An anonymous reader writes "HP researchers have built the first functioning hybrid memristor-transistor chip. Lead researcher Stanley Williams and his team built the very first memristor — the '4th fundamental element' of integrated circuits after resistors, capacitors and inductors — back in April. Memristors can remember their resistance, leading to novel electronic capabilities. The new FPGA circuit uses memristors to perform tasks normally carried out by (many more) transistors and is therefore smaller, more power efficient and cheaper to make, HP says. Memristors could also turn out to be a more compact, faster alternative to flash memory."
The book The Bottomless Well [amazon.com] discusses the concept of "the refined energy pyramid" where each level is smaller and more useful than the one below it. Electricity and computer are two levels medium-high on the pyramid. The books shows how auto technology has been rising up the pyramid with increasing fractions of its energy level at more refined levels. The book says about 15% of a modern auto's energy density (excluding hybrids and plug-ins) is now electrical and increasing. Computing is growing too, replacing items like distributors, etc with more efficent computed actuators and increasing mileage. Memisters will probably more compact implement soem electronic functions the other three are used for now.
Am I the only when that thought memristors would remain the the 5-10 year category for the next couple of decades? Granted, this is just a proof of concept chip but it is moving along very rapidly compared to most 'game changing' advances.
20 years of theory and work just to make the first memristor, less than a year to use the new memristor in a device that actually improves over the standard technology. So when will we see commercially available devices? Next year some time at this rate?
It depends on whether they realize how bad the economic situation is. Right now they still think we are in a recension. The old business model of innovation to drive sales is valid in that circumstance. If they start to think we are going into a depression *cough* then they will cut off research and start fortifying existing tech. I for one hope this technology has enough momentum to carry it through.
"Williams says. Still, he predicts that memristors will arrive in commercial circuits within the next three years."
It seems fast because nobody was talking about these things for the last 30 years. It's only because of technological advances in circuit printing and general computing that we can make these things and integrate them without having to develop a lot of additional technology. The transistor is very old but only after developing a lot of supporting tech have we been able to shrink them down to fit billions in a processor. That same tech can already be applied to memristors. We don't need to wait decades before we can shrink a memristor down to practical levels for ICs.
The reason that memristors were so quick to market is that they were discovered while researching new fabrication processes. This isn't like, say, carbon nanotubes where they're able to make one or two in a lab. For memristors, they had a fabrication process from day one.
Am I the only when that thought memristors would remain the the 5-10 year category for the next couple of decades?...So when will we see commercially available devices?
Ahh, but what's the resolution of that one 'bit'? AFAIK this is an analog device, not at all limited to 1 or 0, but able to be used as such to to the handy-dandy DAC.:-D At first this is a thing for circuit-nerds, but don't despair, as when the circuit-nerds have hand their hands dirty for a minute, computer nerds are going to get to write "programs" for FPGAs, and then not long after that, libraries will emerge that let C nerds (and regular ole' C monkeys) utilize the goodness of these little variable c
Although the HP memristor is a major discovery for electrical engineering theory, it has yet to be demonstrated in operation at practical speeds and densities. Graphs in Williams' original report show switching operation at only ~1 Hz. Although the small dimension of the device seem to imply fast operation, the charge carriers move very slowly, with an ion mobility of 10E-10 cm2/(V s). In comparison, the highest known drift ionic mobilities occur in advanced superionic conductors, such as rubidium silver iodide with about 2*10E-4 cm^2/(V s) conducting silver ions at room temperature. Electrons and holes in silicon have a mobility ~1000 cm^2/(V s), a figure which is essential to the performance of transistors. However, a relatively low bias of 1 volt was used, and the plots appear to be generated by a mathematical model rather than a laboratory experiment.[8]
The application they talk about is the programmable transistors in FPGAs.
You can find FPGAs in all kinds of consumer electronics. Typically, the chip has some onboard flash from which it loads its configuration during power on. These transistors which load from flash during power on are the ones the researchers are talking about replacing with memresistors (memristors?). For this kind of application, a 1Hz write speed is completely reasonable. The write will be done once in the factory.
Who is the idiot modding this as Funny? A SPICE model is an engineering tool used to diagram circuits. It is NOT related to any of the Spice Girls!!!:p
Ultimately, the next few years could be very important for memristor research. Right now, "the biggest impediment to getting memristors in the marketplace is having [so few] people who can actually design circuits [using memristors]," Williams says. Still, he predicts that memristors will arrive in commercial circuits within the next three years.
