4 ms·

Silicon quantum processor with robust long-distance qubit couplings

lisper 9y ago
This paper has made me significantly increase my estimate of the urgency of moving to post-quantum crypto algorithms. Which really pisses me off because I spent a long time learning about elliptic curves, and soon all that knowledge will be useless. Damn those physicists.
- sp332 9y ago
  You'll just have to upgrade to supersingular elliptic curve isogenies! https://en.wikipedia.org/wiki/Post-quantum_cryptography#Supersingular_elliptic_curve_isogeny_cryptography https://en.wikipedia.org/wiki/Post-quantum_cryptography#Supe...
  - lisper 9y ago
    Cool, but do they work for signatures?
    - sp332 9y ago
      No idea if there are production-ready implementations, but in theory, yes. https://en.wikipedia.org/wiki/Supersingular_isogeny_key_exchange#Similar_systems.2C_signatures.2C_and_uses https://en.wikipedia.org/wiki/Supersingular_isogeny_key_exch...
      - TCM 9y ago
        Microsoft seems to have a github repo with the algorithm. https://github.com/Microsoft/PQCrypto-SIDH https://github.com/Microsoft/PQCrypto-SIDH
        ktta 9y ago
        That's still the key exchange code, and not the signature. Microsoft hasn't released any code related to the SDVS extended by SIDH papers linked to the Wikipedia page, because the papers aren't by Microsoft.
joeschmoe3 9y ago
Could I Get the " Explain it like I am 5 TL;DR of this "
- marcosdumay 9y ago
  The researchers invented a way to construct quantum computers that is both less prone to noise (that limits the computer size), easier to set and measure, and easier to manufacture when compared to the state of the art. They didn't build such structure in practice, but got some very comprehensive simulations.
zardo 9y ago
It wasn't clear to me from the abstract, does this allow a complete CPU to be constructed using industry standard tools, or are there still some special or non-scalable steps required? Will the 4004-Q be ready in time for the 2019 holiday shopping season?
- femto 9y ago
  A layman's explanation in the press release: https://newsroom.unsw.edu.au/news/science-tech/flip-flop-qubits-radical-new-quantum-computing-design-invented https://newsroom.unsw.edu.au/news/science-tech/flip-flop-qub... Edit: Also being backed by a new company: "Silicon Quantum Computing Pty Ltd". https://newsroom.unsw.edu.au/news/science-tech/australia%E2%80%99s-first-quantum-computing-company-launches-unsw https://newsroom.unsw.edu.au/news/science-tech/australia%E2%...
deepnotderp 9y ago
The use of the phosphorus donor is interesting. I haven't read it through totally, but it reminded me of this hypothesis: https://www.quantamagazine.org/a-new-spin-on-the-quantum-brain-20161102/ https://www.quantamagazine.org/a-new-spin-on-the-quantum-bra... The other thing that I didn't see is a mention of the temperature, although I presume it may be around room temperature based of previous work with donor-spin qubits. Also note that this isn't an experimental demonstration, this is a theoretical treatment with simulations. Nothing wrong with that, but there was a lot of brouhaha over non-Abelian anyons for topological quantum computing a bit back, but those weren't experimentally demonstrated to even exist. Now, to be fair, these are concrete simulations and are definitely different from the non-Abelian anyon situation because they don't require the existence of particles which haven't been experimentally demonstrated, and IIRC spin-donor qubits have been demonstrated before experimentally, but do know that there can be unexpected pitfalls. For example, in theory QAHE state materials should exhibit no backscatter theoretically, but experimental measurements instead suggest that there is a small amount of backscatter, although the mechanism remains unknown. Nonetheless, hopefully exotic physics in conventional materials can push forward EE devices in the future!
joejerryronnie 9y ago
Sweet! In 18-24 months "quantum computing" will be the new hotness in tech. We've reached peak "Machine Learning" and "Autonomous Vehicles" are yesterday's news. Get in on the ground floor of the next wave of VC handouts! Hit Buck's and casually mention within earshot of the nearest tables that you're working on a quantum block chain project that will revolutionize programatic ad delivery but you're forgoing the hassle and red tape of traditional venture funding and planning your ICO in a couple of weeks - and then place bets on which will hit your table first, a term sheet or the lunch bill.

