back to article What's most important? Bandwidth over kilo-miles, or milli-watts?

AT&T boffins reckon they can fling 400Gb/sec down 12,000km of fibre using a new modulation technique. Meanwhile, IBM's bods say they managed 25Gb/sec over just a few millimetres - but using just 24 milliwatts. Both teams will present their research at next week's OFC-NFOEC conference in Anaheim, California, where the future of …

COMMENTS

This topic is closed for new posts.
  1. AndrueC Silver badge
    Boffin

    but there are plenty more frequencies available below that and we've a lot to learn about how to make the most of them

    I thought the higher 'radio' frequencies were blocked by the atmosphere and were unusable and of course the lower frequencies have insufficient bandwidth to be useful. I'm certainly no expert but I was under the impression we were already using as much EMF spectrum for transmission as was possible.

    I suppose the higher frequencies could be used in a vacuum but that won't be easy to maintain for most applications.

    1. Charlie Clark Silver badge

      For distance you have to use fibre because of the invariable ratio of power to distance squared and this limits you to light frequencies, by no means all of which are currently being used.

      Radio propagation is well understood and, while, there are still plenty of bands available (the ITU carves up spectrum in not the most efficient way) you have to trade bandwidth for propagation.

      The developments are complementary and impressive in their own right. Squeezing more out of an existing underwater cable is cheaper than laying a new one. Reducing power consumption while boosting data transmission will be welcome along the chain: in the server but also in the switches of the various NICs.

    2. TheVogon
      Mushroom

      High frequencies are attenuated, but can be used up to ~ 300Ghz for high bandwidth directional or short range applications.

      The Americans are also using 95GHz for their favourite method of communications: http://en.wikipedia.org/wiki/Active_Denial_System

      1. Rampant Spaniel

        I would pay good money for a usb ads dongle!

      2. Wardy01
        FAIL

        That might be so but that also has nothing to do with this article.

        They are specifically talking about sending light down a fibre and 300GHZ is not a light range frequency.

        The point being, that fibre technology is looking like it's on the horizon of some improvements !!

  2. Fred Flintstone Gold badge

    I am massively impressed by fibre

    I'm still astonished at how they can create glass that's so transparent, light can actually go that distance..

    1. Anonymous Coward
      Anonymous Coward

      Re: I am massively impressed by fibre

      Indeed. If you look into the edge of a piece of window glass, or get a pile of it, the distance you can see through it is just tens of centimetres.

  3. amanfromMars 1 Silver badge

    Unavailable Inapplicable Options in the Grand Design and Greater Schemas of Things Now and Yet to Be

    WHAT'S MOST IMPORTANT? BANDWIDTH OVER KILO-MILES, OR MILLI-WATTS? ... is the same sort of question answered with an understanding of the Churchillian Russian riddle wrapped in a mystery inside an enigma …. LIFE OR DEATH? WHICH DO YOU PREFER? ….. for the significance of one is as nothing without development in experience of the other, ergo are both a coincidental necessity for the greater singularity of fab and fabless purpose?

  4. Robert Carnegie Silver badge

    You don't think of laser beams with millimetre ranges,

    but I suppose it's dusty inside a PC. But maybe these devices are for use in special no-dust enclosures. At least.

    This also probably isn't going to work when the PC's lid is off (unless you have a darkroom! And photographers don't need those any more...) I once had a mouse problem; it was opto-mechanical, I had a desk next to a window, and the laser and sensor in the mouse didn't work in the afternoon, because it got direct sunlight then. So I bought a fluffy mouse cover with eyes and a tail, but "cover" was the main part of the solution, I suppose. You could also use duct tape, but it wasn't widely available then hereabouts.

    1. Ragequit

      Re: You don't think of laser beams with millimetre ranges,

      Interesting. I thought it was weird how they phrased a few things, but I got the impression that they were talking about cabled links that didn't require actual surface space on the board. Perhaps you're right though. They never did mention fiber for the millimeter link. The distance and bandwidth didn't seem high enough in either case. At least for something like a bus between processor/northside/southside. Maybe the plan is to use a bunch in parallel? As far as the dust goes it could be negated in fully submerged liquid cooling, but that makes for an even smaller list of applications.

      I can't help but feel I'm missing something obvious. Guess it's time for more caffeine.

    2. Rob Willett

      Re: You don't think of laser beams with millimetre ranges,

      This won't be used in a air open environment. There will be solid state devices perhaps a couple of mills in size which will have these lasers in. They will not be open to dust, or air or anything like.

      1. S4qFBxkFFg

        Re: You don't think of laser beams with millimetre ranges,

        They will not be open to dust, or air or anything like.

        I was imagining a tripped-out spider, who'd managed to accidentally stand where all eight eyes intersected different data paths and therefore believed it was receiving the mysteries of the universe.

    3. Anonymous Coward
      Anonymous Coward

      Re: You don't think of laser beams with millimetre ranges,

      IPC has penciled in the technology to embed optical fibers within a printed circuit board on their technology roadmap for quite some time (it was a "about 15 years or so" technology as of about APEX 2001 show, IIRC).

  5. John Smith 19 Gold badge
    Thumb Up

    12000 Km -> Trans Pacific cable with *no* repeaters

    Which radically simplifies cable design (no lumpy repeaters to design, no tricky power schemes to drive them).

    That means big savings in up front costs and repair bills as the only things likely to damage the cable are physically cutting them or material breakdown over time.

    The IBM stuff sounds like free space transmission IE no (very short) fibres on the board. Just lots of (highly directional) lights pointed at each other. (or possibly within the same package for intra-chip communications).

    Both very impressive. But I think the Bell work will see application before the IBM stuff.

    1. Fatman

      Re: 12000 Km -> Trans Pacific cable with *no* repeaters

      That (no repeaters) is exactly what the backbone carriers want; fire the laser into the cable, and out the other end come the "bits".

      While you mentioned the up front cost savings, you didn't mention the power consumed to feed existing repeaters on a submarine cable. In a world run by cheapskates, every $ that can be saved for executive bonuses, will be.

  6. This post has been deleted by its author

  7. Kevin McMurtrie Silver badge
    Trollface

    Delivered

    This is great news. Now AT&T can provide slow and overpriced Internet & TV bundles to 400 times as many people.

This topic is closed for new posts.

Other stories you might like