I think you could build one from components if you had data about the memristance function. Start with a voltage controlled resistance element (Gxxx) connected to the two exposed terminals. Add a hidden element of a current controlled current source, sensing on the VCR element, injecting current into a hidden capacitor. The voltage on that capacitor is proportional to the total charge that has passed through the memristor device. The exposed VCR element senses voltage on the hidden capacitor, and uses an interpolated table of resistance vs voltage rather than a linear relationship.
That doesn't capture the hysteretic behavior of the current devices, which stop integrating at the some limiting points, but it's a starting point. You could add such behavior with a few more hidden components (back-to-back ideal zeners across the hidden cap, for example), though getting the right behavior might be a little tricky.
Binary was chosen earlier in computer work for it could represent accurately a digit representive such as 1001 equals 9. Also magnetic core memory could hold only the two states.
With memristors (once they are perfected) can have multi-state such as trinary (base 3) or decinary (base 10) eliminating all of the conversion that is neccessary in the present binary system that require cpu cycles. 123 in the decinary system represents 123 where in binary it would be 1111011 and need to be converted in order to be meaningful.
For instance I have heard for those studying DNA that using base 12 has certain benefits in directly expressing information. Perhaps this will open a whole new arena of possibility that previously could only be simulated in binary.
The mind can imagine many new possibilities if the memristor actually is.
With memristors (once they are perfected) can have multi-state such as trinary (base 3) or decinary (base 10) eliminating all of the conversion that is neccessary in the present binary system that require cpu cycles. 123 in the decinary system represents 123 where in binary it would be 1111011 and need to be converted in order to be meaningful.
Um, for the most part, the computer only has to convert from binary to decimal when it displays base 10 numbers on the screen (ie using the calculator). It's hardly computer intensive. All the operations (add, subtract, multiply, divide) are going to be in it's native binary, no conversion needed.
Computers convert data all the time - this text you are reading now is really just a series of binary numbers converted to ascii or unicode or whatever with lots of other conversions needed to throw it on the screen.
The grandparent isn't as silly as it sounds. Multi-level flash cells don't store binary data, they work, typically, in base 4. Each cell stores a value from 0-3, meaning you only need 4 cells instead of 8 for a byte. Memristor-based systems can accurately store analogue values, and so can potentially store more than one bit of data in a single element. If you could store an entire byte in a single element then you'd get much higher density, although you'd probably want to translate it back into binary when you loaded it into a register, since binary arithmetic circuits are very easy to build (try doing bitwise or in decimal...).
I can write those books and have in the past. My experience stems from 1962 being a systems Analyst for a computer system. In the old days before IC's and even before transistors were a part of computers the vacuum tube was used.
In those days the adder section, rather than being a single chip was composed of discrete components. Once an adder problem was a wire wrap on the back panel that was making poor connection causing it not to promote a "carry" from the previous position.
I really do understand binary but keep in mind that having more than the two states of binary permits a smaller size over all. For instance the 123 I mentioned needs 7 positions in binary but only 3 positions in decinary.
Watch for it. Eventually this will be the going thing. Binary locked us in and was very restrictive while this invention and others in a similar vein will present opportunities we could not imagine before.
The number of cycles needed to convert decimal to binary and back is completely trivial compared to how much easier binary makes all the actual computations.
I'm sure that they will use 2^n states and consider them as groups of bits.
I know. If she were still at HP, she would still be in production. Now that she's used HP as a springboard, she thinks she's going to be a senator or something. So before HP she was just irritating to whomever knew her. Now she gets to irritate the whole world.
What I want to know is what is the recognized symbol for this new element? I see the one on Wikipedia, but is this the international standard, sanctioned by the IEEE?
From TFA: '''In order to be so flexible, however, FPGAs are large and expensive. And once the design is done, engineers generally abandon FPGAs for leaner "application-specific integrated circuits."'''
This isn't really true. The rising fixed costs of an ASIC is prohibitive for low volume embedded projects where a $1 FPGA will do just fine. High performance FPGA chips are about the same cost as a CPU and they are commonly used as reconfigurable co-processors for supercomputing applications or embedded DSP. And I get way more GigaOps per dollar with FPGAs than with a CPU and for much less power.
In general there are two ways to store data (at least in a RAM type memory module). You can either use a capacitor (like DRAM) or a flip flop (SRAM).
The problem with a capacitor is that it looses it's charge after a certain amount of time, and has to be refreshed (which is why DRAM is relatively slow).