This paper has made me significantly increase my estimate of the urgency of moving to post-quantum crypto algorithms. Which really pisses me off because I spent a long time learning about elliptic curves, and soon all that knowledge will be useless. Damn those physicists.

sp332 9y ago
You'll just have to upgrade to supersingular elliptic curve isogenies! https://en.wikipedia.org/wiki/Post-quantum_cryptography#Supersingular_elliptic_curve_isogeny_cryptography https://en.wikipedia.org/wiki/Post-quantum_cryptography#Supe...
- lisper 9y ago
  Cool, but do they work for signatures?
  - sp332 9y ago
    No idea if there are production-ready implementations, but in theory, yes. https://en.wikipedia.org/wiki/Supersingular_isogeny_key_exchange#Similar_systems.2C_signatures.2C_and_uses https://en.wikipedia.org/wiki/Supersingular_isogeny_key_exch...
    - TCM 9y ago
      Microsoft seems to have a github repo with the algorithm. https://github.com/Microsoft/PQCrypto-SIDH https://github.com/Microsoft/PQCrypto-SIDH
      - ktta 9y ago
        That's still the key exchange code, and not the signature. Microsoft hasn't released any code related to the SDVS extended by SIDH papers linked to the Wikipedia page, because the papers aren't by Microsoft.

joeschmoe3 9y ago

Could I Get the " Explain it like I am 5 TL;DR of this "

marcosdumay 9y ago
The researchers invented a way to construct quantum computers that is both less prone to noise (that limits the computer size), easier to set and measure, and easier to manufacture when compared to the state of the art. They didn't build such structure in practice, but got some very comprehensive simulations.

zardo 9y ago

It wasn't clear to me from the abstract, does this allow a complete CPU to be constructed using industry standard tools, or are there still some special or non-scalable steps required? Will the 4004-Q be ready in time for the 2019 holiday shopping season?

femto 9y ago
A layman's explanation in the press release: https://newsroom.unsw.edu.au/news/science-tech/flip-flop-qubits-radical-new-quantum-computing-design-invented https://newsroom.unsw.edu.au/news/science-tech/flip-flop-qub... Edit: Also being backed by a new company: "Silicon Quantum Computing Pty Ltd". https://newsroom.unsw.edu.au/news/science-tech/australia%E2%80%99s-first-quantum-computing-company-launches-unsw https://newsroom.unsw.edu.au/news/science-tech/australia%E2%...

deepnotderp 9y ago

The use of the phosphorus donor is interesting. I haven't read it through totally, but it reminded me of this hypothesis: https://www.quantamagazine.org/a-new-spin-on-the-quantum-brain-20161102/ https://www.quantamagazine.org/a-new-spin-on-the-quantum-bra... The other thing that I didn't see is a mention of the temperature, although I presume it may be around room temperature based of previous work with donor-spin qubits. Also note that this isn't an experimental demonstration, this is a theoretical treatment with simulations. Nothing wrong with that, but there was a lot of brouhaha over non-Abelian anyons for topological quantum computing a bit back, but those weren't experimentally demonstrated to even exist. Now, to be fair, these are concrete simulations and are definitely different from the non-Abelian anyon situation because they don't require the existence of particles which haven't been experimentally demonstrated, and IIRC spin-donor qubits have been demonstrated before experimentally, but do know that there can be unexpected pitfalls. For example, in theory QAHE state materials should exhibit no backscatter theoretically, but experimental measurements instead suggest that there is a small amount of backscatter, although the mechanism remains unknown. Nonetheless, hopefully exotic physics in conventional materials can push forward EE devices in the future!

joejerryronnie 9y ago

Sweet! In 18-24 months "quantum computing" will be the new hotness in tech. We've reached peak "Machine Learning" and "Autonomous Vehicles" are yesterday's news. Get in on the ground floor of the next wave of VC handouts! Hit Buck's and casually mention within earshot of the nearest tables that you're working on a quantum block chain project that will revolutionize programatic ad delivery but you're forgoing the hassle and red tape of traditional venture funding and planning your ICO in a couple of weeks - and then place bets on which will hit your table first, a term sheet or the lunch bill.