Difference between revisions of "20 April 2010"
(a short memristor article) |
(better link to main story) |
||
| (One intermediate revision by the same user not shown) | |||
| Line 2: | Line 2: | ||
In 1971, electrical engineering professor [[w:Leon Chua|Leon Chua]] proposed a theoretical basic electronics component called a [[w:Memristor|memristor]]. In 2008, Hewlett Packard brought the memristor out of theory and into the real world. See [http://www.bit-tech.net/news/2008/05/02/memristor-the-fourth-circuit-element/1 Memristor: the fourth circuit element]. | In 1971, electrical engineering professor [[w:Leon Chua|Leon Chua]] proposed a theoretical basic electronics component called a [[w:Memristor|memristor]]. In 2008, Hewlett Packard brought the memristor out of theory and into the real world. See [http://www.bit-tech.net/news/2008/05/02/memristor-the-fourth-circuit-element/1 Memristor: the fourth circuit element]. | ||
| − | Memristors have similar physical behavior as [[w:Synapse|synapse]]s in the human brain. Production scale memristor memory with tens of billions of memristors should mean that memristors that mimic over one hundred billion synapses could only be a few years away. The human brain has one hundred trillion synapses. A human scale emulation of one hundred trillion synapses might only take one hundred near term memristor chips. Having comparable hardware in terms of numbers of similar components does not mean you can make them behave like a brain, but getting hardware with the right number of neurons and synapses would give researchers a reasonable chance at human brain emulation. The hardware could be ready in the 2020-2025 timeframe for neurons and synapses. [http:// | + | Memristors have similar physical behavior as [[w:Synapse|synapse]]s in the human brain. Production scale memristor memory with tens of billions of memristors should mean that memristors that mimic over one hundred billion synapses could only be a few years away. The human brain has one hundred trillion synapses. A human scale emulation of one hundred trillion synapses might only take one hundred near term memristor chips. Having comparable hardware in terms of numbers of similar components does not mean you can make them behave like a brain, but getting hardware with the right number of neurons and synapses would give researchers a reasonable chance at human brain emulation. The hardware could be ready in the 2020-2025 timeframe for neurons and synapses. [http://www.technologyreview.com/computing/25018/page1/ more...] |
| + | *[http://www.youtube.com/watch?v=rvA5r4LtVnc 6 minute memristor guide] | ||
| + | *[http://www.youtube.com/watch#!v=mUZ2MhZ8SZU&feature=related What is a Memristor?] ''- by Leon Chua'' | ||
}} | }} | ||
Latest revision as of 00:54, 8 June 2010
|
Memristor chips coming soon
In 1971, electrical engineering professor Leon Chua proposed a theoretical basic electronics component called a memristor. In 2008, Hewlett Packard brought the memristor out of theory and into the real world. See Memristor: the fourth circuit element.
Memristors have similar physical behavior as synapses in the human brain. Production scale memristor memory with tens of billions of memristors should mean that memristors that mimic over one hundred billion synapses could only be a few years away. The human brain has one hundred trillion synapses. A human scale emulation of one hundred trillion synapses might only take one hundred near term memristor chips. Having comparable hardware in terms of numbers of similar components does not mean you can make them behave like a brain, but getting hardware with the right number of neurons and synapses would give researchers a reasonable chance at human brain emulation. The hardware could be ready in the 2020-2025 timeframe for neurons and synapses. more...
- 6 minute memristor guide
- What is a Memristor? - by Leon Chua
