Slashdot Mirror
Memristor — 4th Basic Element of Circuits
esocid writes "Researchers at HP Labs have solved a decades-old mystery by proving the existence of a fourth basic element in integrated circuits that could make it possible to develop computers that turn on and off like an electric light. The memristor — short for memory resistor — could make it possible to develop far more energy-efficient computing systems with memories that retain information even after the power is off, so there's no wait for the system to boot up after turning the computer on. It may even be possible to create systems with some of the pattern-matching abilities of the human brain. Leon Chua, a distinguished faculty member at the University of California at Berkeley, initially theorized about and named the element in an academic paper published 37 years ago. Chua argued that the memristor was the fourth fundamental circuit element, along with the resistor, capacitor and inductor, and that it had properties that could not be duplicated by any combination of the other three elements."
but when will we see products that use this?
One more thing to wipe after surfing porn.
Seriously, what the fuck?
My kingdom for some mod points!
Oh my... I think i actually went blind as soon as that hit the screen.
Does it run Linux?
Does it blend?
Does it blend WHILE running Linux?
This would be a laugh. No way for windows users to reboot - the os would be effectively crippled in a few days. lol
As far as I know, human brains don't retain much information when the power is turned off and there's usually some trouble after the power is restored. Furthermore, I'm not sure how power-down information retention relates to pattern-matching abilities.
But what to I know, I had my brain off last night.
It must have been something you assimilated. . . .
If this was theorized 37 years ago, shouldn't we learn about it in physics?
Basically you have Ohm's law which is v =Ri. There is a component for each variable: Capacitors for voltage, inductors for current, resistors for resistance. It is all there, in nice little differential equations.
Yes, this is a great discovery. But please stop with the sensationalist headlines. This is getting out of hand.
This sounds like a great advance. However, computers that don't need to be booted or that boot instantaneously is not new.
TRS-80 Color Computer, for example boots instantly since it runs from ROM unless you are using OS-9.
The TRS-80 Model 100 keeps its file system in RAM and has a separate NiCD to backup the RAM. It boots instantly. The backup lasts months in my experience (even today with old NiCd's).
And any computer can simply be left on... no boot time.
So there's nothing here that cannot be done with a mixture of existing tech, except use less overall power when doing it.
Based on the comments above me I'm not the only one who picked up less than nothing from the article...What is it and how does it work? This is less detailed than a Star Trek Particle description.
Samsung took back my unlocked bootloader because Google wants me to rent movies. They're both evil.
Does this mean I will eventually not be able to use "try rebooting the system" (to try and solve a problem)? In all seriousness, I think this will make us rethink our problem-solving approach. Powering-off is a great way to "wipe the slate clean" as it were.
Would this also inspire new forms of malware/viruses/trojans that persist in memory even after the system is powered off?
And another thing... forensic computer evidence. I guess you could sort of tell what the user was last doing before they turned off the machine.
Depends on how all of this is going to be implemented anyways. Still, some interesting points to consider.
Vivin Suresh Paliath
http://vivin.net
I like
An engineer, a physicist and a mathematician are staying in a hotel.
The engineer wakes up and smells smoke. He goes out into the hallway and sees a fire, so he fills a trash can from his room with water and douses the fire. He goes back to bed.
Later, the physicist wakes up and smells smoke. He opens his door and sees a fire in the hallway. He walks down the hall to a fire hose and after calculating the flame velocity, distance, water pressure, trajectory, etc. extinguishes the fire with the minimum amount of water and energy needed.
Later, the mathematician wakes up and smells smoke. He goes to the hall, sees the fire and then the fire hose. He thinks for a moment and then exclaims, "Ah, a solution exists!" and then goes back to bed.
Hey Taco! Looks like you're using the "infinite monkeys and typewriters" scheme to generate Ask Slashdots again...
I don't understand what makes it a "fundamental" part of a circuit, while say a diode or MOSFET isn't. You can't make a transistor out of resistors, capacitors, and inductors... That's why it always showed up as the magical "voltage-controlled current source" in entry-level circuit analysis courses. I thought the three classic "basic" elements were because they were just the simplest.
Or maybe they're "basic" because every circuit (that's not superconducting), whether or not it contains semiconductors or more exotic stuff, has some amount resistance, capacitance, and inductance. Even if you don't want it, in which case you call it "parasitic". I don't think you're going to accidentally create two separate layers of titanium oxide.
So while I get why this discovery is totally awesome, I don't get what they mean by "fourth fundamental circuit element". Anyone got the skinny?
The enemies of Democracy are
I've just invented the memristor!
Another grand name from the creator of the finglonger.
Klingon programs don't timeshare, they battle for supremacy.
Repeat after me: the researchers constructed a membristor.
Somehow, I don't think these scientists really care about the abstract existence of memristors. Moreover, you can't prove the existence of something that didn't exist before you started. You might be able to proved the feasibility of such devices, but only in mathematics it may be appropriate to say you "proved the existence" of something when you actually have a construction.and then we'll have Leeloo and her multi-pass! Totally cannot wait...
So there you have it - the four fundamental elements are now earth, air, fire, and memristor. We never really wanted water in our computers anyway, so it's good to eliminate it (and don't even think about water cooling your systems - that's sacrilege).
For every post, there is an equal and opposite re-post.
http://technology.newscientist.com/article/dn13812-engineers-find-missing-link-of-electronics.html
This is very interesting stuff. I wonder if these will ever be produced for amateur use, or if they'll only ever find their way into DRAM and such..
