back to article Here's one way past Moore's law: Chips that mix photonics and electronics

As Intel, TSMC, and Samsung race to prove that Moore's law is still relevant with faster and more efficient chips, researchers in China have pointed to a growing body of research revealing one way silicon slingers can achieve higher levels of performance: matrix math accelerators that integrate electronic circuits and photonics …

  1. Pascal Monett Silver badge

    "Provided these opto-processors make it out of the lab"

    You mean, like the carbon nanotubes we've been promised for the last 25 years ?

    Or the revolutionary batteries we still can't buy but have not stopped hearing about ?

    Research is indispensable, but I would really like for these people to stop hyping up a product that doesn't exist and won't for decades.

    It's like that fly at the picnic. Won't go away, and you can't do anything about it except shoo it off.

    Annoying.

    Call me back when they have a product that sells.

    1. Def Silver badge
      Headmaster

      Re: "Provided these opto-processors make it out of the lab"

      Carbon Nanotubes have been used in commercial products for at least eight years now.

      1. Filippo Silver badge

        Re: "Provided these opto-processors make it out of the lab"

        And battery energy density has more than doubled in the last 10 years, while costs have plummeted.

    2. Dave 126 Silver badge

      Re: "Provided these opto-processors make it out of the lab"

      This is a technology news site: it will cover things that happen in laboratories as well as boardrooms. If you only care about what is on shop shelves today, look in a shop.

      I do agree that some technologies take longer to reach maturity than others. But if I feel that I hear too much of them over the years (okay, decades) it's my fault for reading a technology website.

    3. Roland6 Silver badge

      Re: "Provided these opto-processors make it out of the lab"

      >Call me back when they have a product that sells.

      Well, given they say a good application is linear matrix computations, it would seem the production of a linear matrix computations "co-processor" - in a similar vein to the floating-point maths co-processor (remember the 8087 through to 80387 series of math coprocessors), would be in order.

  2. Dave 126 Silver badge

    So, on the way towards Buck Rogers in the 25th Century then: "Ha, their computers are so primitive they still run on electricity!!". (That was an episode from the 80s, in the 90s pop sci TV programme Tomorrow's World had a bit about photonic computing's promise to radically shorten interconnects. Been a long time coming.)

    Whereas computers used on the bridge in Star Trek are evidently so advanced they burst into flames if a photo torpedo hits the far end of the starship - perhaps they run on hydrogen?

    Joking aside, good luck to those working in this area.

    1. simpfeld

      Problems with pure optical computing

      There have been fundamental problems with photonic computing for a long time.

      One is heat dissipation, when an optical gate is off it effectively goes black so just generates heat. Transistors don't do this.

      Secondly, light is quite large. Chips are already have feature massively smaller than optical wavelength, lithography requiring ultraviolet light these days. So any optical chip will have to be larger (features wise) than an equivalent electronic chip.

      So basically for bulk across board/CPU maybe high bandwidth comms is the likely immediate and perhaps only possible use for photonics in the immediate future.

      Surface plasmons are probably a more likely path for actual computation.

      One other thing is that optical fibres have a pretty slow propagation rate has a VF (Velocity Factor) of about 0.6-0.7 (of the speed of light in a vacuum), decent coax can do 0.8-0.9. This higher bandwidth of fibre usually makes this much more preferable but for certain in computer functions this maybe a limiting factor (memory buses perhaps). I just mention this as hardly anyone seems to know this.

      1. Dave 126 Silver badge

        Re: Problems with pure optical computing

        Thank you for a clear and informative post about the challenges and desirability of photonics.

        I wonder if you have any thoughts on using physical objects as computing mediums:

        https://www.quantamagazine.org/how-to-make-the-universe-think-for-us-20220531/

        Cheers

    2. Mike 137 Silver badge

      "Been a long time coming"

      Indeed it has. Optical circuitry (including digital) was being worked on at UCL in the late '80s (and not just about shortening interconnects).

  3. DJV Silver badge

    The boffins have obviously seen the light!

    (see above)

  4. bombastic bob Silver badge
    Meh

    In reference to the 2015 announcement

    seems like we might know where they [possibly] stole the tech from....

    (I have a hard time believing that any kind of REAL innovation can come out of a communist nation, with scientists living under a constant threat of social credit scores and censorship. Such things are NOT conducive to creativity and innovation - and the CCP is well known to have stolen LOTS of tech over the last decade or so)

    The tech is probably usable, at some point. Let's hope hat the "prior art" of 2015 does not allow the CCP to leverage the rest of the world with encumbering patents for [possibly] stolen tech.

    [same kind of thinking with respect to 5G also, in recent years]

    1. Mike 137 Silver badge

      Re: In reference to the 2015 announcement

      "I have a hard time believing that any kind of REAL innovation can come out of a communist nation"

      They did pretty well in the intelligence technology field (e.g. see E. Haseltine, The spy in Moscow Station, pub. Thomas Dunne, USA 2019; Icon Books, UK 2019). And, just for another example, some of their mathematicians have been world leading.

    2. Dave 126 Silver badge

      Re: In reference to the 2015 announcement

      Well, he's attempting to qualify his statements and not all of his words are capitalised. That's progress.

  5. ICL1900-G3 Bronze badge

    Boffins?

    ... Eggheads? What is this, the Daily Mail?

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

Other stories you might like