In our smartphone and tablet reviews we make sure to spend a good amount of time talking about the silicon powering these devices. There’s no reason that handset and tablet manufacturers shouldn’t be held to the same standards as the PC vendors we’ve worked with for years. 

Today the fastest phones are either based on ARM’s Cortex A8 core or a similar architecture as in the case of Qualcomm’s Snapdragon. Starting either late this year or more likely sometime next year we’ll see the first SoCs based on ARM’s first out of order core, the Cortex A9, shipping in phones. The roadmap doesn’t end there though.
 
Later this year ARM will officially announce the successor to the Cortex A9, codenamed Eagle. Today, Texas Instruments is announcing that it is the first company to license the ARM Eagle core.
 
The announcement goes further. Not only is TI licensing the core, but it also helped define the specifications for the core. TI has been working on the design with ARM since  June 2009. As a result, TI expects to be the first to market with SoCs based on ARM’s Eagle core.
 
Unfortunately there’s not much to say about Eagle itself until ARM makes its announcement later this year. TI’s Cortex A9 based SoCs (OMAP 4) will be shipping in Q4, showing up in devices in early 2011. Based on that schedule I wouldn’t expect to see Eagle anytime sooner than 2012. 
 
Eagle’s performance is slated to be much more competitive with future derivatives of Intel’s Moorestown SoC, while power consumption should be similar to existing designs thanks to the 2x-nm manufacturing process it will most likely be built on.
 
We’re still waiting to hear more details about the Eagle architecture but with today’s announcement, something from ARM can't be too far away.
Comments Locked

21 Comments

View All Comments

  • QChronoD - Tuesday, August 10, 2010 - link

    Maybe TI will finally come out with an even fancier calculator?
  • icrf - Tuesday, August 10, 2010 - link

    http://xkcd.com/768/

    I hadn't really thought about it before, but graphing calculators really haven't much advanced at all.
  • name99 - Tuesday, August 10, 2010 - link

    Last time I looked at this (which was about 7 years ago) the TI graphing calculators were based on a 68000.
    They could have been ported to ARM since then, but I can't think of any compelling business reason to do so. It's nice to have a segment of the market to yourself, with, apparently, no-one else in the world, not even some desperate and hungry Chinese or Indian company, interested in competing --- allows you to market the exact same tech product, at the exact same price, unchanged for ten years.

    [I would say the downside to this strategy is that it tarnishes your consumer brand, since it very obviously paints you as having zero interest in innovation; but I can't think of a single consumer product apart from graphing calculators that TI sells nowadays, so they appear to have already achieved that aim.

    They also have not, for example, as far as I know, ported their calculator code (the fancy stuff handling matrices, calculus, etc) to an app that runs on iPhone or Android, which seems to suggest that the division is on complete autopilot, run by people who assume it's one day soon going to stop making money, and with zero interest in changing that situation. ]
  • Exodite - Wednesday, August 11, 2010 - link

    "It's nice to have a segment of the market to yourself, with, apparently, no-one else in the world, not even some desperate and hungry Chinese or Indian company, interested in competing --- allows you to market the exact same tech product, at the exact same price, unchanged for ten years."

    I can't speak about the US situation but over here in Sweden the more popular brand of graphing calculators were always Casio.
  • jabber - Tuesday, August 10, 2010 - link

    The reason could well be they are about 1000th the size of INTEL.

    ARM are a cottage industry compared to the big boys.

    As a shareholder I'm more than happy with that.
  • Exodite - Tuesday, August 10, 2010 - link

    "ARM are a cottage industry compared to the big boys."

    For now, and possibly due to their marketing strategy, but I'd say that ARM is on the up and up what with the ubiquitous prevalence of MIDs in various forms.
  • metafor - Tuesday, August 10, 2010 - link

    There are around 9 billion ARM chips sold every quarter. That absolutely dwarfs Intel's numbers. The thing is, ARM themselves don't make that much money (nor do they make the chips). The ARM licensees do. And if you look at TI, Qualcomm, Marvell, etc. They are anything but "cottage industry".
  • hyvonen - Wednesday, August 11, 2010 - link

    "There are around 9 billion ARM chips sold every quarter."

    What?? Are you saying that, on average, every person on earth is buying at least one ARM-based gadget per quarter???

    Any link/references to prove this seemingly ridiculous claim?
  • Penti - Sunday, August 15, 2010 - link

    It's not that ridiculous it's integrated in everything your car has several ARM processors, your washing machine might have one, your cellphone might have two of them, either as package on package or two separate chips. Pretty much everything with a microcontroller is ARM these days. Devices might implement several ARM-processors in one unit, cars have dozens.

    But the number is of course not accurate, it's billions each year.
  • TareX - Thursday, August 12, 2010 - link

    I actually don't think I'll need anything faster than Tegra 2, till 3D stereoscopic screens become mainstream on mobile devices.

Log in

Don't have an account? Sign up now