- doctea
The possible uses outlined in the article inspire the imagination, but for my money, a technology that remembers everything presents a privacy risk too extreme to contemplate.
http://en.wikipedia.org/wiki/Memristor
You'd think the article would link to it.
How we know is more important than what we know.
Kill yourself freak.
Capacitors don't provide voltage, they resist a change in it. Ditto for inductors & amperage. Although I'll agree it's not the fourth element - it would be the fifth. A NP junction in its various forms would be the fourth.
Have you been touched by his noodly appendage?
Didn't we have the same thing in the 50's and 60's with core
memory? Manufacturers were known to even do the IPL (initial
program load), then power the machine off, crate it, ship it, and
at the receiving end, you'd turn the power on and it was ready
to run right away -- already IPL'd. (Yes, the cores would often
not shift during cross country shipping).
Great an cool tech that will windows suck more as it will stay in ram and not fully unload it self when you shut down.
http://www.eetimes.com/news/latest/showArticle.jhtml?articleID=207403521
- doctea
As for the others, they are components. For instance, a resistor R fits in dv = Rdi. A capacitor C fits in as dq = Cdv. An inductor as dphi = Ldi, and a memristor fills in the missing dphi = Mdq.
According to their description, it's a memory cell that retains information without needing power. In other words, non-volatile RAM. It actually (according to their description) looks pretty darn similar to FeRAM.
Of course, to "Read" this stuff, you have to pass current through it to measure its resistance... so then you have to refresh it back to state... not sure how serious the power savings can be compared to simply improving the power required for other memory devices there.
In fact, the whole article is a bit misleading. They've come up with a new "memory element", just like how FeRAM and MRAM and now PRAM are new memory elements jockeying for position, and each of those comes with the same pie-in-the-sky pronouncements: computers that retain state when turned off without needing to cache to disk, lower power consumption/resiliency, neural networks, blabbity blah blah blah.
Some of the real jokes come later - pattern recognition, facial recognition, etc. Those come from either improving your software, or actually making a non-binary machine so that it's easier to express multiple states, not from just having a new way to store the same data set.
In a way, the stuff in this article reminds me a lot of people eating with their butts.
What a lame fifth element. There's not even a Leeloo.
Check out my sci-fi/humor trilogy at PatriotsBooks.
PROM's, EPROM's, EEPROM's, FLASH, etc., have been around for a very long time. Is this "memresistor" different/better because it's denser and cheaper to manufacturer? Unless I missed it, the article never cites any advantage over existing non-volatile memory technologies.
9/11 Eyewitnesses to Explosive WTC Demolition 1 of 2
got the wrong episode.
We've had Non-volatile state storage for ages (eg. FeRAM and floating gates (as used in flash) and battery backed up RAM). State storage is only part of the picture.
Whatever the mechanism, freezing state is not sufficient to instantly boot a modern computer. Pretty much all modern computers have some communication with an external device that needs to be renegotiated and reconnected, be that a mouse, disk or network.
Engineering is the art of compromise.
+5 real information
No, think linux suck less more mac os X as computer turned on all day night even more.
suggests that it's something like an "analog" digital potentiometer. There was some experimentation for a while (and I guess still is) using analog computers and circuits that work like this.
The first computer I ever built was simply 3 potentiometers and an analog milliammeter (circa 1969). Setting two of the pots to particular positions on a dial would cause the meter to move up. The third pot was turned until the meter went back to zero. The dial on that pot then showed the "answer". It could be used to do simple multiplication and division.
I suppose those pots did in fact hold a sort of "memory". The memristor would probably be most useful for sensors and the type of "computation" I mentioned above.
pardon my ignorance in digital electronics - I thought capacitors "store" charge for a while with some decay rate, can this be comparable to the function of "memory"?
Sweet gravy no. I'm having a hard enough time learning inductors/capacitors. Good thing this kind of stuff does hit university for a good 15+ years.
Here: CNet Writeup
Discussion of why a memristor is new, and more about how it works.
Looking for a Rails developer in Chapel Hill?
The caption makes it sound to me like...well... a MOSFET. but that's just cuz the caption was written poorly.
The theory makes it sound to me like.... well... a Flash memory cell.
So what's the big deal?
Anyone got the skinny?
Yes but please don't tell anyone. I'm having a hard enough time trying to get girls to like me as it is.
Well I could blab about it all I wanted and there's still little chance of a girl finding out, so I think you're safe.
The enemies of Democracy are
The analogy with Aristotle's Law of Motion and Newton's Second Law against circuit design theory (basically, charge:flux :: force:acceleration is correct, and charge:voltage :: force:velocity is wrong) is the most interesting thing in the EE Times article.
I'd love to hear comments about that.
i'd hit it so hard, if you pulled me out you'd be the king of britain [bash.org]
If they mean passive components that function with voltage other than input signals, or devices with only two connections, shouldn't they count diodes? And what integrated circuit actually uses parasitic inductance (that from the bonding wires connecting the chip)? The only semi-real inductance-for-a-purpose I've seen in a monolithic chip isn't available without the chip being powered, and is essentially synthesized by use of capacitance in the feedback loop of an inverting amplifier.
And if we're cheating and using power to simulate inductors, haven't we already got memistors of sorts? Things like simple latching flip-flop behavior, or memory on the capacitance of stored gate charge? Or something that has memory with power off, as in flash-memory.
If they want to invent a new basic device with memory I've got one. Call it a fliptode.
