Posted on 04/09/2010 5:46:29 AM PDT by Erik Latranyi
Hewlett-Packard says it has made a discovery in electrical engineering that could one day lead to computer systems and handheld devices that are dramatically faster and more energy efficient than what's possible using today's technology.
The discovery relates to technology based on a basic element of electrical engineering called a "memristor," which an HP Labs team first demonstrated in 2008 as a resistor with memory. While researchers initially believed the memristor-based technology would only be useful in storage devices, they have recently learned that it can perform logic.
If such technology can one day be used in developing a new type of chip, then the silicon will be able to perform computations where data is stored, which could greatly increase processing speeds, according to HP. CPUs used in today's computer systems process information brought from other parts of a computer system.
HP has designed a new architecturewithin which multiple layers of memristor memory can be stacked on top of each other in a single chip. As a result, the company believes devices incorporating memristor-based chips could hit the market in five years.
(Excerpt) Read more at informationweek.com ...
/mark
ping
It’s always the “if” isn’t it.
Too many just pump out press releases, me I want to see production samples. Trust but verify.
Impossible. Scientists long ago argued everything that could be invented has been invented. Trust the scientists.
Starting countdown to the applephiles saying, “so basically it will allow HPs running Microsloth to blue screen that much faster”
And if it holds up, then this is the first fundamental advance in electrical engineering since the invention of the transistor.
I’m still waiting for magnetic bubble memory.
Now that would really be neural networking!
I tend to agree - I’ve been following these developments in the trade press. Sure there are always the breathless pronouncements of how this is the next big thing - but in this case I think it may live up to the hype.
This would be wicked-cool in FPGAs :-)!
Also, LOL at the “Apple User” pic ... I am stealing that one :-)!
Nah. Memristors and memristive systems have been discussed and worked on for decades. The name and theory goes back to the 1970's.
This work at HP is a useful advance, no argument, but 5 minutes on the Wikipedia page about Memristors puts this in context.
This article is breathless journalism, mostly. Cheers for the team at HP, yes; but let's not get too carried away.
They were too slow, flash drove by before they ever got out of the garage.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.