Moore's Law On Hold?
show comments
  • Marcus V

    Moore’s Law as it’s thought of today– which is actually an elaboration on what Moore said, not just his words as gospel– is a little more subtle than “Make transistors smaller! Make them faster!”

    It’s not (necessarily) about transistors, it’s about processing power; and it’s not just about processing power, but processing power per unit cost.

    The death of Moore’s Law was hailed when transistors stopped getting twice as fast every 24 months, but processing power per unit cost has still been dropping accordingly. That’s the important part, and while there are some limits on that metric as well, the article didn’t really speak to them at all.

  • Walter,

    A new way of designing chips is required. We need chips that don’t use any power while they are waiting for something to happen. And that do not use much power when something happens.

    There is such a chip.

    More computing power than a desktop computer and it uses about 1 tenth (0.1) of a watt at full clip.

    The current version is optimized for control applications. There is no reason a desktop version (or an array of these chips) couldn’t be used for a desktop machine. Even better would be using them in a laptop. The battery might last days.

  • Luke Lea

    Trends that can’t go on forever don’t. Besides heat not speed is the bigger issue today.

  • I was in error the chip does over 90 B(illion)IPS and uses 650 mw (that is a little over six tenths of a watt) and microwatts if completely idle. And what ever in between is needed to get the job done.

© The American Interest LLC 2005-2017 About Us Masthead Submissions Advertise Customer Service
We are a participant in the Amazon Services LLC Associates Program, an affiliate advertising program designed to provide a means for us to earn fees by linking to and affiliated sites.