Slashdot Mirror
First Quantum Computing Gate on a Chip
An anonymous reader writes "After recent success in using quantum computing for superconducting qubits, researchers from Delft have formed the first Controlled-NOT quantum gate. 'A team has demonstrated a key ingredient of such a computer by using one superconducting loop to control the information stored on a second. Combined with other recent advances, the result may pave the way for devices of double the size in the next year or two--closer to what other quantum computing candidates have achieved, says physicist Hans Mooij of the Delft University of Technology in the Netherlands. Unlike today's computers, which process information in the form of 0s and 1s, a quantum computer would achieve new levels of power by turning bits into fuzzy quantum things called qubits (pronounced cue-bits) that are 0 and 1 simultaneously. In theory, quantum computers would allow hackers to crack today's toughest coded messages and researchers to better simulate molecules for designing new drugs and materials.'"
I know grammar has been taking a hit in society as of late, but now even our computers are blatantly spewing out double negatives?
We're not in for an unrough ride, gentlemen.
I find it interesting that the first electronic computing gates devised were the AND/OR gates, using basic diode logic. Quantum computing research develops the NOT gate first. I think this has something to do with the esoteric nature of quantum computing. AND/OR gates require two inputs to change to a single value, where NOT is merely an inverter. The idea of entanglement makes the inversion process a likely first step in quantum research.
For those wondering why this is important, the first true electronic gates were invented in the early 1920's. This predates point-contact transistors by about 20 years, invented in 1947. 60 years later, here we are with transistor computing in every aspect of our lives.
At the rate quantum computing is advancing, I think we can expect to see quantum transistors (in the lab, at least) by 2020. A true useful quantum computer may be available less than 50 years from now. Hopefully by then someone will pick up the slack and have the Linux kernel ported to the Q-CPU architecture!
khasim (12/9/06): In a blind taste test, more people preferred Coke over the Pepsi that I had previously pissed in.
Not Jokes:
It's a Quantum Gate.... NOT!
games journalism blog
Sound a lot like Tribbles to me.
but how can they test it when the output is always either 0, meh, pfft or 1?
They whose government reduces their essential liberties for temporary security, receive neither liberty nor security.
I'm no quantum theory expert by a very long shot, but it was my understanding that there are 32 quantum states of electrons, not just on/off (1/0) like in the binary computer world. So, if we now have a quantum NOT gate, doesn't that mean there are 32 possible states of the NOT gate? Also, according to the article the CNOT gates they created can be both 0 and 1 simultaneously. In my mind this would cause errors and actually stop the flow of information instead of speeding it up.
Someone with some understanding of this stuff please elaborate, before my head asplodes.
In theory, quantum computers would allow hackers to crack today's toughest coded messages.
That's an overstatement. A quantum computer will not suddendly magically crack the strongest codes. Yes, certain algorithms designed for quantum computers, like Grover's algorithm, will reduce the time needed to find the key of a symmetrical cipher with about half the number of bits in the key. However, given for example a 256-bit key you would still have ~2^128 keys to check and afaik 2^128 still takes quite sometime to crack....
They're opening the quantum gates now? They're insane! Who knows what might pour out of them... I hope they're at least doing it on the moon.
The future of the human race is up to one lone marine now. Thanks a lot, scientists.
Dude you're getting a Delft!
Can anyone remember the name of that assembler that only had the 'not' operator? Maybe it's time for a port :)
> the result may pave the way for devices of double the size in the next year or two
Well, at the current rate of progress, we might see a Quantum Pentium III in about 26-52 years, depending on whether its "next year" or "two". I might be dead of old age by then.
At home you will use these for ever more sophisticated rendering of artificially intelligent virtual reality porn.
At work it will be more useful in the advanced simulation of a mechanical process for imprinting letter glyphs on sheets of wood fiber.
Help stamp out iliturcy.
"Mooij you're gettin' a Delft!"
If a qubit is both 0 and 1 at the same time, what is the point of inversing it? Would it then be 1 and 0 at the same time?
(comment 2):"How come Delft U has been able to perform a CNOT with two qubits using superconducting technology? I thought Rose/D-wave claimed it was extremely difficult to do discrete quantum gates with superconducting technology. What are the present & future limitations of the Delft "quantum computer?"
Rose IGNORED the question. The quantum computer built by D-Wave is an adiabatic computer (which is an analog computer), whereas the Delft people have built a discrete gate quantum computer. Does the Delft computer make D-Wave's computer obsolete?
on whether someone is observing (or not, which is not and in negative logic).
"National Security is the chief cause of national insecurity." - Celine's First Law
Dangit, and I'm having enough trouble in computer science as it is, without all this fuzzy simultaneous 0/1 nonsense.
Hmm, seems like they've successfully performed a NOT on Moore's law.
Stop! Dremel time!