A flip-flop can be used to store one bit, but in order to do that two separate latches are needed. One latch has 2-4 gates each having 2-8 transistors. All of a sudden there are over a dozen transistors used to store one but.
A memristior does not actually help any calculations, it just hopefully make storing data much more efficent
Modern SRAMs usually use fewer transistors than that. 6T SRAMs are common, for example. They use a pair of inverters to store the state (4 transistors) and a pair of transistors to connect the inverters to the data lines (6 total). The write operation then involves a drive signal with more power than the inverters, thus forcing the state change.
They can replace SRAM cells, which take multiple transistors. They also retain state without needing power (similar to FeRAM, which can also replace SRAM). However, unlike FeRAM the memristor can also store analog values.
In the digital realm they're likely useful mostly as memory. However, given the analog properties, they could be useful in creating "fuzzy" neural nets.
"Hybrid memristor-transistor chip" doesn't really sound ridiculous to me. Would you have preferred he created a marketing name to describe a new fundamental technology?
Current technology contains stuff like "metal-oxide semiconductor field-effect transistors," we just don't call them that very often.
Hybrid, transistor, and chip? No, no problem whatsoever. My problem is with memristor, and memristance. They were made up to describe a resistance "setting" being remembered by a variable resistor. And frankly, the only problem I have is that the word originated for written text, where it is fine. I, however, am a big fan of the spoken language and the flow of a well constructed sentence. For a car analogy, a memristor is like a faulty transmission in the car that is a sentence's flow. It lurches and degrades the overall ride.
I believe what they mean is that the memristor is "fundamental" in that it provides a symmetrical link between electric charge and magnetic flux. A resistor is the link between current and voltage, a capacitor - voltage and charge, an inductor - current and magnetic flux, and a memristor - magnetic flux and charge. I think. I'm sure someone with a better grounding in electromagnetic physics can clarify this a bit.
It is the fourth fundamental linear circuit element. Transistors (I = k*V^2) and diodes (I = e^V) are not linear. Resistors V = k*I, capacitors I = k*dV/dt (the derivative is a linear operation), inductors V = k*dI/dt, memristors V=k(t)*I are linear.
The reason there are four linear circuit elements is clear if you write these equations in terms and flux and charge.
Hybrid, eh? (Score:3, Funny)
But does it get better gas mileage??
Unsurprisingly, yes (Score:5, Interesting)
Parent
Re: (Score:3, Funny)
Well that was faster than expected... (Score:5, Interesting)
Am I the only when that thought memristors would remain the the 5-10 year category for the next couple of decades? Granted, this is just a proof of concept chip but it is moving along very rapidly compared to most 'game changing' advances.
20 years of theory and work just to make the first memristor, less than a year to use the new memristor in a device that actually improves over the standard technology. So when will we see commercially available devices? Next year some time at this rate?
Re:Well that was faster than expected... (Score:4, Insightful)
Parent
Re:Well that was faster than expected... (Score:4, Funny)
Right now they still think we are in a recension.
We're in a critical revision of a text?
Parent
Re:Well that was faster than expected... (Score:5, Funny)
Thank you! I'll be here all week!
Parent
From the article (Score:5, Insightful)
"Williams says. Still, he predicts that memristors will arrive in commercial circuits within the next three years."
It seems fast because nobody was talking about these things for the last 30 years. It's only because of technological advances in circuit printing and general computing that we can make these things and integrate them without having to develop a lot of additional technology. The transistor is very old but only after developing a lot of supporting tech have we been able to shrink them down to fit billions in a processor. That same tech can already be applied to memristors. We don't need to wait decades before we can shrink a memristor down to practical levels for ICs.
Parent
Re:Well that was faster than expected... (Score:5, Insightful)
Parent
Re:Well that was faster than expected... (Score:4, Funny)
Am I the only when that thought memristors would remain the the 5-10 year category for the next couple of decades?...So when will we see commercially available devices?
In 5-10 years.
Parent
Re: (Score:3, Interesting)
Ahh, but what's the resolution of that one 'bit'? AFAIK this is an analog device, not at all limited to 1 or 0, but able to be used as such to to the handy-dandy DAC. :-D At first this is a thing for circuit-nerds, but don't despair, as when the circuit-nerds have hand their hands dirty for a minute, computer nerds are going to get to write "programs" for FPGAs, and then not long after that, libraries will emerge that let C nerds (and regular ole' C monkeys) utilize the goodness of these little variable c
size? (Score:2)
Re:size? (Score:5, Funny)
Does anyone know what size features the chip was etched at? um? nm? That might give a clue how close it is to being used in other products.
m. It could be a while.
