Hewlett-Packard Labs has now put out a FAQ on memsistors:
http://www.hpl.hp.com/news/2008/apr-jun/memristor_faq.html
This is a fascinating development in electrical theory for me. I've had just enough electronic training to be dangerous to myself, but nobody else in the world. I like the theory part. I'm lousy at taking the theory from my head and put it into my hands. The first electronic school I went to was in the Navy and was about the old electron diode vacuum tubes. That happened in the late Fifties, and the transistor had just arrived on the consumer market about this time. The first portable radios like the Walkman were just being fitted with solid state transistors. I've been to a couple of other schools and studied electronics in the mid-80s just as personal computers arrived. I bought my first computer in '88. I'm not a nerd. This was forced on me by circumstance. The real reason I bought my first computer was to replace an electric typewriter. The ease of editing on a computer before I hit the Print button was immediately recognizable as a superior advantage. Particularly to someone who typed as badly as me.
The memsistor FAQ is really well written in a way a layman might understand the basic idea of memsistors and how they form the fourth cornerstone of electrical theory. I didn't expect to understand what i read in the FAQ. but I did. The head of the team that made the prototype and wrote it up for Nature magazine wrote the FAQ and used how water flows through pipes to illustrate, and I worked for twenty odd years fitting and welding pipe. The way he explained it I understood exactly what it was the missing link of electric theory. It remembers it's ongoing condition when it's turned off. It stays right where it was when the power or water is turned off. When the power or the water is turned back on it starts from the configuration it was in when it was turned off. That's why it can be used for "instant on" computing when it's used to make Random Access Memory. It doesn't "do" anything to remember it's contents as memory. It just stays like it was left, and when you turn it on again it starts from there. That means that unless some change is introduced it doesn't ever have to be turned on beyond static recognition of where it's at. Energy doesn't have to be expended to refresh or change anything about it's unchanging state of being unless change is required. I may have made that last part up.
DRAM has to be refreshed ever so often for it to maintain it's integrity as a zero or one. If the power is turned off, then the memory stored in the cells loses it's assigned value. When the power is turned on again, the DRAM has to have it's memory slots refilled from the hard drive again before data processing can happen. Random Access Memory made from memsistors wouldn't have to be refreshed from the hard drive. You could just turn on the power again and everything would be just like you left it when you cut the power off. I may have missed the mark here. I may have understood memsistors to work a little like ROM memory, and I don't know why, but I'm sorta positive what these people have done is not as if they reinvented programmable memory. What I don't understand is that if all the parameters present in a memsistor at any given moment stay the way they exist as when you turn off the juice, what does turning the juice back on do to the frozen state the memsistor was left in. Does it have the same affect as a BIOS battery that holds the BIOS code in ROM memory while the motherboard is turned off.
No battery is needed to keep the information in the memsistor even if you turn off the power to the circuit it's a part of. What happens to the data stored in the memsistor when the circuit power is turned off and then turned back on. Doesn't that alter the way it was left? What I'm curious about is, does any of the regular circuit power pass through the memsistor if it's contents doesn't need to be changed. I'm thinking about efficiency here.
Will the presence of memsistors in a digital circuit make it more efficient because no energy is exerted keeping the memory registers refreshed. I wonder if the minuscule amounts of electricity involved in the various brain waves have biological memsistors or the equivalent thereof. When I upsurge into conscious awareness when I wake up from a nights sleep, it doesn't take me long to jump right back into the patterns I retired from the night before. This might have something to do with Alzheimer's. Old people's bionic memsistors short out. '-)
I do not attempt to tell the God's own truth here because I don't know what the truth is or hardly ever. I try to capture the drifting thoughts that randomly appear in my imagination for reasons I may not understand. I don't know if the content I capture with these words is true or false. The Comments settings are turned off to prevent me from having to defend what amounts to little more than fanciful, sometime crude speculation. Great moments in our lives never return.
