fancied running smallpt. So I spent a couple of hours getting it to compile under Visual C++ Express 2010, which is completely free and you can download from here.

The code assumes you’re using the open-source GCC compiler and his code includes some Linux/gcc programming tricks that don’t work under Windows, but a bit of tweaking later I had it rendering the Cornell-box scene.

It is awfully compute-intensive though, taking over 12 minutes to render a grainy 100-samples-per-pixel version on my Core 2 6300 everyday office PC:

Aha! This was a perfect opportunity to put my new quad-core Core 2 Q9400 system, that our lovely IT department built me a couple of weeks ago, through its paces. I added a few lines to the code of smallpt to get it to give me an overall time in seconds for the complete render, and set it going.

Straight away, render time came down to 252 seconds – just over four minutes.

Then I remembered my dual-Xeon workstation muscle machine, originally a test ‘white box’ from Intel that, ahem, never found its way back to them. The only reason I don’t use it as an everyday machine is its excessively loud industrial-level cooling system. But with its dual, quad-core Xeon processors, which cost some frightening amount of money when new, this was the perfect job for the Beast.

I set up the machine in a corner of the PC Pro Labs (well away from complaints about the noise) and installed Windows 7 Ultimate x64, just to make the test fair since that’s what’s running on my other PCs.

Then I fired up smallpt.exe and postponed making my next cup of tea, knowing it would rip through the render before I could even rise from my chair.

Oh.

Turns out my once-mighty eight-core workstation, barely over three years of age, is now slower for raw compute speed, and by a heck of a margin, than my quad-core machine.

In fact its two Xeon X5340 CPUs took 493 seconds to churn through the smallpt render: getting on for twice as long as my quad-core.

Deflated, I switched off the machine, then wandered over to Mike Jennings in his own corner of the Labs, engrossed in a graphics-card group test for the next issue of PC Pro.

“What’s the CPU in your test rig, Mike?”

“Oh, it’s a Core i7 980X. Six cores. Really fast!”

“Ah. Fast you say? Um, mind if I use it when you’re done?”

“Sure.”

So I did.

It’s not often I class a computer as astonishingly fast, but hell’s teeth this one certainly is.

The render completed in 73 seconds. That’s almost three-and-a-half times faster than my nearly-new Q9400 machine, and nearly seven times faster than my not-exactly-old, dual-Xeon workstation that was worth a good four thousand pounds when it was new.

Let’s consider those results on a per-socket basis.

With this pure-CPU, highly multithreaded task, the latest generation of enthusiast-level Intel CPUs are over thirteen times faster per processor than the professional-level Xeon CPU of three-and-a-bit years ago. And about five times faster per core.

I knew all this before, but seeing that machine chew through the render with such ferocious speed really brings home the level of engineering achievement that Intel continues to manage, year after year.

Try it yourself

If you want to try the unofficial PC Pro smallpt render test on your machine, you can download my compiled version here.

But wait! The multithreading needs the Microsoft OpenMP support DLL, vcomp90.dll, and the program won’t work without it.

The free-but-faffy way to get it is to install the Microsoft Visual C++ 2008 Redistributable Package from here.

Once the redistributable is installed, search for vcomp90.dll – it should be hiding in a subfolder somewhere within C:\Windows\winsxs – and just copy it to the same folder as smallpt.exe.

Now double-click the smallpt.exe file and the renderer will open in a command-prompt box, churn away for a while and save the rendered image file to the same folder when the render is complete. It’ll also give you the time taken to render when it’s finished.

You can open the resulting .ppm image using GIMP for Windows.

Let us know your results, for machines both old and new.

Has anybody out there got a machine that will break the minute mark?

Update:

Check out the posts below and you’ll see that Intel itself has risen to the challenge. Read all about the superchilled Intel test rig.

First, there’s the screen size. With a 9in e-ink display and relatively narrow bezel, it feels a lot more like a portable screen than a portable computer, unlike the Amazon Kindle. Once you switch it on you can immediately see it’s not just big – it’s high resolution too, at 1,024 x 768. That’s the highest resolution we’ve seen, with the rest of the competition at 800 x 600.

It wasn’t just me who was impressed. Our eBook expert Stuart Turton, who’s with me in Germany, couldn’t suppress a grin when he first tried to navigate his way around the interface.

