Tuesday, February 01, 2005

Crossbar latch & fast computers

This looks like an impressive discovery in nanophysics. Today, Hewlett-Packard has announced in Journal of Applied Physics
that they invented a "crossbar latch", a new cheap, nanometer-sized component at the molecular level that has the potential to replace the transistors and speed up the computers by three orders of magnitude.

The patent was actually registered already in 2003:

It will probably take years before the new computers can be developed using this technology, if it ever happens.

See also Hewlett-Packard's memristors good for memory chips.


  1. It will work within 5 years, I think!

  2. "the switches in the experimental crossbar latches only flip at about a tenth of a second. "


    a hundred million times speed-up will need to be achieved to come anywhere close to today's silicon.

  3. Sorry to post a comment on an unrelated topic, but
    your past extensive discussion of Larry summers begs the question.
    Today, another University Professor actually risks being fired due to something he wrote other people found offensive.
    (According to Drudge, the Board of Regents has allready asked him to resign).

    Would just like to ask if Summers freedom to say offensive stuff does not also apply to Prof. Ward Churchill.

    To me, the only significant difference is that in one case the offensive statements, said at an official function by the university's president, might discourage Women to study science at Harvard. In the other, I fail to see how the offensive statement written by Ward Churchill might materially damage anyone.

    Yet it seems the "liberal academia" is much tougher on violating Churchill's rights than Summers's.

  4. The professor in Colorado enjoys free speech, too, and I am supporting free speech for him.

    For others: Ward Churchill compared the World Trade Center *victims* to Nazis. I think that this professor should have all the freedoms in the USA, but as soon as he transmutes his opinions into acts, he should be dealt with much like other terrorists in the war on terror.

    Incidentally, I was pleased by the large number of terrorists that have been eliminated in Iraq in the last 6 months or so.

    Also, comparing Professor Summers with his completely mainstream, kind, common sense opinions to this perverse Ward Churchill is a bit over the edge, is not it? Maybe Ward Churchill needs a medical assistance more than prison.

  5. I will keep on erasing most messages by "quantoken". Most of them are full of absolutely weird hypotheses and accusations, and I don't want this trash to grow on my blog.

  6. "... a new cheap, nanometer-sized component at the molecular level that has the potential to replace the transistors and speed up the computers ..."

    I see that lumo has discovered 'the hype', a cousin to 'consensus'. No, wait, LQG is a hype... :-)

    Ordinary silicon, at room temperature, will in all likelihood ([1-10{-99}] in lumo parlance) continue to be the practical material for years to come, and thereafter as base material and for interfacing.

    The latch use polymers, which today are mainly a few high-temperature types for passive functions (insulation and/or topography).

    As a tip, look at carbon nanotube technology (nanoelectronic or nanomechanic); carbon is already a familiar material for Si/SiGe straining, and compatible with much of the processing temperature budget. (Each step must processed at succesively lower temperatures; or high temperature at very short time, it's a non-linear relationship.)