What is Digg?Sponsored by Travelzoo
$52 and Up—Airlines Slash Fares On Peak Holiday Flights. view!
travelzoo.com - This year, waiting until the last minute is NOT the best strategy. See why.
145 Comments
- quill, on 05/28/2008, -1/+110"Memristors--the fourth passive component type after resistors, capacitors and inductors--were postulated in a seminal 1971 paper"
I love it when we can deduce, via math and/or theoretical physics, the existence of something before it can actually be detected/invented. (I like to use Neutrinos as an example of this.)
Science: It works, bitches! - brstilson, on 05/28/2008, -2/+79For those of you not familiar with this stuff, this is big. This is invention-of-the-transistor big, except even bigger, and we saw what transistors did for electronics.
- Gojirra, on 05/28/2008, -2/+43Can we call it something cooler, like a flux capacitor?
- offthewagon, on 05/28/2008, -1/+40So, when robots make slaves of us all we can thank the memristor?
- darkcooger, on 05/28/2008, -1/+31I especially liked the part where all of our electronic theory has been based on the wrong variables. This is really two discoveries in one, and there's no telling what other new things might be just around the corner with a whole new perception of electronic theory.
- EarlOfLade, on 05/28/2008, -1/+31I foresee a Nobel Prize for Chua and Williams.
- lordewoks, on 05/28/2008, -0/+26"We want to build memristor-based devices that operate in a manner similar to how the synapse works in the brain--neuron-like analog computational elements that could perform control functions where decisions must be made involving comparisons as to whether something is larger or smaller than something else. We are not building a neural network yet, but we think that using the memristor in its analog mode with our crossbar is a pretty good representation of a neural net."
Sweet, here comes Data! - EricAnderton, on 05/28/2008, -0/+20I agree. This is very, very big. HP is going to start printing their own money before too long.
* partially or wholly analog computers
* outright replacement of flash and other solid-state storage circuit designs
* more reliable chips at higher circuit densities than before
* rendering obsolte a good portion of EE cannon (i.e. time to re-write the textbooks) - vroom101, on 05/28/2008, -0/+19Here's an excellent explanation of how the memristor works via the IEEE Spectrum article "The Mysterious Memristor" by Sally Adee (May 2008) at http://www.spectrum.ieee.org/may08/6207 :
The reason that the memristor is radically different from the other fundamental circuit elements is that, unlike them, it carries a memory of its past. When you turn off the voltage to the circuit, the memristor still remembers how much was applied before and for how long. That's an effect that can't be duplicated by any circuit combination of resistors, capacitors, and inductors, which is why the memristor qualifies as a fundamental circuit element.
The classic analogy for a resistor is a pipe through which water (electricity) runs. The width of the pipe is analogous to the resistance of the flow of current--the narrower the pipe, the greater the resistance. Normal resistors have an unchanging pipe size. A memristor, on the other hand, changes with the amount of water that gets pushed through. If you push water through the pipe in one direction, the pipe gets larger (less resistive). If you push the water in the other direction, the pipe gets smaller (more resistive). And the memristor remembers. When the water flow is turned off, the pipe size does not change. - Gojirra, on 05/28/2008, -2/+19I know it sounds crazy, but maybe you could read the article that has been conveniently linked from this page?
- vroom101, on 05/28/2008, -0/+17Yes.
- kirijitsu, on 05/28/2008, -0/+16Memristor awesomeness aside, I think the most important part of the article is how Chua notes we've been misunderstanding electronic theory at a basic level all along.
"Electronic theorists have been using the wrong pair of variables all these years--voltage and charge. The missing part of electronic theory was that the fundamental pair of variables is flux and charge," said Chua. "The situation is analogous to what is called "Aristotle's Law of Motion, which was wrong, because he said that force must be proportional to velocity. That misled people for 2000 years until Newton came along and pointed out that Aristotle was using the wrong variables. Newton said that force is proportional to acceleration--the change in velocity. This is exactly the situation with electronic circuit theory today. All electronic textbooks have been teaching using the wrong variables--voltage and charge--explaining away inaccuracies as anomalies. What they should have been teaching is the relationship between changes in voltage, or flux, and charge."
The age of electronics just got a shot of some serious performance-enhancing drugs. - twiztidsinz, on 05/28/2008, -0/+15More time finding *new* stuff, less time bitching, please
- darkcooger, on 05/28/2008, -0/+14No, that is not already the case. Transistors gets less and less efficient at smaller scales due to leakage currents. The smaller the scale, the higher the percentage of current is leaked. I understand that Intel has made a lot of headway in reducing the leakage current, but I suspect that physics simply makes it impossible to eliminate it completely without a fundamental change in the way transistors are constructed.