Initially unable to work out how to move through the menus with the keys, Stuart instinctively prodded at the screen instead. And, to his amazement, it worked. This is a touchscreen that doesn’t look like a touchscreen – it doesn’t have the drastically reduced contrast and isn’t nearly as prone to the off-putting glare of the Sony Reader Touch. Asus seems to have made the touch-sensitive layer almost invisible, so you don’t lose the all-important paper-like appearance of the e-ink screen.

Dig around in the interface and it’s clear that Asus isn’t going down the minimalist feature route. It looks like and behaves like a proper eBook reader, but there’s a stack of extra applications in there, including a web browser.

Tap that browser icon and, after a pretty lengthy wait, up it pops. Text is entered via a handwriting-recognition system, where you draw individual letters into a box, or you can pop up an onscreen keyboard. That’s as far as we got with the browser – it’s listed as a beta app in the menu system and we couldn’t make it actually load a page.

Elsewhere there’s a music player app, and in the eBook application itself you can select text with a swipe of the finger, copy it and then paste it into the text editor app. Currently it’s too slow to be very useful as a study tool, but the Asus rep we spoke to was at pains to point out that the firmware is still being developed.

And we hope that it does get some more polishing before release, since the speed of operation is currently the E Reader’s only downfall. Page turns take around three seconds, and tapping on an icon results in a long wait before anything happens.

The E Reader will be an open platform, able to read all the common formats including ePub and PDF, which is encouraging.

Despite the extra apps, battery life is still officially only being quoted in terms of page turns – currently 10,000. We’re told that Asus is “still fiddling” with batteries in order to squeeze the most out of it while using the other apps.

This is a very impressive eBook reader debut for Asus. The build quality is more than adequate, the screen is great and the extra features look usable rather than bolted on for no reason.

It’s not going to be the cheapest of eReaders, but it’s not outrageous either at a projected £250 inc VAT for the version with 11g wireless.

Asus says it’s looking into the possibilities for a 3G version with integrated SIM too.

Launch in the UK is set for the end of May. We can’t wait.

Despite not being fully finished, two demonstration units are running on Asus’ stand. We spent a while playing, in the company of a slightly nervous-looking Asus rep.

The guts of the machine come as no surprise, and include 11n wireless and the now-standard netbook-style spec list of an Atom N270 processor, 1GB RAM and a 16 or 32GB SSD. At the back there are three USB2 ports, HDMI and VGA video outputs plus Ethernet and audio.

The most interesting part is the touchscreen built into the right-hand side of the unit. It has all sorts of clever little apps.

At first glance the screen display looks like a similar interface to standalone-app gadgets like the Chumby. The difference here is that the screen isn’t a separate bolt-on device running a linux derivative – it’s actually a secondary Windows monitor. That means the interface is pretty responsive and a lot more sophisticated than you’d expect. You can even bring up the Windows XP desktop on it if you want, albeit at a very odd portrait aspect ratio.

When you’re not using it to navigate menus, the touchscreen becomes a nice big touchpad to control the mouse on the main display.

There’s a proprietary wireless video transmitter to go with the Eee Keyboard, and although it’s technically an option Asus expects most units to be sold with one. There’s enough bandwidth to show HD video at 720P, and it was working flawlessly on the display stand with the receiver unit a few feet away. Unfortunately Asus was a bit skittish about us wandering off with one of the only two working samples in the universe, so we weren’t allowed to move the Eee Keyboard away from the receiver to see what the range was like.

So, assuming you want a PC in a keyboard – personally I don’t but I can see there might be some who do – the issue with the Eee Keyboard is the price. Asus tells us that it will launch at 550 euros including the video transceiver, which given the exchange rate will probably translate into something around £550.

It’s an awful lot to pay for a low-powered computer running Windows XP, and we’re yet to be convinced that the software running the secondary screen is solid enough not to act up at inconvenient moments.

Our slumped shoulders when it seemed there was no 3G are now square again – you will at least be able choose a 3G version. So essentially you can choose between a big iPod (no 3G) or a big iPhone (with 3G). And despite there being no mention of GPS capability during the launch event, we’re very much relieved to say the 3G versions will have it.

But the real question is nothing to do with what Apple’s done: it’s the basic issue of the tablet form factor itself. Exactly how are you going to use that lovely-looking on-screen keyboard? Some elaborate preliminary tests – involving a make-believe iPad in the shape of an A4 paper pad – reveal the mechanics of typing on the screen of a flat tablet leads to some awkward, uncomfortable and stupid-looking positions. You basically have to raise both knees to a semi-foetal position, or do what, judging by the demonstrations, Apple people themselves are already adopting as the default iPad-using position: cross one leg.