It's like a diode, but by applying a pulse above a specified voltage in the non-conducting direction it flips the anode and the cathode around. I suppose a third programming pin could be used instead.
Trivia bits - A long time ago Apple used various combinations of grounding three connector pins to identify the operating resolution of a monitor. Eventually they needed more combinations. They got more, and maintained backwards compatibility, by allowing diodes between the pins, with the polarity choices (diode direction) also adding combinations.
Isn't a fuse a write once memistor? That's more or less what field-programmmable ROMs contained.
For more fun do something with a chip full of tunnel diodes!
If this is all too complicated, bring back core memory
This one took quite a bit of thinking, although this wikipedia article summarizes it best.
A transistor may be approximated as a variable current source. Similarly, many applications of transistors are as "active" devices, which supply external power to the circuit being considered.
A diode is effectively nothing more than a voltage-controlled switch. In a DC circuit, it simply passes current through (with a small voltage drop that can be approximated by an inline negative voltage source).
Likewise, all transistors can be abstractly considered as networks of diodes. This is why they are inherently binary devices, and why computers "think" in binary.
The classical circuit elements (Resistor, Capacitor, Inductor) each fundamentally affect the electromagnetic properties of the electrons flowing through said circuit.
Resistors impede the flow of current; a capacitor is a current "bucket" that also blocks DC signals in AC circuits; and an inductor builds up a sort of inertia for the flow of current, through the creation of a magnetic field.
The distinction is hazy, but I think I can see it where it comes from.... when a diode/transistor does something, it affects of the "layout" of the circuit, rather than directly affecting the electrons flowing through it.
The memristor is extremely interesting, as it blurs the line between analogue components and solid-state devices, and provides exciting possibilities for the development of analogue computing and data storage.
Even more exciting is that they can already be made smaller than transistors, and two can be combined to create a device that functions analogous to a transistor.
Considering that we're quickly approaching the limits of Silicon-based technology, this invention may very well offer the key to the true "next generation" of electronic devices, and may very well be as significant to our generation as the transistor was to the previous. This is Nobel Prize-worthy stuff we're talking about.
Kudos to HP for supporting "true" R&D. They most definitely will be reaping the benefits of this one for years to come.
-- If you try to fail and succeed, which have you done? - Uli's moose
Resistance of metals can increase with the temperature of the metal (see http://en.wikipedia.org/wiki/Resistivity). With that, you could try this at home. Take a length of wire, and thermally insulate it. Put a lot of current through it to heat it up. Wait. Read the resistance of the wire later on. The resistance of the wire now is slightly higher than it was (resistivity is rather small, and you would probably only see a tenth of an ohm difference). A higher resistance means a "one" was previously written to it. A lower, "normal", resistance means it was a zero. So, to write a one, dump lots of current into it. To write a zero, let it cool down.
Of course, this is neither fast or power efficient, but it works to some degree (pun!).
Now, if I'm correct:
A) This requires current to flow through the first wire, so where's the memory?
B) Aside from the probably-more-linear relationship, how is this different from JFET's or BJT's?
I mean, the transistor is a device where you run a current or voltage to point A, changing the resistance between points B and C. Can someone explain the difference?
>> Standing on head makes smile of frown, but rest of face also upside down.
Diodes and all types of transistors work on the same principle. If this doesn't, then it's something new. If it is, then it's just exploiting formerly misunderstood properties of a transistor. Resistors and capacitors, of course, are completely different from transistors; resistors are just harder to push electrons through and capacitors store them. They're fundamentally different beasts in that we are interested in totally different properties of each (although a transistor certainly has both resistance and capacitance, and a capacitor has resistance, and a resistance ostensibly has some capacitance... everything else seems to.)
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
The linked article, and even the Wikipedia page about it, are leaving me very puzzled.
The fundamental circuit elements are all "fundemental" because they have very simple mathematic functions that describe their behaviour.
For resitors, it's just Ohm's law: V = I*R
For capacitors and inductors, there are integrals involved (which, for you paranoiacs, means they "remember" the past, just like this memristor thing).
Capacitor: V = 1/Capacitance * Integral( I )
Inductor: I = 1/Inductance * Integral( V )
Inductors can be seen as the "dual" of capacitors (just swap voltages and currents in the equations.) Resistance can be seen as having a "dual" of conductance (again, swap voltages and currents).
So, where does this "memistor" fit in? What is its mathematical function? Does it have a dual? Perhaps a "forgetistor"?
There's a difference between the memristor and flip-flops and any other device that mimics it. This is one device, not made up of transistors or capacitors, simplifying a circuit considerably. Also it scales beautifully to the nanometer size, allowing for smaller, simpler fast memory without need of capacitors.
I think what the fundamental elements have in common is that they have a linear transfer function, whereas transistors and diodes are non-linear.
I am not a crackpot.
It seems to me from reading a few articles, that this is a non-linear device. For the record I'm just finishing EE junior year. It seems the applications for these devices are like flip-flops (a la the memory talk) and to act like axons. Using analog analysis on flip flops shows that they are non-linear. (I thought through a few different logics, CMOS, TTL, and diode logic). Axons, as far as I can tell from the internet and Carver Mead's book: Analog VLSI and Neural Systems, are non-linear as well. Why is linearity important? Well resistors, inductors and capacitors are linear. Diodes are non-linear and aren't considered basic circuit elements. Linearity, by the way, means that if I input signals x1 and then x2 into a circuit, I get y1 and then y2 as outputs; then I put in a third signal, A*x1 + B*x2, the output is A*y1 + B*y2. Though honestly, without specific equations for the device, I couldn't be sure.