Transistors are capable of more than on and off -- they can handle many intermediate stages of charge (fundamentally low, medium, high), which computing applications do not currently exploit. Why not add a third state by using technology that already exists? What are the benefits of quantum computing over the existing versatility of transistors?
by turning bits into fuzzy quantum things called qubits (pronounced cue-bits) that are 0 and 1 simultaneously
Sounds like any ol' woman to me, nothing to worry about, we have been handling it for centuries.
Custom electronics and digital signage for your business: www.evcircuits.com
You'd never know if an article was a dupe or not.
Engineering is the art of compromise.
This is awesome no it's not!
Having taken a class on quantum computing last semester I would really like to add in some facts here. First to say qbits are both 1 and 0 at the same time is not entirely clear. Qbits are represented by arrays of length 2. These can be either [1,0] or [0,1]. Where the confusion occurs is when these are a superposition of the two, which means basically means that there is a probability that the result would be one of the two. What results from this is knowing the result most of the times, but sometimes being uncertain. For the uncertain cases there are ways to use the probabilities where in almost all cases only the more probable case will result. Also it is not completely correct to say we have no idea of how these will work. We have a pretty damn good idea, we just have not tested it yet. In fact, most of quantum computing is just simple linear algebra, as the qbits can be represented as arrays and the gates that control them can simply be represented by 2 by 2 matrices. Obviously this is only the basics of it, not touching on entanglement or any algorithms(which can all be represented by multiplying matrices). Anyway I did a pretty bad job of explaining all that, but the point is that this is a big deal and we are way ahead of understanding how these things should work in the future over understanding how to make a machine that will make them work.
I want to log into that machine and run some quantum Perl scripts on it. Nothing like an existing library of code to kickstart a new architecture.
--
make install -not war
How does a quantum NOT gate work exactly? Normal NOT gates make zeroes into ones and visa versa. So this makes something from a simultaneous one and zero into a simultaneous zero and one?! How does this even help perform calculations? How do you use that info? *Not zero or one, but both? WTH?* Sorry if this is an obvious question, Discrete mathematics isn't in the curriculum for aerospace engineering. Does this do tons of simple calculations very fast? If not, I don't see as much of an application in it. Maybe if it could do some higher level calculus *in hardware* then it has real value.
Last Post!
Qubits represent a probability of being a 0 or 1. Observing a qubit destroys that probability, and you "read" only a zero or a one. You don't actually know what the probability was of reading that value, only what the value was at the moment you read it (and you can't read it again - by reading it, you've destroyed the state).
If I were to make a creative leap I would say, something that can be utilised which has two binary states at one time could somehow turn in to something that could result in erroneous random but recorded switching which then could in turn develop into an evoloving set of data that could potentially become self aware.
Or it could just spit out junk.
Either way my vote is that it should be called "Deep Thought".
-JB
"I love deadlines. I love the "whooshing" sound they make as they pass by." - Douglas Adams.
IAMAQP (I am not a quantum physicist) but the theory I read explains a system gaining processing power from shared computing of a single processor replicated across multiple realities. Each qubit is a calculated answer by a machine in one reality and the culmination of those answers assumedly gives you the correct response. David Deutsch wrote a book on this called "The Fabric of Reality" that works through the concept of a basic Turing machine - where computers all come from - and how this can be re-worked into a quantum processor.
There's a lot more math to it than that, but the idea is a simpler approximation formula replicated infinitely across realities gives an accurate response much faster than any single reality calculating the absolute answer.
Cooler yet is that if they're actually making functional quantum gates does this mean the processor power is actually being derived from other realities? That would be awesome and totally Outer Limits material.
-Matt
--- Need web hosting?
Disclaimer: I am a quantum information scientist
Qubits represent a probability of being a 0 or 1. Observing a qubit destroys that probability, and you "read" only a zero or a one.
This is at best an incomplete description of what happens. Qubits are quantum states, not probabilities. Quantum states are sometimes called "probability amplitudes", in that taking the square of the magnitude of the coefficient for a particular basis state gives you the probability of getting that state if you measure in that basis. There are a few very important points: (1) we're dealing with complex numbers, and things work in such a way as to give us the possibility of "interference" of probability amplitudes; (2) quantum states are real states, not just probabilities representing our ignorance of which classical state you'll find when you measure.
A brief intro to the math:
Let's take some qubit in some arbitrary state, which we'll call | psi > (I'm using Dirac notation). We can completely describe the state as follows:
| psi > = a | 0 > + b | 1 >,
where a and b are complex numbers, and have the property that |a|^2+ |b|^2 = 1. We see that we have an uncountably infinite number of possible states for just a single qubit. If psi were a classical bit instead of a quantum bit, we could use essentially the same description, except that the requirement on a and b would then be that exactly one of them is 1, while the other is 0 (only two possible states). If psi were a "classical" analog "bit" or a probabilistic bit, the requirement would be that a, b in [0,1], and a+b=1.