Admittedly the on-screen keyboard does look very usable, but only if it’s on a flat surface – we simply don’t think it’s going to work very well otherwise. Apple has conceded as much by launching a separate keyboard and stand at the same time as the iPad itself.

Personally I’m very much looking forward to using the iPhone OS on a larger device – it’s an interface bordering on genius and the new additions to the UI with the pop-up context menus make it look like a proper desktop operating system. I just wish it wasn’t restricted to running on a tablet.

Not everyone is as lukewarm as I am. Barry Collins is positive: “I think Apple has created the first desirable tablet PC, and unlike Microsoft the company’s understood it needs an entirely new interface, not just a desktop OS patched onto a tablet”

David Bayon’s view: “Primarily as an eBook reader, I would genuinely quite want one. I’m just not sure I’d pay the premium for all the other stuff in it that I don’t really want though. It could be the eBook reader that could prove to me that eBooks could work, but it’s too expensive for me to want one.”

PC Pro editor Tim Danton isn’t afraid to share his views: “As ever Apple’s made an impressive piece of hardware, and I’m pleasantly surprised by the low price, but a big question remains: who’s actually going to buy it? Jobs claims it’s amazing as a web browser, but it needs something else – a killer App, if you’ll excuse the shocking pun. This could turn out to be newspaper content like the New York Times, it could be the iBooks app, and I hope that all the promises of partnerships with publishers turns out to be true. Imagine the readability of a newspaper with video clips from yesterday’s football match. But that sort of content won’t appear without a lot of investment, and how long will it take before Apple sells enough to justify the expense to publishers?”

The burning question before its release was whether the iPad could turn the awkward, not-one-thing, not-the-other nature of every other tablet device ever released, and somehow make it brilliant by coming up with some unforeseen technical breakthrough. It doesn’t really look like that’s happened.

So the iPad so far looks unlikely to be the breakthrough device the iPhone was. Overall it certainly has potential, particularly as the new delivery medium for magazines and newspapers. But we’re simply not convinced the tablet form factor is the solution to a problem nobody knew they had. The superb iPhone interface in a bigger device means it will wipe the floor with every other tablet device of course, but then that was never going to be very difficult.

The general consensus at the time was that it was probably, maybe, possibly a good idea in the long run, but a gimmick as it stood. We reckoned it was your typical early-adopter-only product.

Well, I’m eating our collective words now that I’ve spent the weekend playing with one.

Now I’m not one to be lured by gadgets for gadgets’ sake – in fact I hate them. But I like this one. I like it a lot. And I like it for four reasons:

1. It works.

It may be a first-generation design, but it does exactly what it says. And the projector stuffed into its innards doesn’t massively affect normal picture-taking abilities aside from making it a bit chunkier than your average modern compact. In fact, most of the time you can forget that it has a projector in it at all.

2. The first time you see your photos magically appear on the nearest wall, you’ll giggle.

And the second time, and the third. It’s just great, and absolutely everyone I’ve showed it to makes the same child-like, gurgly happy noises and immediately wants to play.

3. It’s an idea with a future.

As left-field as the whole concept initially sounds, as soon as the giggle-effect has subsided you start thinking about other things to do with it. The first one that sprang to my mind was using the camera as a movie projector. And lo! As with any compact these days you can shoot movies on the S1000pj, but you can then immediately project them, which induces even more silly giggly fun. Disappointingly, it doesn’t have a video input so you can’t feed, say, a standalone camcorder or netbook into it. But it probably wouldn’t take much to add that input to the next model.

4. The integration and pricing make it a disruptive technology.

How many people do you know who might go out and pay £400 for a dedicated pocket projector? I’d hazard a guess that would be none. How many people do you know who might go out and pay £400 for a new camera? Probably at least a few. With the S1000pj you’re not paying several hundred quid for a niche gizmo with no obvious purpose, you’re paying it for the latest Nikon camera. You carry it around with you because it’s a camera. Everyone wants a camera: it’s not an expensive solution looking for a problem, unlike a dedicated pocket video projector. The projector adds roughly £150 to the price – there’s a fair minority who’ll pay that for a cool feature in their new toy, even if they think they probably won’t use it all that often. Hence, Nikon is likely to shift more pocket projectors into the market than the pocket-projector manufacturers themselves.