Parent
Re: (Score:3, Interesting)
You joke but look at this
http://en.wikipedia.org/wiki/Memristor#Titanium_dioxide_memristor [wikipedia.org]
Although the HP memristor is a major discovery for electrical engineering theory, it has yet to be demonstrated in operation at practical speeds and densities. Graphs in Williams' original report show switching operation at only ~1 Hz. Although the small dimension of the device seem to imply fast operation, the charge carriers move very slowly, with an ion mobility of 10E-10 cm2/(V s). In comparison, the highest known drift ionic mobilities occur in advanced superionic conductors, such as rubidium silver iodide with about 2*10E-4 cm^2/(V s) conducting silver ions at room temperature. Electrons and holes in silicon have a mobility ~1000 cm^2/(V s), a figure which is essential to the performance of transistors. However, a relatively low bias of 1 volt was used, and the plots appear to be generated by a mathematical model rather than a laboratory experiment.[8]
1Hz? Next!
Re:size? (Score:4, Insightful)
the plots appear to be generated by a mathematical model rather than a laboratory experiment
This is what I would say "NEXT!" for, but to each their own.
Parent
Re: (Score:3, Informative)
1Hz? Next!
The application they talk about is the programmable transistors in FPGAs.
You can find FPGAs in all kinds of consumer electronics. Typically, the chip has some onboard flash from which it loads its configuration during power on. These transistors which load from flash during power on are the ones the researchers are talking about replacing with memresistors (memristors?). For this kind of application, a 1Hz write speed is completely reasonable. The write will be done once in the factory.
Re:size? (Score:5, Informative)
Does anyone know what size features the chip was etched at? um? nm? That might give a clue how close it is to being used in other products.
The memristors made in April were 50 nm wide
Parent
Spice model (Score:3, Insightful)
Is there a spice model available?
Re:Spice model (Score:5, Informative)
Who is the idiot modding this as Funny? A SPICE model is an engineering tool used to diagram circuits. It is NOT related to any of the Spice Girls!!! :p
Where's the link to MetaModerate?? grrr...
Parent
Re: (Score:2)
I don't know... I didn't get my degree in EE or anything, but I was able to guess what "spice" was from the context.
So I say it's no excuse.
Re: (Score:2)
FTA:
Ultimately, the next few years could be very important for memristor research. Right now, "the biggest impediment to getting memristors in the marketplace is having [so few] people who can actually design circuits [using memristors]," Williams says. Still, he predicts that memristors will arrive in commercial circuits within the next three years.
So probably not yet. From what I've seen just from a quick Google search, it will involve updating not just the models but the core code as well. [cadence.com]
Re:Spice model (Score:5, Informative)
I think you could build one from components if you had data about the memristance function. Start with a voltage controlled resistance element (Gxxx) connected to the two exposed terminals. Add a hidden element of a current controlled current source, sensing on the VCR element, injecting current into a hidden capacitor. The voltage on that capacitor is proportional to the total charge that has passed through the memristor device. The exposed VCR element senses voltage on the hidden capacitor, and uses an interpolated table of resistance vs voltage rather than a linear relationship.
That doesn't capture the hysteretic behavior of the current devices, which stop integrating at the some limiting points, but it's a starting point. You could add such behavior with a few more hidden components (back-to-back ideal zeners across the hidden cap, for example), though getting the right behavior might be a little tricky.
Parent
**** SPOILER ALERT **** (Score:5, Funny)
Re:**** SPOILER ALERT **** (Score:5, Insightful)
Parent
More than 2 states are now possible. (Score:5, Insightful)
Binary was chosen earlier in computer work for it could represent accurately a digit representive such as 1001 equals 9. Also magnetic core memory could hold only the two states.
With memristors (once they are perfected) can have multi-state such as trinary (base 3) or decinary (base 10) eliminating all of the conversion that is neccessary in the present binary system that require cpu cycles. 123 in the decinary system represents 123 where in binary it would be 1111011 and need to be converted in order to be meaningful.
For instance I have heard for those studying DNA that using base 12 has certain benefits in directly expressing information. Perhaps this will open a whole new arena of possibility that previously could only be simulated in binary.
The mind can imagine many new possibilities if the memristor actually is.
Re:More than 2 states are now possible. (Score:4, Informative)
Um, for the most part, the computer only has to convert from binary to decimal when it displays base 10 numbers on the screen (ie using the calculator). It's hardly computer intensive. All the operations (add, subtract, multiply, divide) are going to be in it's native binary, no conversion needed.
