HP Claims Invention Can Speed up Computers

This is not an incremental increase, mind you, but "reinventing the computer at the molecular scale." Sadly, there's no telling when this will leave the lab or trickle down to the rest of us.

"Hewlett-Packard Co. said Tuesday that its researchers have proven that a technology they invented could eventually replace the transistor, a fundamental building block of computers.

"In a paper published in Tuesday's Journal of Applied Physics, HP said three members of its Quantum Science Research group propose and demonstrate a 'crossbar latch,' which provides the signal restoration and inversion required for general computing without the need for transistors.

"HP said in a statement that the technology could result in computers that are thousands of times more powerful than those that exist today."

bah. I hate hp.

[quote author=quasi-modo link=topic=10408.msg97993#msg97993 date=1107402686]
I hate hp.
[/quote]
How dare they make technological advancements!

That would be a VERY impressive invention, if it's feasible to create and use. The most significant limitation on processor advancement is (and has been for some time) "how many transistors can we cram on the chip without them getting crossed up."

If the transistors can be completely replaced with something smaller and/or more efficient, this would be a major, major breakthrough.

So then I won't have to overclock to get all the speed neccessary? I have  my AMD 2000 overclocked so I can play Half Life 2. (Running at 2.04 gHz)

Wow.  That news could also explain why HPQ has gone up 2.2% today.  I'm assuming they announced this crossbar latch after they have filed the patenting paperwork, so this could be huge.  The only downside would be if it is too expensive to massproduce for a long time still, and thus only mainframes would benefit from it for a decade or more.  Thus it would have little effect on stock prices in the near term other than to replace IBMs share of mainframe sales.  However, IBM could lease the technology from HP and let IBM continue to make the mainframes.

If this works out, we can say goodbye to this...

[img]http://developer.intel.com/hardwaredesign/pix/devsite_img2.jpg[/img]
We shall create a processor with...one hundred BILLION transistors!

Would be way COOL to see a huge jump like that.  Imagine something they release in a few years compared to something from the government in 1960.  In the year 2010, 50 years & ALOT of progress.

For anyone interested, I found HP's press release on the subject (including an answer to Grok's patent question):
http://www.hp.com/hpinfo/newsroom/press/2005/050201a.html