The downsides

So the light output is only 10 lumens – compared to several thousand for a standard desktop projector, or 50 for the likes of the recent generation of pico-projectors like the Dell M109S. But that 10 lumens is brighter than it sounds*. In a worst-case test scenario, projecting onto a whiteboard right next to the windows in the PC Pro office, we managed to project an image with about a 10in diagonal; move the camera any further back than that and it gets too dim to be useful. Put it in a darkened room, though, and you’ll manage an image size of a couple of feet no problem at all. It looks great.

Battery life is a bit of an issue, but again less than you might initially think. It’ll last a maximum of an hour in projector mode, and that’s assuming the battery’s fully charged, which a camera kept in your pocket usually isn’t. But you’ll rarely want to use the projector for more than twenty minutes at a stretch unless you’ve a particular yearning to revisit the 1970s and bore your audience into an early grave with your holiday slideshow.

So let’s say in five years’ time battery life has increased two-fold and the efficiency of LEDs has done the same, which I’d say is a reasonable expectation. That gives you a four-fold increase in brightness, or a four-fold increase in battery life for the same brightness. And four hours of projection time is perfect for films.

Futurology is a mug’s game, but I’d go so far as to say there’s an even chance of integrated projectors becoming a standard tick-list feature on compact cameras. And once they make their way into lots of cameras, the prices come down and people start wanting them, they’ll pop up in laptops and netbooks too. And all thanks to the off-the-wall punt that Nikon is taking with the S1000pj.

Plenty of things might derail that prediction of course, not least the fact that your average branch of Jessops doesn’t have too many darkened rooms, making the projector a tricky thing to demonstrate. But the S1000pj is the first true hardware innovation I’ve seen in a long time that really looks like it has legs.

UPDATE:

So here’s an image I put together that best represents how things look in a darkened room. In fact in this case a more or less totally dark living room at about 10 o’clock last night. This picture is two exposures blended together so that you can see both camera and projected image. It’s not a mock-up, it’s the actual image that the camera was projecting in that actual position, I just took a long exposure and  shorter exposure without moving anything and blended them to get the best feel of how it actually looks. For scale, those orange and green things at the bottom of the shot are the cushions at either end of my sofa:

Not too bad, eh? That’s an image of a good four feet diagonal. Again, remember this was more or less total darkness so you’re seeing the best it’ll do, but even so it’s impressive if you ask me.

*Human response to brightness is roughly logarithmic rather than linear, in other words the brighter a light is the less sensitive we are to it. So a 100-lumen projector only appears about twice as bright as a 10-lumen one.

Not quite as daft an idea as I’d initially thought, though. A highly scientific survey of Facebook that I just conducted has revealed that 32.3% of all photos contain the picture-taker themself, in that highly endearing cuddling-up-to-your-other-half/friend/total stranger, sticking your arm out and taking the picture backwards sort of way. The screen on the front of the ST500 and ST550 models lets you do it without taking half a dozen shots of the tops of your heads first. Obviously, in the picture you see on the left, all the front screen is showing is a picture of a camera. This is because I was taking a picture of it at the time. Imagine, instead, a happy scene with laughter and smiling.

But wait, there’s more! Not only will the front-mounted screen mirror what the lens can see, it’ll do other things too. It’ll show you a numerical countdown for self-timer shots, or a happy smiling face to indicate to your friends exactly when the camers is about to take the shot, so they can pretend to be happy too, and smile.

Best of all, it will (if Samsung is to be believed) make your children photogenic no matter what. It will manage this trick by showing a terrifically cute animation of a clown on the front screen, thus ensuring the little blighter darling will be enraptured, smile delightedly and radiate angelic photo-friendly happiness at just the right moment.

There’s even going to be a facility to download new animations to the camera, which will – as our Samsung spokesman so eloquently intoned – “reinvigorate the product in terms of the child mode function”. Indeed.

Also on the agenda were Samsung’s newest camcorders, reviews of which will be gracing this site soon. Apparently, while the sale of hard-disk camcorders has been falling, flash-memory-based ones are on the up. And little wonder given their small size. And also their ruggedness, since according to our Samsung man, “a hard-drive product, you shake it about and the needle will skip”.

Well, quite.