I don't get what they mean by "fourth fundamental circuit element"
There are four fundamental circuit variables; current, voltage, charge, and flux.
We can define the relationships between charge and current and between flux and voltage. (charge as an integral of current, flux as an integral of voltage over time)
A resistor provides a function to relate voltage and current.
A capacitor provides a function to relate charge and voltage.
An inductor provides a function to relate flux and current.
Until now we did not know how to construct a passive device which would provide a function relating charge and flux. The only remaining combination of these fundamental variables.
That's a good explanation that ties together what I already knew about RLC circuits with the piece that was missing, thank you.
The enemies of Democracy are
A memristor is a passive, two port element that "remembers" how much current has passed through it. That is, its resistance depends on how much current has previously passed through it.
Artificial memristors have been fabricated in labs using active elements (that require a power source, just like a transistor), to demonstrate its operation and potential application.
TFA is interesting, because this is the first time a real memristor has been demonstrated. i.e. it uses no active elements and requires no external power to actually behave like a memristor.
Does the computer stay locked up when we reboot?
Well, if I had to guess I would say they are commenting more on functionality than form of the basic devices. Using capacitors, inductors, and resistors sealed in a vacum tube will give you the voltage following voltage source (tube amplifier), as well as a rectifier (wikipedia has a nice illustration of a tube rectifier at http://en.wikipedia.org/wiki/Rectifier).
I grant you semiconductor devices are much better at those functions, hence why we use them, but their abilities weren't without precedent when they were created.
As far as the article goes I would take it with a grain of salt. This looks very familiar to the molecular memory from a few years back (I think also reported by HP labs, or possibly IBM), where it turned out instead of storing a charge in a molecule the researchers were just spot welding the connection leads together by putting too big a current accross the leads (sort of like flash memory, only it wore out much faster).
Punching a hole in the titanium dioxide layers with an overvolt would definatly change the resistance, the question is can you change it back, and how many times can you repeat the cycle with the same device?
Not quite correct. A transistor does directly affect the electrons that flow through it. I'm particularly thinking of MOSFETs, which I work with. There's a gate that directly affects how much current will flow through the channel.
I think the distinction has to do with linearizing the circuit. When you decompose a transistor into a voltage controlled current source, you get a linear element. Rs, Ls, and Cs are already linear. But I don't get why the memristor should be considered a new fundamental circuit element. It doesn't sound linear. Actually it just sounds like a current controlled resistor. I'm sure it's quite useful, but I don't see why you can't break it into linear circuit components.
The ultimate goal of science is to unify all forces of nature to a single law that can be silk-screened onto a T-shirt.
The difference between a memristor and FeRAM is that because the memristor is constructed without using any transistors, it can be used as a kind of analog memory. Instead of just storing 1's and 0's, it's resistance is an analog value anywhere in the range of on and off. Of course you can still use it to store digital data, but the real fun will come when you interconnect these things to emulate the analog behavior of the brain. This is where the claim of pattern recognition and facial recognition come in. They're not actually talking about software there but the actual analog capabilities of circuitry built with memristors.
The other amazing thing about memristors is how small they are. The articles state that you can emulate a transistor by connecting a few memristors, and that transistor is smaller than any we have today. Also it states that the memristor actually performs better at smaller sizes. This really is neat stuff.
For some reason the my comment where I actually explain what's going on is not modded up to where people will see it. A memristor M satisfies the equation dphi = Mdq. It is one of the six differential equations that can be written that way taking two of the four fundamental circuit variables current, voltage, charge, and magnetix flux. The other five were well described and embodied in electrical elements: this sixth one finishes the symmetry.
Thank you mo. Scrolling down, I saw the usual arrogant, "I'm smarter than any dude in a lab" type posts explaining that this memristor was just a memory cell and what was the big deal? Hey Slashdot geeks, you may be smart but so are a lot of other people. Knee-jerk cynicism is a pointless waste of time, okay?
Your post on the other hand actually explains the difference and deserves an 'informative.'
- None can love freedom heartily, but good men; the rest love not freedom, but license. -- John Milton
"Linear" in what sense? True: you (essentially) get out the same frequencies you put into a system made up of these elements, and there is a (roughly) linear relationship between voltage across and current through these individual components. I'm a bit hesitant to really term them "linear", though, because for a nontrivial definition of "linear", any combinations of them should also be linear, and that's definitely not true: they shift around the poles and singularities of the transfer functions, and the amplitudes and phases of what you get out are definitely affected. Once a resistor is thrown into the mix, the energy of the system (at least that stored in E and B fields) is no longer conserved, and the differential equations become inherently nonlinear.
It doesn't get any more succinct than that.
I think you were on to it with your second point about it being basic because every circuit has some amount of each fundamental element. I've actually seen this a bit in resistance measurement stuff where bipolar excitation improves the accuracy of the measurement due to this effect.
The abstract of the original paper from ieee.org:
"A new two-terminal circuit element-called the memristor characterized by a relationship between the charge q(t)equiv int_{-infty}^{t} i(tau) d tau and the flux-linkage varphi(t)equiv int_{- infty}^{t} v(tau) d tau is introduced as the fourth basic circuit element. An electromagnetic field interpretation of this relationship in terms of a quasi-static expansion of Maxwell's equations is presented. Many circuit-theoretic properties of memistors are derived. It is shown that this element exhibits some peculiar behavior different from that exhibited by resistors, inductors, or capacitors. These properties lead to a number of unique applications which cannot be realized with RLC networks alone. Although a physical memristor device without internal power supply has not yet been discovered, operational laboratory models have been built with the help of active circuits. Experimental results are presented to demonstrate the properties and potential applications of memristors."
