ReRAM ??

Only if it includes a global warming component, and some stem cells.

Lots of worthless PhD stuff out there. Weaned to engineering level, I found people addressing modeling...

sentences

It has that too, it uses fuzzy logic.

exciting stuff there, can't wait to see the meminductor and memcapacitor.

I call shots on the memsuperconductor.

sentences

Powered by a fuel cell.

sentences

Running on only water.

There are 4 fundamental circuit variables: voltage, current, charge and flux.

Current is defined as the time derivative of charge and (Farday's Law) voltage defined as the time derivative of flux.

The fundamental 'elements' are defined by a relationship between two of the above 4 variables.

A resistor is the relationship between---------- voltage and current. A capacitor is the relationship between--------- charge and voltage. An inductor is the relationship between--------- flux and current.

However, if one assumes symmetry, there is a 'missing' 4'th between--------------- charge and flux.

That is the theorized 'memsistor'.

Draw an XY graph. X axis current charge Y axis voltage ^ flux v

The 'elements' fit in the formed quadrants. Going clockwise you have in the lower left an inductor. Upper left is the resistor and upper right is the capacitor. The lower right is either 'missing' or a 'memsistor'.

(view fixed font)

voltage ^ resistor | capacitor | current charge | inductor | ????? v flux

I'll stick my neck out and say I'm not convinced because all the supposed 'memsistors' I've heard of depend on altering the physical structure of the device. In HP's, for example, ion migration is what alters the device's resistance. But in the commonly known 3 'elements' there is (ideally) no alteration of the device's physical structure and, indeed, such alteration would be a 'defect' or 'real world' deviation from the 'ideal' element. Examples would be heat altering the 'fundamental' resistance of a resistor (we don't call a thermistor a new 'element') or capacitor plates 'bending' from the attractive forces.

I don't doubt that useful devices could be constructed nor do I doubt this 'phenomena' could be a significant 'real world' issue in the nanoscale (just as leakage is) but it strikes me as a transition from the 'electric' to 'physical' realm rather than a new 'element'. A transducer, for example, interacts with the 'physical' realm but we don't call a speaker a new 'element'.

On the other hand, things get a bit fuzzy at the atomic level, so maybe, but, at any rate, that's the theory.

(view fixed font)

voltage ^ resistor | capacitor | current charge | inductor | ????? v flux

It?s a "flux capacitor"!

It might also be modeled by a battery or a motor. Instead of

Paul

Not necessarily. The antidiagonal entries on the plot of four quantities store energy, while the upper-left, at least, dissipates power. Perhaps the entire main diagonal dissipates.

- Resistance is the ratio of voltage to current; the product is power (real dissipation, nonconservative).

- Capacitance is the ratio of charge to voltage; the product is energy (stored, conservative).

- Inductance is the ratio of flux to current; the product is energy (stored, conservative).

- Memristance is the ratio of charge to flux (or the inverse, whichever); the product is technically energy * time, or momentum * length. Doesn't really make sense.

Tim

The units for a memristor is said to be Ohms. It dissipates just like a resistor. This has been used to argue it's not a new fundamental device since, it is claimed, a new fundamental device would require a new unit of measure. I'm not saying that's 'true', just that it's an argument made. Note, however, that they're both a ratio of (in generalized terms) 'quantity vs potential' with the difference being one is the differential of the other. e.g. dq-idt. It might be better to call the memristor unit Rt, or OhmT, to accommodate the 'memory' function.

Or, put another way, it strikes me that saying 'ohms' for a memristor is akin to noting the volt and current of an inductor or capacitor at a specific moment in time. It's valid for that moment but is different at others and the same goes for the memristor's 'Ohms'. It changes over time, depending on flux and charge, so Ohms is an insufficient measure of it's overall, time variant, characteristic.

A battery is not a passive device and a motor is not fundamental as it can be modeled by fundamental devices.

Someone mentioned the diagonal. Yes, the inductor-capacitor diagonal are energy storage elements. The resistor-memristor diagonal are dissipative elements.

From considerable searching I conclude that the ReRAM is something your door-to-door sweeper salesman might be hawking. Lots of claims. Little do. Those who _were_ going to do non-volatile RAM with it are "delaying" entry into the market for several years.

Even ran across a thesis and several papers where the "demonstrated" Spice model had dangling nodes ;-) ...Jim Thompson

Well, we're mixing apples and oranges here. Whether you can make devices reliably, economically and in quantity with the desired density is another whole matter from whether you can make one at all and plenty of people have demonstrated working devices of one form or the other.

What I heard is that HP's delay isn't for manufacturing reasons, per see, but because their partner doesn't want to encroach on flash sales at this time.

It's the angle of the dangle that matters, right? ;)