ARM’s big reveal today of its latest processor design came with the clear message that the company will use its market weight and mobile heritage to fend off challenges from Intel.

Speaking here at the ARM Developers Conference, CEO Warren East began the hyping exercise around the Cortex-A9 processors. ARM plans to provide both multi- and single-core versions of the new product, which should arrive in some early devices by 2010. The new chip is aimed at powerful, portable products and should show up to eight times the performance of the ARM chip in the iPhone, according to East.

ARM has spent years building up partnerships across a broad set of markets. To that end, the company announced that NEC, Nvidia, TI, Samsung and STMicroelectronics have all committed to using the Cortex-A9 design.

ARM hopes to leverage partners like these against Intel, which is re-entering the mobile device market after trying and failing at the effort once already. Next year, Intel should release a product code-named Silverthorne that relies on the x86 instruction set and consumes about one-tenth the power of Intel’s mobile chips. Another chip called Moorestown should appear in 2010 and consume well under a watt, putting the product more in line with ARM’s processors.

But where Intel looks to push on its own, ARM relies on a host of friends.

“You are not just engaging with ARM people. You are engaging with a huge partnership,” East said, during his speech. “We believe that working in partnership creates much better solutions.”

As the market leader, ARM has proved reluctant to address Intel’s new mobile charge during public forums. Why acknowledge this bothersome upstart until it has proven technology in the mobile device market – even if the upstart has billions at its disposal?

East, however, tackled the subject during an interview with The Register.

“This is not an ARM versus Intel thing at all,” East told us. “This is a bunch of semiconductor companies. It is that collection of companies – not ARM – that is active in that space.”

As East puts it, ARM remains content to live in a world of “anonymous ubiquity” where those companies that license its products do the dirty marketing and sales work. Let Intel flex its muscle and go up against these giants, ARM seems to say. Let Intel see if these other vendors want it in the game.

After a bit more prodding, East was far more direct about his thoughts on Chipzilla.

“I don’t actually believe there is any struggle to be had really,” he said. “It is almost no contest. We are an order of magnitude ahead. We have more than 200 semiconductor licensees.

“Intel is a very big company, but they are not bigger than the whole ARM collective. They are very good at spending on marketing, but the ARM collective spends more. I just don’t think there is that much of an issue to be had.”

Ah, but perhaps ARM will relent on its anonymity a bit with the Intel threat looming. Why not shove an ARM logo on every iPhone that goes out the door?

“I won’t pretend that is not a debate that we’ve had,” East said. “We live in a business-to-business world, and it costs money to make those sorts of demands. Also, I am not entirely convinced that following the 'Intel Inside' campaign is something the industry will be ready to do again anyway.”

So, that would be a no.

A9 - Bingo!

Looking back at the Cortex-A9 part, you’ll find an 8-stage superscalar, multi-issue pipeline with speculative out-of-order execution. The device can crunch four instructions per cycle at greater than 1.0GHz.

As mentioned, ARM will ship the multi-core Cortex-A9 MPCore and the single-core Cortex-A9. The multi-core chip (up to four cores) will handle more demanding tasks such as churning through multimedia software, while the single-core part will go after lower-cost, lower-end parts of the mobile market.

All told, the Cortex-A9 products can show up to 8000 aggregate DMIPS in performance while eating up less than 250 milliwatts.

ARM says the power of its chips will make them best suited to a oncoming new class of mobile devices that apparently deliver where ultra mobile PCs and smartphones have failed. But, even if some new type of device does not emerge, the fresh ARM chip should improve the performance of existing applications such as video on phones while needed far less power than today’s chips.

Should the new devices take off, ARM think it is better suited to address market needs than a company like Intel with a tradition in the power-hungry PC market.

"Power is really important here and clearly the evolution from the phone end is some way ahead versus the evolution from the PC in terms of delivering a sensible amount of battery life to a user," East said.®