The other piece of this, it occurs to me, is that while diodes and transistors and other semiconductor devices do provide new ways for circuits to react, they are nonlinear whereas the memristor is a linear device, just like a capacitor, inductor, or resistor.
The three elements we're used to (R, L and C) relate four things, potential, current, flux and charge. R relates current and potential, L current and flux and C potential and charge. The thing that relates flux to charge is this newfangled (compared to the other three) thing called a memristor. The other two relations (potential/flux and current/charge) are fundamental conservation laws.
At least that's what my quick Googling on the subject turned up.
Probably not modded up because your comments contain the phrase "differential equations" ;)
The real "Libtards" are the Libertarians!
Imagine the possibilities once our computers can remember information! Oh, wait...
Memory that could survive power off is old hat, both with various forms of ROM as well as RAM (core memory for example). What makes this new component different?
So what you're saying is that it's sorta like a capacitor, but instead of voltage, its function operates on flux.
How many gigawatts can it handle?
Dmitri B. Strukov1, Gregory S. Snider1, Duncan R. Stewart1 & R. Stanley Williams1
Anyone who ever took an electronics laboratory class will be familiar with the fundamental passive circuit elements: the resistor, the capacitor and the inductor. However, in 1971 Leon Chua reasoned from symmetry arguments that there should be a fourth fundamental element, which he called a memristor (short for memory resistor)1. Although he showed that such an element has many interesting and valuable circuit properties, until now no one has presented either a useful physical model or an example of a memristor. Here we show, using a simple analytical example, that memristance arises naturally in nanoscale systems in which solid-state electronic and ionic transport are coupled under an external bias voltage. These results serve as the foundation for understanding a wide range of hysteretic current-voltage behaviour observed in many nanoscale electronic devices2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19 that involve the motion of charged atomic or molecular species, in particular certain titanium dioxide cross-point switches20, 21, 22. Here's the final paragraph of the research paper, which discusses some of the implications: The rich hysteretic i-v characteristics detected in many thin-film, two-terminal devices can now be understood as memristive behaviour defined by coupled equations of motion: some for (ionized) atomic degrees of freedom that define the internal state of the device, and others for the electronic transport. This behaviour is increasingly relevant as the active region in many electronic devices continues to shrink to a width of only a few nanometres, so even a low applied voltage corresponds to a large electric field that can cause charged species to move. Such dopant or impurity motion through the active region can produce dramatic changes in the device resistance. Including memristors and memristive systems in integrated circuits has the potential to significantly extend circuit functionality as long as the dynamical nature of such devices is understood and properly used. Important applications include ultradense, semi-non-volatile memories and learning networks that require a synapse-like function. There's also a Nature News and Views, but I think that might also need a subscription.
Why can't we still undo mods without posting comments?
My spoon is too big.
Um, flux and voltage?
Current and charge?
There are some who call me
How is this different from the Warburg Impedance that shows up in electrochemical cells?
Resistors I see affecting a circuit immediately. They do make changes, but there is no delay and it's a sort of 1:1 relationship. They "do not change".
Inductors and capacitors affect a circuit over time. As they are affected more, they produce more effect, and vice versa. They have a sort of memory, but it's usually short term and is the sort of thing that is "used" by the circuit immediately. They "change after awhile".
Transistors and diodes are different in that they cause distinctly different behaviors based on the immediate input. They "change now".
This memsistor seems to have yet a different response. It serves as a circuit memory somewhat like a capacitor, but is specifically designed to affect a circuit at a much later time. They "change later".
I think this is what they are trying to get at.
I work for the Department of Redundancy Department.
Prof. Bernard Widrow developed a "memistor" in 1960 which was used to build early neural networks. It was a bit larger and a bit slower than what's described here, but did essentially the same thing.
In the early neural networks it was used to adjust the gains associated with the various inputs.
Breaking Into the Industry - A development log about starting a game studio.
I wish that hadn't been posted by an anonymous coward. I want to add them as a friend so that their future posts are more visible to me.
Linear as in "not discontinuous". Diodes and transistors have knees, gaps and other discontinuities in their voltage / current functions.
exp(-k*t) is a linear solution, isnt it...? why do you say that adding a resistor to a circuit makes it nonlinear? an RLC circuit has a pseudoperiodic solution, cos(w*t)*exp(-k*t)... are you saying this is not linear?
erm....
What about voltage and flux?
Theres 4*3=12 combinations altogether. Not 4!
OK, I didn't RTFA, but this is slashdot right? Aren't the gates (and channels) of current (no pun intended) transistors only several atoms wide? So several "memristors" can be created in the space of several atoms? I guess leakage current places limitations on how "large" of a memristor value can be created?
The other doodads you talk about, like transistors, are classified as active elements.
It would have been good if the editors would have inserted a clarifying "[passive -eds]" to the submission.
An honest question, here...
I thought that only moving charges, i.e. current, could produce magnetic flux. Is there another method of magnetic flux production going on, here? If it is still flux caused by moving charges, how is it different than the inductor?
>> Standing on head makes smile of frown, but rest of face also upside down.
