When do you think Quantum Computers will be out?

[dav1d]

When do you think Quantum Computers will be out? Most scientists think 20-30 years but we have them in the lab already, they're just 1.5mill and only about 6 qubits.

 

[xts]

They are already out. A month ago first 128-qubit commercialy available machine was sold for 10M$. Is it worth its price - what applications may it be really used for?

Quantum computers are not as universal like 'classical' ones. You can't write 100,000 lines of C++ code. So if the algorithm (AQO for this box) is designed for optimisation, it doesn't mean you really may find a minimum of any function you like.

Great question: what this D-Wave machine is really able to compute?

 

[Cosmo Novice]

They are already out. A month ago first 128-qubit commercialy available machine was sold for 10M$. Is it worth its price - what applications may it be really used for?

Quantum computers are not as universal like 'classical' ones. You can't write 100,000 lines of C++ code. So if the algorithm (AQO for this box) is designed for optimisation, it doesn't mean you really may find a minimum of any function you like.

Great question: what this D-Wave machine is really able to compute?

This "quantum" computer has been built by a 55man company in a short time period. The Dwave machine has quantum elements and using quantum modelling in its algorithms but there is now way on Earth that is a fully operational 128 Qbit computer.

Currently top labs in the world struggle to entangle 10 Qbits for brief periods of time - I do not believe Dwave have not only entangled 128Qbits but can now use these effectively for computational purposes. My scepticism is also one shared by the Quantum Computing research field - it can be stresses that D-wave One has never been shown to be opertaional under quantum computing mechanisms - they show the end result of the computations but not the process to get their. To clarify the computations performed for Lockhead & Martin by the D-Wave One could have been performed by a number of high powered supercomputers.

This is not to say process is not being made - just do not believe the hype, remember after all they have a product to sell!

 

[xts]

That's what I expected from D-Wave.
Anyway, even single 10M$ box in a short time for 55 people, deducting expenses, still makes something like 100k$ per head - not terribly bad ;)

10 years ago IBM claimed to fraction 15 into 3*5 (which is a special easy case for Shor's algorithm)
Did anybody make any progress since then? 21 maybe?
Should I feel safe with my 4-bit long RSA signature ?

 

[Cosmo Novice]

That's what I expected from D-Wave.
Anyway, even single 10M$ box in a short time for 55 people, deducting expenses, still makes something like 100k$ per head - not terribly bad ;)

10 years ago IBM claimed to fraction 15 into 3*5 (which is a special easy case for Shor's algorithm)
Did anybody make any progress since then? 21 maybe?
Should I feel safe with my 4-bit long RSA signature ?

The algorithm is actually easy to solve - ordinary computers can do it - its the order finding process that is difficult. So yes you should be ok with your RSA key at least for the foreseeable future!

Quantum computing is still a long way off IMHO - QCs can develop in very specific roles at the moment - but until you have QCs that arent mechanism specific, then application can be limited.

 

[Cosmo Novice]

Just to give people a little more info:

http://www.forbes.com/sites/alexknapp/2011/10/31/lockheed-martin-installs-quantum-computer/"

and referenced article

http://www.scottaaronson.com/blog/?p=639"

Scott Aaronson is something of an authority in this field - his opinion is definitely more authoratative than D-Waves commercial hype!

Hope this helps.