- arunforce, on 05/28/2008, -2/+16I'm curious to when it will appear in the market.
- EarlOfLade, on 05/28/2008, -0/+12Well, it may allow you to store the entire iTune catalog on a device the size of a USB stick, along with a few thousand HD movies..
So, it may not be much of a use to you but... - numb, on 05/28/2008, -1/+12FTA: "Memristors will enable very small nanoscale devices to be made without generating all the excess heat that scaling down transistors is causing today."
- talonstriker, on 05/28/2008, -0/+10Aw man, I just finished my circuit theory course this semester. Will this thing start to replace memory chips in the future?
- vroom101, on 05/28/2008, -0/+9Not so fast. The memristor can operate as a digital device OR analog device. From the article:
As Chua predicted, Williams is already thinking about creating new types of devices with HP's crossbar architecture beyond a simple memory device. "If we push current through it hard and fast, it acts like a digital device, but if we run current through it gently and slowly it acts as an analog device," said Williams. "We are already designing new types of circuits in both the digital and analog domains using our crossbar architecture. . . . - fasda, on 05/28/2008, -0/+9I see a patent for HP worth Billions of dollars
- cdawzrd, on 05/28/2008, -0/+9It varies its resistance based on change in voltage... it has a "memory" for the last voltage change going through it. Or at least that's my understanding.
- DharmaDog, on 05/28/2008, -3/+12Repeat, but still very cool.
Yea, HP! - EricAnderton, on 05/28/2008, -0/+9Exactly. Cue the sound of thousands of EE students who just realized that they'll never be able to sell their books back.
- bgrah449, on 05/28/2008, -1/+10"A concept is useless with no possible execution." Yeah, except there was a possible execution. They just didn't know what it was. Don't knock them for getting the baton halfway to the finish line in the relay race.
- vroom101, on 05/28/2008, -0/+9The theory is already known. HP has already built the device. I seriously doubt it'll take 10-15 years for this technology to hit the market.
- majordanger, on 05/28/2008, -0/+9Not to Fret.. When I was in school, we used 30 watt soldering irons,punch cards and Fortran.
I don't use too much of that today.
Now I'm trying to figure out how to probe a signal on a 568 ball DSP the size of my thumbnail.
Remember, They did teach you how to read. - willy3121, on 05/28/2008, -1/+9Bad timing = spending the last four years getting a bachelor's degree in EE instead of the next four :(
- trevorh, on 05/28/2008, -0/+8The memristor can have it's resistance value changed by direct current it will remember that value but alternating current can be can be used to read the resistance value of the memristor with out changing the value.
This is not a read once technology it can be read over and over until DC is run to to change the value of the memristor. - oneoverzero, on 05/28/2008, -0/+8They can patent a technique for making them, but not the idea itself.
- vroom101, on 05/28/2008, -1/+8Imagine this: An Intel or AMD processor chip running at 3 GHz, with ALL the memory, i.e., the maximum of memory the CPU can address, built using the memristor on the same wafer (chip) as the CPU logic -- and the memory is non-volatile, i.e., when the power is turned off the memory isn't erased. See http://digg.com/hardware/Missing_link_memristor_cr ... or
What need would there be for a CPU cache? - PhrosTT, on 05/28/2008, -1/+8like bose einstein condensate
- PhrosTT, on 05/28/2008, -1/+8HPQ Stock ?
- flickr, on 05/28/2008, -2/+9Sarah Connors has failed...
- hfactor, on 05/28/2008, -1/+7Hmm, I'd rather give credit to the man who invented the aqueduct than to the men who built one. Without someone postulating that the world is round, how could someone set sail to prove it? Also, lots of mathematical theorems have been proved without even dreaming of an application - and decades later they found one in a previously nonexistant field (like cryptography)...
- edmcguirk, on 05/28/2008, -0/+6While I am an EE from the previous century, I think that the idea that the memristor equations are based on flux and charge instead of voltage and charge probably means that those graphs you refer to need to be changed.
The memristor doesn't fit on the old graphs, we need to create new graphs where the different devices will all fit. - rmxz, on 05/28/2008, -0/+6I think they're overstating things a bit. Every EE student in history has seen passive components with memories that can increase their resistance when a high enough current passes through them. Your old car has them (called fuses) and your house has them (called circuit breakers) both of which increase their resistance with a current. And at non-nano-scales people made passive devices with similar properties that could reset themselves (bubble memories, core memories, etc).