I suppose 36 years ago "memristor" might have seemed cool. But it just doesn't feel right. Even "memistor" is better than "memristor". For that matter, so is "resimory".
How about "flasistor" or "resistash"? I know! "Storistor"!! Yeah, that's the ticket!
I don't know if I'm too late to get this modded to any level where someone will actually read it (and possibly answer), but does this mean that you feel such a device should be patented by HP? There is always talk here about just "taking the next step" technologically. This is a great invention, but it was also theorized/predicted 30+ years ago. Does that mean it's 'just the next step' or does it actually meet the 'non-obvious'ness requirement?
You're absolutely right. I was applying homogeneity to be a necessary condition on the solution, which implies that f(t)==exp(-k*t) ==> f(a*t) = exp(-k*a*t) != a*f(t).
That would explain why she could remember everything right from being cloned, though.
Have you been touched by his noodly appendage?
Until now we did not know how to construct a passive device which would provide a function relating charge and flux.
Uh, in 1985, the flux capacitor was perfected by Dr. Emmett Brown. It requires a mere 1.21 jiggawatts to bias.
Research shows that 67% of those who use the term "research shows", are just making shit up.
So shouldn't this technically be called a flux capacitor?
I wonder if they'll make some big enough to use on a regular circuit board with a soldering iron. I'd love seeing what kind of cool circuits could be made with these.
The mempacitor & memductor can do it better... and they're only 5-10 years from production. :)
From the grandparent: We can define the relationships between charge and current and between flux and voltage.
Also you are double counting some relationships because they are reflexive.
But at least you eliminated the identities and didn't claim 16 combinations.
Not quite.
4! / (2! * (4 - 2)! ) = 6.
2 relationships are given as basic integrals, now the other 4 are described by resistors, capacitors, inductors, and memristors.
Missed that bit!
Still not clear how it works though, or what exactly it's storing. Surely resistance is a function, and it must be storing charge or flux, or some quantum effect perhaps?
There are some who call me
so, by that logic, wouldn't a better name for it be a flux capacitor?
If I had a nickel for every time I had a nickel, I'd be richcursive!
TFA is content free. Every sentence contained in it described the non-linear phenomena of Hysteresis. Of course, linear elements like resistors and caps can't replace it. Duh...
But what's new about hysteresis? Magnetic recording and core memory and re-writeable optical storage are all based on this idea.
How did this get on slashdot?
Some drink at the fountain of knowledge. Others just gargle.
And that six-transistor radio I listened to as a young boy must have been receiving digital signals in 1965.
Look what the computer age has done: analog has been forgotten, even by bright minds.
But luckily not by everyone. The brightest minds still are aware that even the fastest digital 'gates' are fundamentally analog. Actually, especially the fastest....
But repositories of common knowledge are being filled with well-intentioned, but less-than-half-baked treatises and misinformation.
Transistors are IN NO WAY networks of diodes. Yes, some (bipolars) have a p-n junction, but there is no way that diode theory explains amplification. And don't get me started in fets (junction or insulated-gate). No diodes in those (except the packaged-in protection diodes in some). The jfets can be used to rectify, but that's not their nature.
I'm assuming some CS major wrote that wiki about something they didn't understand. Please don't be offended - that's not a stab at CS majors, but it was obviously someone "web-active" and had some (but inadequate) exposure to electronics.
(Disclaimer - I didn't read this wiki, so I'm taking for granted that moosesocks has lifted it from a wiki verbatim)
Ugh.
And to the GP - there is a fundamental difference between passive and active components.
The grandparent just had it backwards: voltage is the integral of flux over time.
The only problem I can see is that even if this has the speed of digital with the range and versatility of analogue, the inherent drawbacks with analogue still exist, namely non-exact reproducibility of data, code or programs in general, and obviously loss of data over generations.
Digital is just so.... precise.
Why OpalCalc is the best Windows calc
In principle how does this differ from Magnetic Core Memory? http://en.wikipedia.org/wiki/Magnetic_core_memory It is a better way to implement it; but, how does it differ enough to be a forth to be added to Caps, Resistors, and Inductors? Tim S
It sounds like an active componont ,like a semiconductor. Or maybe just a resisor. Maybe they could make a mempacitor and memductor too. I'm glad they are excited .
A picture is worth a thousand words...
Now for the kicker - I was introduced to SPICE in EECS 105 at good ole UCB (Winter '74) and the prof for the course was none other than Leon Chua. Thoroughly enjoyed the course and ended up getting my one and only A+ at Cal.
N.B. The license that Pederson used to distribute SPICE was probably what the CSRG used as a basis for distributing BSD.
In case anyone has a nature subscription:
http://www.nature.com/nature/journal/v453/n7191/abs/nature06932.html
"I object to doing things that computers can do." -- Olin Shivers, lispers.org
A more technical description is available here.
Diodes are not fundamentally binary devices. They are more useful than passive elements for circuits dealing with discrete data, because they have non-linear behavior which has several discrete identifiable regions. Every diode has three of these (> turn-on voltage, with forward current flow), ( -Zener voltage, with negative current flow), and (between those, with negligible current flow). All diodes have a Zener breakdown voltage, so-called "Zener diodes" are diodes where the Zener voltage is carefully controlled and doesn't damage the junction permanently. Within each of these regions, and in the transitions, the behavior is still continuously variable, and it turns out that diodes and transistors are awfully useful for analog signals as well. What do you think drives a transistor radio? Or an audio amplifier?
So don't blindly trust your source.
