What's new
  • ICMag with help from Landrace Warden and The Vault is running a NEW contest in November! You can check it here. Prizes are seeds & forum premium access. Come join in!

'Skynet' gets its basic building blocks

Verite

My little pony.. my little pony
Veteran
What happens when Moores law gets blown outta the water? When computer systems can program themselves? The time is sooner than you think.

Tiny Brain-Like Computer Created

The most powerful computer known is the brain, and now scientists have designed a machine just a few molecules large that mimics how the brain works.
So far the device can simultaneously carry out 16 times more operations than a normal computer transistor. Researchers suggest the invention might eventually prove able to perform roughly 1,000 times more operations than a transistor.

This machine could not only serve as the foundation of a powerful computer, but also serve as the controlling element of complex gadgets such as microscopic doctors or factories, scientists added.

The device is made of a compound known as duroquinone. This molecule resembles a hexagonal plate with four cones linked to it, "like a small car," explained researcher Anirban Bandyopadhyay, an artificial intelligence and molecular electronics scientist at the National Institute for Materials Science at Tsukuba in Japan.

Duroquinone is less than a nanometer, or a billionth of a meter large. This makes it hundreds of times smaller than a wavelength of visible light.

The machine is made of 17 duroquinone molecules. One molecule sits at the center of a ring formed by the remaining 16. The entire invention sits on a surface of gold.

How it works

Scientists operate the device by tweaking the center duroquinone with electrical pulses from an extremely sharp electrically conductive needle. The molecule and its four cones can shift around in a variety of ways depending on different properties of the pulse - say, the pulse's strength.

Since weak chemical bonds link the center duroquinone with the surrounding 16 duroquinones, each of those shifts too. Imagine, for instance, a spider in the middle of a web made of 16 strands. If the spider moves in one direction, each thread linked to it experiences a slightly different tug from all the others.

In this way, a pulse to the central duroquinone can simultaneously transmit different instructions to each of the surrounding 16 duroquinones. The researchers say this design was inspired by that of brain cells, which can radiate branches out like a tree, with each branch used to communicate with another brain cell.

"All those connections are why the brain is so powerful," Bandyopadhyay said.

Since duroquinone possesses four cones, each molecule essentially has four different settings. Since the central molecule can simultaneously control 16 other duroquinones, mathematically this means a single pulse at the machine can have 4^16 - or nearly 4.3 billion - different outcomes.

In comparison, a normal computer transistor can only carry out just one instruction at once, and only has two settings - 0 and 1. This means a single pulse at it can only have two different outcomes.

Putting it to work

The idea is to hook this new gadget up with other molecules - either copies of itself or different compounds other scientists have invented. For instance, researchers have created a host of machines just a molecule or so large over the last decade or two - motors, propellers, switches, elevators, sensors and so on. The new invention might offer a way to control all those other compounds to work as a whole. Indeed, Bandyopadhyay and his colleagues revealed they could hook up eight other such "molecular machines" to their invention, working together as if they were part of a miniature factory.

This invention could serve as the controlling element of complex assemblies of molecular machines, Bandyopadhyay suggested. One future application for such assemblies "could be in medical science," he told LiveScience. "Imagine taking assemblies of molecular machines and inserting them into the blood, perhaps if you wanted to destroy a tumor inside the body."

The device currently is operated with an extremely sharp electrically conductive needle - specifically, that belonging to a scanning tunneling microscope, a bulky machine far larger than the 17 molecules in question. However, Bandyopadhyay hopes that in the future they can issue commands to their invention using molecules that deliver electric pulses instead.

The device needs to be made in vacuum conditions at extremely cold temperatures - about -321 degrees F (-196 degrees C). Bandyopadhyay said it could be operated at room temperature, however.

More powerful still

Bandyopadhyay added they could expand their device from a two-dimensional ring of 16 duroquinones around the center to a three-dimensional sphere of 1,024 duroquinones. This means it could perform 1,024 instructions at once, for 4^1024 different outcomes - a number larger than a 1 with 1,000 zeroes after it. They would control the molecule at the center of the sphere by manipulating "handles" sticking out from the core.

"We are definitely going to 3-D from 2-D immediately," Bandyopadhyay said.

Bandyopadhyay and his colleague Somobrata Acharya detailed their findings online March 10 in the Proceedings of the National Academy of Sciences.

http://news.yahoo.com/s/livescience/20080311/sc_livescience/tinybrainlikecomputercreated
 

Echoes

Member
That sound is this topic going over my head...

That sound is this topic going over my head...

lolwut





Nice new avatar, by the way.



EDIT: Going to take me a while to research some of this. For the time being, if Verite says it's bad...it's bad.
 
Last edited:

nycdfan042

Its COOL to DROOL!!!!!!
Veteran
SKYNET?? isnt that the name of the company in the terminator that built the software for the robots?? LOL as long as they come up with little robots that clean the penis pate off my dill banger...
 

twojoints

Member
i dont know... skynet... i just dont think its realistic. i mean, the technology? sure. but i dont see computers suddenly turning evil on us... because theyre not even happy towards us to begin with. theyre just computers, no matter how much AI we give them. thats exactly what it says, "artificial" intelligence. something about a human soul that could never be replicated...
 

Dr Dog

Sharks have a week dedicated to me
Veteran
twojoints said:
i dont know... skynet... i just dont think its realistic. i mean, the technology? sure. but i dont see computers suddenly turning evil on us... because theyre not even happy towards us to begin with. theyre just computers, no matter how much AI we give them. thats exactly what it says, "artificial" intelligence. something about a human soul that could never be replicated...


