Flagged.
If you're going to take the time to make a comment, how about making a comment worth reading? "Flagged" doesn't mean anything useful.
'Unprecedented' discovery could propel quantum computers to reality
31–40 of 52 posts
Re: 'Unprecedented' discovery could propel quantum computers to reality
#32If you can store values for a given time frame with such time frame higher than the delay of the circuit, why don't we just build a latch and extend the time frame indefinitely? Or are these diamonds only settable once, and then you have to build another one?
Because when you read a qubit, you collapse the superposition and end up with a boring old 0 or 1. Which is what you want when the algorithm is completed, but during the execution of the algorithm you want to do operations on the superpositions. So you cannot read the value and feed it back because in the process, you destroy it.
Re: 'Unprecedented' discovery could propel quantum computers to reality
#33"For qubits, they can hold a value of “1” or “0” as well as both values at the same time. Described as superposition,"... Am a bit frustrated reading that same description for years and still having no clue what it really means. Both values at the same time... huh? What is the physical property that is existing in two states at once? And how does that feature translate into speedier computation? Is it so hard to unde…
Think of it like this. What if you represented your computer bits as a arrow on a compass. You have also the plane that the compass is on. The arrow points in one direction for 1 and in another direction for zero. Regular computation would have you doing operations on bits. You could only do one operation at a time on one bit and you are able to flip them given a recipe. A probabilistic computer allows you to do some…
Superposition can be described as follows. You can correlate two qubits together magically.
That's entanglement not superposition.
Re: 'Unprecedented' discovery could propel quantum computers to reality
#34I'm nitpicking just a little, but isn't discovery always about something unprecedented? Otherwise it's not that different from the rest of the scientific work people do. As a disclosure, I'm a physicist, having worked on quantum computing during my PhD. I think it's really cool, though long qubit lifetime doesn't take you far, the devil is in the details, for example good (reliable, low noise and tunable) qubit-qubit…
I think a discovery that improves the storage time of qubits by 6 orders of magnitude is unprecedented. Also, even if this technique has terrible qubit-qubit interaction properties, it would still be useful for storage. You could just read the qubit out, do computation in a different medium, and store the result back.
Re: 'Unprecedented' discovery could propel quantum computers to reality
#35Can someone maybe describe the details of "could propel" and what exactly "reality" mean? I've always been told "when quantum computers become reality you can forget about everything you learned on computer science". For me reality would be that one can use it in production, even if it's a complete niche. In other words is it "we got a step closer" or "we potentially solved our only (major) problem"?
As far as we know, its definitely just "we got a step closer". The advances that we will need to make about our understanding of quantum systems are probably significant, but quantum computers themselves don't seem that promising as of now for one important reason - we have had a very hard time coming up with algorithms that run faster on quantum computers (asymptotically) compared to classical computers. The only significant ones right now are Shor's algorithm (factoring numbers in log(n)^3 time) and Grover Search (searching through N unsorted values in sqrt(n) time). So it's definitely a niche for now, but with two qualifications:
1. Search is a very important niche, it comes up everywhere and could effectively let us say "fuck it, lets brute force it" in a lot of places where we currently use complex algorithms.
2. The knowledge that we gain while trying to build one will probably be useful somewhere.
As for "could propel". There is a set of five criteria (or 7 if you consider networking) called the DiVincenzo criteria that roughly state the properties a system needs to satisfy to be a viable quantum computer. One of these is "long decoherence times compared to gate operation times", which means how long the system of qubits can store quantum information before the noisy environment collapses it to classical information. We compare this to gate operation times because the relative times are what is important. If you have a decoherence time of 2 seconds, but it takes 3 seconds for a CNOT gate to operate, its not very useful. I don't know about the gate operation times of quantum computers based on diamonds. Even so, decoherence times are usually on the order of ms, so a 2 second time is probably a very significant achievement in that criterion. That's also been one of the difficult criteria to crack, so that adds to it. Additionally, theres no need to use this medium as the "RAM" or cache of a computer, this could potentially be used as the equivalent of a HDD on a quantum computer.
Re: 'Unprecedented' discovery could propel quantum computers to reality
#36Earlier quoted context omitted.
> Am a bit frustrated reading that same description for years and still having no clue what it really means. Both values at the same time... huh? What is the physical property that is existing in two states at once? Ever heard of this quantum mechanics thing?
You act like quantum mechanics is an easy thing to understand.
Well, other people act like monads are an easy thing to understand. YMMV.
Re: 'Unprecedented' discovery could propel quantum computers to reality
#37Earlier quoted context omitted.
Think of it like this. What if you represented your computer bits as a arrow on a compass. You have also the plane that the compass is on. The arrow points in one direction for 1 and in another direction for zero. Regular computation would have you doing operations on bits. You could only do one operation at a time on one bit and you are able to flip them given a recipe. A probabilistic computer allows you to do some…
Superposition can be described as follows. You can correlate two qubits together magically. That's entanglement not superposition.
Yea, sorry. I haven't studied quantum computing so I am trying to do a plain English explanation to further my understanding.
I should have described entanglement as the correlation of two qubits at after their interaction when they are separated.
Superposition should have been described as both particles being able to take any state in a set of two spherical vectors.
I will correct this when I explain it further. Thanks sp332!
Re: 'Unprecedented' discovery could propel quantum computers to reality
#38This I believe is how certain aspects of the mind exists, information being temporarily stored and interactive with other systems. Theoretically then systems of the brain/mind should be able to be duplicated. This would allow for managing of large amounts of information, at very high speed, with no-heat generation. The only heat generated would be through the support systems / foundation - I suppose much like how the…
It reminds me of scifi movies where computers are made of crystals
Re: 'Unprecedented' discovery could propel quantum computers to reality
#39Can someone maybe describe the details of "could propel" and what exactly "reality" mean? I've always been told "when quantum computers become reality you can forget about everything you learned on computer science". For me reality would be that one can use it in production, even if it's a complete niche. In other words is it "we got a step closer" or "we potentially solved our only (major) problem"?
Quantum computers are used in production, as you say in a complete niche. Lockheed Martin bought a quantum computer (and associated research & development) from D-Wave for use in their "most complex computations". [1] There's no further notice of what Lockheed has done with it, but at least there's one quantum computer in private hands designed to do production work. [1] http://www.marketwire.com/press-release/d-wave…
in my understanding the jury is still out on whether D-Wave actually has a quantum computer or not, e.g. http://www.scottaaronson.com/blog/?p=954
Re: 'Unprecedented' discovery could propel quantum computers to reality
#40"For qubits, they can hold a value of “1” or “0” as well as both values at the same time. Described as superposition,"... Am a bit frustrated reading that same description for years and still having no clue what it really means. Both values at the same time... huh? What is the physical property that is existing in two states at once? And how does that feature translate into speedier computation? Is it so hard to unde…
Michael Nielsen explains why explaining quantum comouting is inherently difficult (TL;DR our intuition works classically and thus a complete classical model of a quantum computer would imply that a quantum computer could be simulated classically) http://michaelnielsen.org/blog/quantum-computing-for-everyon...