"I can't imagine how things could get any worse!" (some guy) "That could just be failure of imaginatioÂn on your p
Galatians 5:19-21
Indirectly:
Matthew 5:17-19
Note; the law's still there, but the parole board takes true remorse into huge consideration.
Ugh. You're right.
I'm a physicist who seriously needs to catch up on sleep and brush up on electronic theory.
I completely ignored FETs in that vomitous output of incoherent thoughts.
NPNs and PNPs still sort of fit my original description though....
Mods --- feel free to send my original post into oblivion. It's flat-out wrong.
-- If you try to fail and succeed, which have you done? - Uli's moose
Anyway, the concept is neat. But one problem I had is that the paper says that the memristance (M) needs to change with charge (or flux) to be interesting. But the other fundamental elements (R = dv/di, C = dq/dv, and L = dphi/di) all have basic components that assume it's fixed. Circuits using resistors, capacitors, and inductors all assume the R, C, and L are fixed, except in special cases where it is explicitly says otherwise. So why does a memristor get to have a changing M value? Supposedly because if it didn't it would just look like a resistor and would be a boring element. But I think there must be more to it. But so far I haven't figured it out.
Thanks for your explanation.
The ultimate goal of science is to unify all forces of nature to a single law that can be silk-screened onto a T-shirt.
Fourth Basic Element? And then some...
Fourth Horsemen of the Apocalypse may be more like it. Mankind vs Machine coming sooner than we all think and this discovery just might make AI that much more possible and powerful in our lifetime... okay, maybe I'm getting a bit carried away, sure I am... If that helps you sleep better at night, more power to you.
Imagine an electronic circuit 1000 times more powerful than our own brain, comes to maturity(programmed) in minutes, 100% resilient and lasts for centuries.
Mankind invented the word consciousness; I wonder what word the Machine will use...
Run Sarah!!! RUN!!!!!!!
It's not the some of the functions of a memristor can't be duplicated with a combination of the other 3. It is that to do it with the other 3 would require a much larger circuit to be designed whereas the memristor would be a simpler, smaller circuit. Smaller circuit = greater density.
Well, there's spam egg sausage and spam, that's not got much spam in it.
>Kudos to HP for supporting "true" R&D. They most definitely will be reaping the benefits of this one for years to come.
*Maybe*. Having a working device is only one part of the game, being able to mass produce it cheaply is the other part, there's plenty of cool gizmos which we aren't able to mass produce so they go nowhere..
why do I keep on thinking this is the name of something in Futurama?
Compare Chua's "Linear and Nonlinear Circuits" to any undergrad circuit theory textbook today. The difference is absolutely ridiculous.
He's also the inventor of "Chua's circuit", a neat little thing that does chaos-theory things.
When I was at Berkeley, Chua was teaching some course about nonlinear networks and neural processing. The description sounded interesting, if a little out there.
Seconded. Analogue memory, and it's tiny (therefore cheap) already. If it's as good as it looks then in the short term it will "just" mean a continuation of die shrinks and new types of memory, but the longer term applications could be revolutionary, the obvious one being neural nets.
I was under the impression HP had pretty much gutted their R+D arms in favour of just selling printers/consumables these days, it's good to know they're still doing fundamental stuff like this.
Every week there is news of fabulous new technology that will make RAM cheaper, faster and more plentiful.
I suggest we make a tag - "newram" - so that we can search out a list and poke finger and laugh at it later.
Unfortunately, I am too stupid to figure out this tagging thing. Will someone else rise to the occasion?
Galatians - ditto.
Matthew 5:17-19 - give me a break - it's a huge step to claim that not destroying 'the law' amounts to any condemnation of homosexuality whatsoever.
Of course it is, but you won't get the funding to continue development if you don't make a whole lot of might and could style predictions.
It's hype - resistors, capacitors and inductors are known as *linear* components because any circuit you can make from them has outputs which are linear functions of the inputs. memristors (what a terrible name) share this property when not doing anything (i.e. just storing constant values and not actually being *used* as memories) but, just like the diodes and MOSFETs you mention, are not linear devices and so are not at all fundamental in the same way as R, L and C.
:)
For what it's worth, Stan Williams is something of a hype artist. (Think 'IBM R&D press release department' here) -
see for example "HP Labs Scientists are Building Computer Chips in a Whole New Way" from 1999:
http://www.hpl.hp.com/news/molecules_that_compute.html
it was more or less bullshit then - reasonable enough at some simple level, but not at all justifying the hype. These memristors will likely turn out just the same - I'd wager you a beer they'll be all but forgotten 5 yearas from now.
Yes, I remeber core memory and the poor buggers who made it by hand.
Do you remeber what the John Von Neumann / Vannevar Bush debate was about?
"TFA is content free."
Guess you missed the reference to the fact that Nature has published HP's claim of being first to demonstrate such a device. Maybe you could have beat them to it with nothing more than a sewing kit.
And did you exchange a walk on part in the war for a lead role in a cage? - Pink Floyd.
(Someone had to say it..)
A really evil root kit could conceivably find unused EEPROM in your video card and write itself there. When your system boots, the video card itself provides the hooks and starts the rootkit BEFORE the OS starts. It is the proverbial red pill/blue pill scenario.
However, this has never been seen in the wild, it would definitely be video card specific, and would have a very limited infection rate. So we are actually pretty safe.