What happens if we measure psi? It depends on the basis we choose to measure in, but if we go to measure psi in the {| 0 >, | 1 >} basis, we'll get | 0 > with probability |a|^2, and | 1 > with probability |b|^2. Figuring out probabilities for other bases requires only a basis transformation (simple linear algebra).
Now, this qubit business seems horribly messy--we have an infinite number of states for a single qubit--how can we possibly describe the action of a two-qubit gate like controlled-NOT (CNOT)? Fortunately, quantum mechanics is linear, which means that if we describe how a gate operates on each of the possible input basis states, we've completely specified the gate. For two qubits, we can use the following basis: {| 00 >, | 01 >, | 10 >, | 11 >}. Labeling the rows and columns in that order, we get the following truth table for the CNOT gate:
1 0 0 0
0 1 0 0
0 0 0 1
0 0 1 0
In other words, if the first bit is 0, do nothing to the second bit. If the first bit is 1, flip the second bit.
It turns out that CNOT plus a bunch of different single qubit gates is universal, meaning you can use that set of gates to implement any "quantum circuit".
A 'bit' is simply shorthand for "binary digit". Quantum digits, however, aren't binary, since they can represent much more than a simple 0 or 1. By adding the 'Qu-' to the term, we are essentially calling them "Quantum Binary Digits," which is in itself an oxymoron.
Qubits are quantum bits in the sense that there are two basis states (often | 0 > and | 1 >). In some systems, qutrits (quantum trinary digits) are more natural, and occasionally, you'll hear "qudit", which is a d-level quantum digit, for some arbitrary d.
Given how often I have to say or hear the word "qubit", I guarantee it wouldn't be more fun if it had an extra syllable.
For those with access the paper can be found here (PDF). Dr.Dobb's and PhysOrg also have the story.
After trying to wrap my head around the physics here, I feel like I just want to call it 'quit'.
True confidence comes not from realising you are as good as your peers, but that your peers are as bad as you are.
The NOT gate is a simple bit inverter, but the CNOT gate (CONTROLLED-NOT) has two inputs, using the 2nd bit to invert or not the first bit. The article mentions the CNOT gate, not the NOT gate. In classic digital electronics, the CNOT gate equals the XOR gate:
http://en.wikipedia.org/wiki/Cnot
Cracking encryption algorithms? Sounds like the MPAA/RIAA may need to flex the DMCA yet again.
Out of modpoints but really liked a post? 1BDkF6TtmmeZ3yqXbz9yhdYVqRYnwFoXDj
If a qubit Q is both zero and one at the same time, then surely its complement !Q is also both zero and one at the same time? If you had qubits Q and R which were both 0 and 1 at the same time, then wouldn't (Q & R) be all of {0, 0, 0, 1} at the same time (so more likely 0), (Q | R) be all of {0, 1, 1, 1} at the same time (and so more likely 1), and (Q ^ R) be {0, 1, 1 and 0} at the same time (so equally likely 1 and 0)?
Just because a wave function has to collapse into one eigenstate when it is observed, doesn't necessarily mean that will collapse into the one you were hoping for! And you don't need to lock a cat in a box with a time bomb to prove that.
Je fume. Tu fumes. Nous fûmes!
One NOT gate ought to be enough for anyone.
How is this a step forward, I thought we already had gates that were NOT quantum gates...
- First they ignore you, then they laugh at you, then ???, then profit.
...that is the question
Next comes Quantum Coupled Ethernet to revolutionize communications. Quantum pair sets working to provide two point to point channels.
Within a Duo Quantum, 10GigE, or the AQ (Athlon Quantum) same difference with a hyperchannel set devoted to other chips scattered all over the world.
QCE is the "Zero Mile Solution", you heard it here on Slashdot first!
Bandwidth in long haul optical fibers will be a thing of the past, those companies long term value goes down the tube! Economics of each end drives the food chain of interconnected networks at each expensive junction melting into a lower cost anywhere possibility. Then low and behold, even your cell phone works in a mine shaft or in space 100 million miles away, with nary a cell phone tower. Oops cell tower syndicate companies now have a value equal to toast per site.
Smart phones will have more bandwidth than the little screens know what to do with.
This is our world in the next decade folks.
Remember, those who short the telecommunications economy, from cable, cell and phone with a dash of even satellite television challenged... Well now you can beat up on Wall Street ahead of those gullible institutional bankers trying to schill extra profits from an apparently revitalized set of companies. Well guess what folks, this is the last hurrah for them in a Quantum Communications world the amount of winners will be very small and the all of us get a better deal than the toll keepers to video and voice want to rain on your parade with yet another DRM scheme.
Wall Street is so damn blind to this!
http://www.aisnota.com/slashdot/ Welcome to Logic and the Future