Now this memristor *IS* a really really cool device - a nano-scale circuit breaker that can reset itself (lower it's resistance again) - and I think it will revolutionize electronics. But is it really as magical as they're trying to make it sound? - Elliuotatar, on 05/28/2008, -0/+5This may have been news a little while ago, but this is the first article I've seen which explains that it changes the fundamental understanding of how all electronics work.
- jrhelgeson, on 05/28/2008, -1/+6HP can't patent this, can they? This is a basic scientific discovery, not an invention. If they patent this 'technique' then we really need to just storm the patent office and burn it down. Talk about killing innovation!
- vroom101, on 05/28/2008, -0/+5I submitted the printer-friendly link, so that maybe a problem. Try reading the article from the main link: http://www.eetimes.com/showArticle.jhtml?articleID ... (www.eetimes.com/showArticle.jhtml?articleID=207403521)
- HonoredMule, on 05/28/2008, -1/+6Modern scientific attitudes seem to dismiss the notion that we can still be fundamentally wrong about something...EVEN if we have put the incorrect understanding to good, productive use or manage to fit it into surrounding scientific theory.
This IS big, and I eagerly anticipate the new innovations that this may realize. But at the same time I love seeing the scientific community humbled back to a point where they may actually be open to these discoveries again. - HonoredMule, on 05/28/2008, -0/+5Thank you to the poster for linking the printable version...it's much appreciated.
- trevorh, on 05/28/2008, -0/+4HP can certainly patent the techniques they used to produce the memristor.
- Yez70, on 05/28/2008, -3/+7God forbid someone else may not have seen this story a month ago, huh?
- EricAnderton, on 05/28/2008, -0/+4But what I want to know is: is the memristor a "read once" technology when it comes to using it to remember something?
I'm not seeing the relationship between past voltage applied and the resistive state of the component as a useful property. Unless there's some kind of resistance change curve that can be sensed at low voltage, or a threshold at which the memristor doesn't change, how do you "read" it without changing its value? - Sapulator, on 05/28/2008, -0/+4I question the assumption that it is the fourth fundamental circuit element. First of all, it appears that in practice, it is not a linear device as are the RLC components. Also, (while I haven't done my research), I wonder how it fits into the S-plane (frequency domain)? Creating the pythagorean triangle, a resistor represents the horizontal (real) impedance, an inductor is the positive J reactance, and the capacitor is the negative J reactance. Where is there room for a fourth element? Is a memristor's impedance considered part real and part imaginary? It doesn't make sense to me. It would be greatly appreciated if someone could link me to a site explaining this phenomenon.
- sfazzio, on 05/28/2008, -0/+4While I agree with your sentiment, I don't quite think this is a prime example of that. This is more of a case of: "according to the math and theoretical physics, it'd be really useful if we could make a device with these characteristics."
bose-einstein condensate is a much better example. - vroom101, on 05/28/2008, -0/+4@EricAnderton I think, after a bit of re-reading and reflection, the memristor appears to be a read-multiple-times device.
And now I'll defer to the experts for instruction and correction... - dsmx, on 05/28/2008, -2/+6Short answer Yes, Long answer it will take 10-15 years for it to be cheap enough to use in everyday applications.
- EricAnderton, on 05/28/2008, -0/+4That's the beauty of this. It's all 100% based on existing chip-fabrication technology.
Just the shape of the circuits and atomic composition of the parts involved are the only things that allow them to take advantage of this previously eschewed "side effect". So the only real R&D they have is perfecting the design and coming up with ways to use it practically. There is no other hard or expensive part left, as far as I can tell. - drgreenberg, on 05/28/2008, -1/+5One ought to read the "Controversies" section here: http://en.wikipedia.org/wiki/Memristor
The framework in which this device is being placed as well as its actual significance isn't as clear cut as the articles from HP and Chua (so far, the main sources of promotion and enthusiasm for the device) have implied. While the device may prove to be important, the math that implies that the device is the fourth missing element may actually just be a distraction, with the real value simply being that a measurable parameter of the device ... the resistance ... can be changed in a nonvolatile manner. Well, it may also be possible to take, say, a capacitor and construct it so that the capacitance can be changed in a nonvolatile manner. If so, it'd have similar hysteretic (memory-storing) properties despite not being one of the "old" exisiting trio of 2-terminal passive devices. In the end, the importance of this device may come down to its size and scalability and not to any new fundamental principle. -
Show 51 - 100 of 149 discussions
The Digg Toolbar for Firefox lets you Digg, submit content, and keep track of Digg even when you're not on the Digg site. Download the official 
© Digg Inc. 2009
— Content created and posted by Digg users is