Intel PC hegemony facing ARMy attack
ARM set for '13% share by 2015'
ARM-based processors will bite off a significant chunk of Intel's PC hegemony in the next few years, achieving a 13 per cent share of the PC-processor market by 2015.
That's one conclusion of "Worldwide PC Microprocessor 1Q11 Vendor Shares", an IDC report released Thursday.
Notice that IDC's report focuses on PC processors – not merely processors in general. It doesn't take into account all those ARM-based processors that are already powering tablets and smartphones.
"While media tablets are a competitive factor at the system level because of competition for share of consumer wallets vs PCs," the report's author, Shane Rau, told The Reg in an email exchange. "Media tablets are not PCs in IDC's taxonomy, and the processors inside them are not PC processors. I define a PC processor by the design intent of the manufacturer."
And IDC sees that design intent of ARM-based processor manufacturers aimed squarely at Intel's turf – just as AMD, newly revitalized with its Fusion line of accelerated processing units (APUs), is aiming to grab a larger stake in the PC-processor marketplace.
ARM-processor manufacturers are starting small in their PC ambition. As Rau told us, "While there are some ARM-based netbooks (which are called mini-notebooks in IDC's taxonomy and are part of the PC taxonomy), the volumes are very small."
We asked Rau whether Microsoft's plan to port Windows 8 to ARM was a major reason for his projection of a 13 per cent ARM PC-processor market share in a mere four years, he agreed. "Yes," he answered. "Windows 8 moving to ARM is the major assumption in the long-term forecast for ARM in PCs. As Nvidia, Qualcomm, and TI announced at CES, they'll support Windows 8 with ARM-based SoCs, so that's a clear sign of intent to design for PCs and so the market direction."
Rau's "yes", however, was followed by a big ol' "but". "Other factors include ARM architectures that support PC functionality (with features like 64-bit processing, 64-bit memory addressing, high-speed interfaces, and large caches) and PC applications coming from ISVs," he cautioned. "If any major part of the necessary ecosystem for PCs falters, then ARM in PCs will falter. This is a long-term roadmap."
In other words, yes, Intel does face a challenge from the lower-power end of the PC-processor market, but a lot of independent ducks must line up in their correct rows for Chipzilla to feel major pain.
What's more, Wednesday's announcement of Intel's "tri-gate" 22nm processors may portend a better power/performance future for a company that has been late to the low-power, high-performance game.
That said, as Intel moves down the power ladder, ARM-based processors such as Nvidia's quad-core ARM Cortex-A9 Kal-el are moving up the performance rungs. And then there are those muscular ARM Cortex-A15 parts scheduled for late 2012 or early 2013.
Thing haven't been so much fun in the world of PC processors since Motorola/IBM/Freescale's PowerPC dropped out of the mêlée some years back. ®
COMMENTS
Design intent
"Media tablets are not PCs in IDC's taxonomy, and the processors inside them are not PC processors. I define a PC processor by the design intent of the manufacturer."
The original design intent of the ARM processor was for use in a PC, namely the Acorn Archimedes, and later the RISCPC. They were simply more successful selling to the embedded market where conformance to a specific chip type (ie x86) to run a specific OS (windows) wasn't a big deal.
Fun!
Certainly very interesting times. Just imagine the inquiry inside Intel should ARM succeed in snatching a large and damaging market share; just how did a pokey little design house from somewhere flat, cold and wet without even a small fab to their name manage to out maneouvre the mighty Intel? I would be very interested to know if ARM's design team staff count is larger or smaller than Intel's.
If this does indeed come to pass, ARM will definitely have been a 'slow burner'. It's taken 20ish years to get this far, not exactly the fastest growth curve we've ever seen.
X86 has had a very impressively long run so far but the fantastic growth of mobile and datacentre applications has really underlined the penalties of the x86 architecture; power consumption. Intel are trying to keep up with clever silicon manufacturing processes, but you can't escape the fact that an ARM chip implemented on the same processes is smaller, cheaper and lower power. They once had an ARM license (StrongARM / Xscale) but disposed of it and haven't managed to compete since. Big mistake?
Intel could win if they bought ARM and wiped them out or renamed the ARM instruction set as x86-lite. I'm amazed that they haven't tried to do so as yet. It would raise the mother of all Competition Commission / SEC antitrust inquiries, and I don't think that Intel would win that one.
Would be nice ...
... having an alternate architecture available (as in readily available, reasonably priced laptops, desktops, servers). Also, the ARM model of licensing the designs to umpteen chipmakers would seem like an healthier one than the current Intel hegemony.
Acorn customers were right after all?
Nice to see that people that bought into the Atom, BBC micro and the Archimedes are seeing the fruits of their endeavours at last.
The Windows effect
I have a feeling that a large part of the predicted increase is because Microsoft has announced that Windows 8 will run on ARM as well as on x86. Industry predictors like IDC tend to ignore all other than Windows when talking about PCs (though they might be aware of a one-digit fairly steady presence by Apple), so they would be taken completely by surprise if something other than Windows took a two-digit percentage of the PC market (defined as desktop and full-size laptop computers). So if they predict a two-digit share by ARM, this is probably because they believe a two-digit percentage of Windows-8 PCs will use ARM.
I hope that we will see a broader picture than that, with ARMs in PC-class computers based on non-Windows operating systems such as Android, WebOS and Linux.