Computers convert data all the time - this text you are reading now is really just a series of binary numbers converted to ascii or unicode or whatever with lots of other conversions needed to throw it on the screen.
Native base 10 has been done before (basically ignoring bits representing 10-15) and all that was found was that it wasted space as conversion in those scenarios are beyond trivial. Here's a book for you:
http://www.amazon.com/Code-Language-Computer-Hardware-Software/dp/0735611319 [amazon.com]
Parent
Re:More than 2 states are now possible. (Score:4, Informative)
Parent
Re:More than 2 states are now possible. (Score:4, Interesting)
Thanks for the suggestion on the book.
I can write those books and have in the past. My experience stems from 1962 being a systems Analyst for a computer system. In the old days before IC's and even before transistors were a part of computers the vacuum tube was used.
In those days the adder section, rather than being a single chip was composed of discrete components. Once an adder problem was a wire wrap on the back panel that was making poor connection causing it not to promote a "carry" from the previous position.
I really do understand binary but keep in mind that having more than the two states of binary permits a smaller size over all. For instance the 123 I mentioned needs 7 positions in binary but only 3 positions in decinary.
Watch for it. Eventually this will be the going thing. Binary locked us in and was very restrictive while this invention and others in a similar vein will present opportunities we could not imagine before.
Parent
Re: (Score:2)
I'm sure that they will use 2^n states and consider them as groups of bits.
Wow (Score:4, Interesting)
Re:Wow (Score:5, Insightful)
Parent
Re: (Score:3, Insightful)
Neural nets? (Score:3, Interesting)
This memristor technology sounds like it could be an ideal device for implementing neural nets. Anyone working on that, I wonder?
-jcr
Re: (Score:3, Funny)
huh?
This Memristor technology sounds like it could be an ideal device for mounting lasers onto sharks. Anyone working on that, I wonder?
-SNS
Symbol (Score:2, Interesting)
FPGA FUD from TFA (Score:3, Informative)
This isn't really true. The rising fixed costs of an ASIC is prohibitive for low volume embedded projects where a $1 FPGA will do just fine. High performance FPGA chips are about the same cost as a CPU and they are commonly used as reconfigurable co-processors for supercomputing applications or embedded DSP. And I get way more GigaOps per dollar with FPGAs than with a CPU and for much less power.
Re:How does it replace multiple transistors (Score:5, Informative)
Parent
Re:How does it replace multiple transistors (Score:4, Informative)
Parent
Re: (Score:3, Informative)
They can replace SRAM cells, which take multiple transistors. They also retain state without needing power (similar to FeRAM, which can also replace SRAM). However, unlike FeRAM the memristor can also store analog values.
In the digital realm they're likely useful mostly as memory. However, given the analog properties, they could be useful in creating "fuzzy" neural nets.
Re: (Score:3, Interesting)
Is there any free circuit simulation software that includes simulated memristors for me to fool about with?
I still don't get quite how they work.
Re: (Score:2)
"Hybrid memristor-transistor chip" doesn't really sound ridiculous to me. Would you have preferred he created a marketing name to describe a new fundamental technology?
Current technology contains stuff like "metal-oxide semiconductor field-effect transistors," we just don't call them that very often.
Re: (Score:2)
Hymtric?
MOSFET is a pretty common "spoken acronym" in the electronics world.
Re:Unfortunate (Score:5, Interesting)
Parent
Re:Unfortunate (Score:5, Funny)
Parent
Re:Unfortunate (Score:4, Funny)
Parent
Unfortunate:Commitment. (Score:2)
"Now, who wants to try their hand at coming up with a better name?"
Hysterical Electronics:They lose their cool before you do.
Re:Unfortunate (Score:5, Funny)
Autopots! Transist and roll out!
Parent
Re:Feds Deregulating Hybrid Corn for ETHANOL?? (Score:5, Funny)
Parent
Re:4th? Try 5th fundamental (Score:5, Informative)
Parent
Re:4th? Try 5th fundamental (Score:4, Informative)
It is the fourth fundamental linear circuit element. Transistors (I = k*V^2) and diodes (I = e^V) are not linear. Resistors V = k*I, capacitors I = k*dV/dt (the derivative is a linear operation), inductors V = k*dI/dt, memristors V=k(t)*I are linear.
The reason there are four linear circuit elements is clear if you write these equations in terms and flux and charge.
Parent