Perhaps it’s because it’s (relatively) early in the morning, but Intel appears to be making light of its €1 billion fine, levied for not playing fair with AMD. Or am I just reading too much into a feeble marketing line?

It’s a nice enough addition to the legendary DSLR range that began with the 300D in 2004, but the price is flabbergasting.

Granted, the new model has new features – HD video and a better (much better) screen – but none that justify a price hike of over 50% from the previous model. Yes that’s right – over 50%. When we reviewed the 450D at launch last year, the price with 18-55mm lens was £600. The new model is a sneeze short of a round grand, at a list price of £969.

A thousand pounds. For a camera that not even Canon will pretend is designed for professionals (or even very serious amateurs).

It’s not Canon’s fault, of course. Having recently finished writing a digital camera group test (which you’ll be able to read in the next issue of PC Pro, on the shelves around 15th May), an overriding theme was the fact that the pound sterling’s weakness (and the general state of the economy) is having a huge effect on prices across the board.

So anyway, I seem to recall some economist somewhere once saying that you could chart the progress of the economy in terms of the price of Mars bars. Apparently they track the state of fiscal play very well.

So I’ve decided to create a new economic index, designed to chart the progress of the economy in terms that techies can get to grips with.

It’s called the CameraIndex(tm), and it’s based on the price of successive models in the EOS xxxD camera line, from the groundbreaking EOS 300D in 2004 to the new EOS 500D. It’s exclusively revealed in the graph below:

Eeek. No wonder we got used to technology getting cheaper and cheaper. And even less wonder that a price hike is such a shock.

Just for fun, I thought I’d check my theory of a correlation with the wider economy, by superimposing the value of the UK’s FTSE 100 index – a fair-ish indicator of the general state of the economy – at the same points in time:

That’ll be quite a nice inverse correlation then.

If you’d like to license the use of the CameraIndex(tm), do please drop me a line. Subscription fees will be levied in f-stops (or possibly hundredths of a second – I haven’t decided yet).

It wasn’t the most surprising launch of a server processor ever, but server events aren’t usually renowned for thrills and spills.

Some interesting figures emerged though. They show in just two slides the market reality behind Sun being swallowed up by IBM a few weeks ago, and the sheer dominance of Intel when it comes to processors in everything from enterprise servers to netbooks.

With apologies for the shoddy photography, take a gander at the image below, and click on it for an enlargement if you’re struggling with the fuzziness:

Intel’s Tom Kilroy used this slide to show how the server market in general is growing, which it plainly is, from something like 250,000 units shipped worldwide in 1990, to around 8 million today. But it’s more interesting if you note the key: the blue part of the graph is Intel’s share; the red is competitors’.

The interesting part is that if you squint at the bottom-left of the graph you’ll see it’s all red up until 1992. In other words, back in the late 1980s and early 1990s Intel wasn’t selling any processors at all into servers. Bring it up to 2008 and you’ve got 7 out of 8 million – about 87% – with Intel inside (**BONG!!** bong bong bong!).

That’s a mighty fall for the old purveyors of RISC (as opposed to Intel’s CISC) architecture and what used to be a high-value, high-margin business. It’s always been one of Sun’s main markets.

If you look at another slide – even bearing in mind this is Intel’s own marketing material so was never going to come out in favour of the opposition – the future of non-x86 CPUs is looking decidedly ropey:

A tenth the cost and almost three times the performance of IBM’s Power6? Even taking the required pinch of salt into account, it’s hard to argue with numbers like that.

Intel’s commoditisation of the server market has followed unstoppably from its domination of desktop PCs. The economies of scale in using the same processor architecture and technology in servers, desktops and laptops seems to be pushing x86 unit cost down to a level that nobody else can begin to match. Add to that the compounding effect of the ever-increasing R&D budget that comes of selling so many processors – Power6, for instance, is still on 65nm fabrication when Intel is steaming ahead at 45nm – and you’ve got a terrifying level of domination.

In fact the only place in the market where Intel – or at least x86 architecture – is not dominating is at the very bottom end. With mobile-phone processors and ultra low-power embedded systems, the likes of ARM and Motorola are still hanging on.

In a conference call a late last year, Intel’s Pat Gelsinger stated it would be “decades” before Intel could displace ARM in very low-power devices.

Time will tell whether that was actually a very conservative estimate. I certainly wouldn’t bet on it.