What do you mean unrealistic?
Sure on the terminator scale, sure, but just look at what could happen to a military software program, quite easily could something given, some sort of early AI. They(programs) could see people as a threat, and if they are programmed to take action against threats, bamm a mild catatastrophe

Sure this may sound far fetched, but remember that alot of things that we have now are the result of Hollywood
1. Cell phones
2. Plasma Tv
3. The taser
4. email
5. The pc even

All in someones imagination at one time, skynet cannot be that far off
 

Verite

My little pony.. my little pony
Veteran
If these things actually do replace the transistors in computers its going to be crazy the computational power they will have. Even as complex as the micro-processor has got it still just does things sequentially really fast with transistor gates that only do two states, '0's and '1's. To have 16 states for one input is an insane jump in processing. To take that to 1024 states for each input and you basically have rewritten Moores law and what we know about the limits of computational power.

At that point its not a matter of programing AI, all you would have to do is give the thing a handful of different programming routines and tell the thing to program itself.

The only save factor is the stuff isnt able to replicate outside some harsh lab conditions, if it does get to the replication at room temp level then it should get to be some wicked technology.
 

tngreen

Active member
Veteran
wow that is scary as hell! brain-like power in nano-machines? not good. i know there is good intentions with this but there could be some very bad consequences.
 

C6H6

Member
Way too many unknown factors the application of technology like this is years and years in the future way past out lifetime. Many factors still need to be controlled in order for any chance of this being instituted.
First problem duroquinone (C10H12O2) while it’s a fairly stable molecule but you start subjecting it to conditions like the ones needed to do anything like what their talking about, its going to generate heat. The thing would be subjected to an unfathomable amount of electrical pulses every second which would generate mad amounts of heat. With duroquinone melting point of about 104-114C (228-237F) without proper cooling it would simply break the compound down rendering it useless. To cool something like that would take a very unreactive substance with a high specific heat capacity to absorb the heat and transfer it away from the compound. And with the structure of duroquinone having both polar and nonpolar regions something like water for example, would create a temporary bond and when those electrical pulses came they would simply dissipate in the water, become distorted and most wouldn’t make it to the receiver that would be needed that’s equally as small to interperate the signal else its just a fancy electrical coil.
chemistry lesson, the reason plants need light is because the chemical in the chlophyll absorbs the light energy from the sun. The the electrons get excited and jump into the next energy level and when they fall back, The energy is then transferred from the chlorophyll molecule to different areas of the plant where its needed for synthesis of different compounds the plants needs. Point being the energy that is absorbed goes somewhere and isn’t just reemitted, that what heat it. with duroquinone being smaller then the visible spectrum id be worried about other forms or electromagnetic radiation like UV light to name one. It would absorb the energy (light energy) and have no place to go and would be reemitted in the form of heat leading to the eventual degradation of the duroquinone. Exact same principal why light make THC less potent it breaks down the weaker bonds. That’s also why certain prescription drugs come in those orangeish bottles, the light is strong enough to break down some of them.
Having said that It’s a phenomenal idea. This kind of thing that can put mankind into a new era. Great read.
 

ItsGrowTime

gets some
Veteran
C6H6 said:
First problem duroquinone (C10H12O2) while it’s a fairly stable molecule but you start subjecting it to conditions like the ones needed to do anything like what their talking about, its going to generate heat. The thing would be subjected to an unfathomable amount of electrical pulses every second which would generate mad amounts of heat. With duroquinone melting point of about 104-114C (228-237F) without proper cooling it would simply break the compound down rendering it useless. To cool something like that would take a very unreactive substance with a high specific heat capacity to absorb the heat and transfer it away from the compound

Im no nuclear physicist but I know enough to get myself into trouble. Wouldn't this be the same principle as what we currently use to harness nuclear power? I saw your mention of water but water itself never comes in contact with nuclear cores, but rather uses the heat from the nuclear reaction to heat the water to a high enough temp to propel a steam turbine. I don't see how someone couldnt take the same principle to a smaller level to remove heat and transform it into steam to propel something else. The scale is the only difference IMHO. In other words, take the heat created from this intelligent reaction and force it to propel a turbine that powers a droid, a la Terminator.
 
Last edited:

C6H6

Member
correct the water never touches the reactor it go through a series of tubes that are close enough to the reactor that the radiant energy put out gets absorbed and pulled away. in order to make it work with the duroquinone youd need a barrier separating the molecule from the coolant but it would have to be incredibly close to the duroquinone in order to absorb that radiant energy and you be back at square one worring about the interactions between the compound and the material the barriers made of.
Water was just an example because of its high specific heat capacity
 

C21H30O2

I have ridden the mighty sandworm.
Veteran
i think your gunna have to think of something other than water to cool something smaller than a nanometer.
 
G

Guest

Verite said:
If these things actually do replace the transistors in computers its going to be crazy the computational power they will have. Even as complex as the micro-processor has got it still just does things sequentially really fast with transistor gates that only do two states, '0's and '1's. To have 16 states for one input is an insane jump in processing. To take that to 1024 states for each input and you basically have rewritten Moores law and what we know about the limits of computational power.

At that point its not a matter of programing AI, all you would have to do is give the thing a handful of different programming routines and tell the thing to program itself.

The only save factor is the stuff isnt able to replicate outside some harsh lab conditions, if it does get to the replication at room temp level then it should get to be some wicked technology.
Replicators, ie Stargate.

Resistance is Futile.
 

Latest posts

Latest posts

Top