- I live the greatest adventure anyone could possibly desire. - Tosk the Hunted
I don't like the name "memristor". It's too close to one of the other three fundamental names. Of course, "flux capacitory" comes to mind, but it also uses one of the other names in its name. So I propose calling it a "fluxor". What do y'all think? (of course, they're used on the starship Outreach :-) )
What this discovery does is completes the set of possible passive devices; we've found them all now, there will be no more. Everything else you're talking about are active devices, which I would argue cannot be reduced to a fundamental set.
To contrast the two, let's think about how we would define the fundamental set of semiconducting active devices. The simplest is a diode: it's just a PN junction. Everything else is made of different arrangements of P and N material, plus insulators. A bipolar transistor is a PNP or NPN arrangement, and there are two ways to run one "backwards" and get a diode, one using a PN path, the other NP. But you can't just solder a pair of discrete diodes together and get a bipolar transistor; the physical arrangement of the materials matters. And if arrangement of materials matters, is a FET fundamental, or is it just part of the class of devices called "transistors"? Where do UJTs, SCRs, triacs, and IGBTs fit? I say these are all variations on a theme with infinite possibilities. It's like with music: you can't draw a line around certain sounds and say "those within are musical notes, and those without are not".
By contrast, the set of passives is a fixed-size set, with each element being completely unlike any other. An ideal inductor has no capacitance and no resistance; an ideal resistor has no capacitance or inductance; and an ideal capacitor has no resistance or inductance. (Real ones have parasitic elements of the others, but there's no fundamental mathematical requirement for this.) We can draw a line around this set and say, "this is fundamental, all passive circuits are built on these elements".
Bravo, bravo! Although greatly off-topic and repulsive, I admired one who writes as vividly as you! The use of delightful (although quite disgusting), picturesque descriptions coupled with the writing stamina to keep the reader entertained, made for a fine story, indeed. So while everyone hates on you (and now me), I see the beauty behind it all, which is typical of me. Write on, you fetid writer, you.
There are six possible pairs of any four objects. So this isn't the only remaining combination. According to your post, there still aren't devices to relate voltage and flux, or charge and current.
d(Psi) = M dQ
So, Doc Brown wasn't that far off?
Perhaps they were thinking of piecewise linear modelling, where any non-linear component, including transistors, can be modelled using linear components, ideal diodes, and ideal voltage sources...
This topic is old enough and long enough. Didn't we agree that in such circumstances, Microsoft bashing should be modded informative?
I always thought the big three were the Solder, the Fun, and the Magic Smoke. (amen)
UTF-8: There and Back Again
What does this mean for computing? It seems to me that this has enormous implications all along the chain of how we think about computer design, logic circuits, and programming. Am I wrong in understanding this as a fundamental change in how things will/might work?
How is this different from an EEPROM memory cell, technology which has been around for a long time? I would submit that we already have the ability to retain memory when the power is off!
Sorry I'm late: I had to recharge my flux capacitor.
They finally invented the flux capacitor!
-- Scott Now that brings us to the topic of materials in other fields of life that we observe daily but not closely which would model future inventions in EE/CS - basically we need to do the layman's equivalent of mathematicians' relentless theoretical pursuits - extending analogies across fields of study. That would lead to mostly science-fiction, but some ideas in there would be big, really big.
Like if you were watching ketchup really closely in grad school and it struck you that the history force applied directs the speed of the flowing ketchup well into future instants of time, and you sat thinking - hmm, now that can be used in some semiconductor theory or some data structure in comp sci, or some algorithm in math/cs, or maybe explain some heating/convection at ocean-floor fault lines, or in shaping airplane wings...
A brute-force analogy system.
We could come up with something interesting in a couple of thousand tries - and if not, we publish the crap possibilities as fodder for Open-Sci-Fi stories
Sounds too far-fetched, I know, but tha's what everyone calls a math nerd who then goes on to become "immortal in history" by virtue of the same nerdity.
Hackers have long memories. It works both ways.
Actually, you can consider digital circuits as low-pass devices.
Deducing from TFA, it sounds like this works on the principle of the old magnetic core memory- you have to "tickle" it to read the data.
4th passive element. A transistor is a active device.
Resistors, capacitors, inductors, and this new memristor are all passive devices and fundamental parts of all circuits. All materials have intrinsic resistance, capacitive (Electric Field effect), inductive (Magnetic Field effect), and now the theoretic moving of charged dopants. They show up with varying degrees of importance. Diodes and the other active devices are all a result of using voltage and/or current (as well as charge and magnetic flux) to change mostly the resistive property of a material. In superconductors resistance (and likely this memsistor) is 0 which is far different from not being relevant. Now at nanometer scales, it appears that this memristor is dominant. As this property is flushed out in the near future, VLSI (Computer chip) circuit design will change. As designers take this new property into account for analysis of current circuits sizes can shrink. As designers start to apply this in novel ways some very cool circuits will be made. Hope this is helpful. sorry for errors.
Programming you say? There is a new discussion group on google - http://groups.google.com/group/memristor-computer-programming You are welcome to join. Please forward the web page to others who may be interested. Jaxs
There is a new discussion group on google - http://groups.google.com/group/memristor-computer-programming You are welcome to join. Please forward the web page to others who may be interested. Jaxs
can someone please help me with the flux capacitor proof? i can get as far as saying that M = Vdt/CdV where M = memristance, V = voltage, C = capacitance
The more interesting thing about Galatians 5:19 is that it's actually talking about sex outside of marriage. So if you have the correct religious affliation (Episcopalian), that doesn't even apply.
Have you been touched by his noodly appendage?