Will Darksend tx be deanonymized by Quantum computers?

[JuanSGalt]

Assuming that these quantum computers could not break the cryptography that governs crypto-coins.

Could the coin join implementations of privacy, like darksend, be broken by quantum computers in the future?

This is a critizism of darkcoin that I picked up somewhere, curious what you think.

thank you!

 

[oblox]

Your probabilities are 3^# of rounds. You tell me if you can prove without a reasonable doubt who sent what and where. Focus on >8 rounds.

 

[JuanSGalt]

Great, old transactions with less than 8 rounds of darksend would than be vulnerable.

 

[oblox]

Great, old transactions with less than 8 rounds of darksend would than be vulnerable.

Even at 4 rounds, it's 1 in 81. Not without reasonable doubt...

 

[vertoe]

With quantum computers the last thing I would do is deanonymizing darksend transactions.

They break all known encryption, so why bother scanning blockchains when you can have all coins at once?

 

[crowning]

Could the coin join implementations of privacy, like darksend, be broken by quantum computers in the future?

They break all known encryption, so why bother scanning blockchains when you can have all coins at once?

This!

But in case some institution doesn't need the money, but wants to invade your privacy, they still couldn't.

To break Darksend, you don't need to break Darkcoins encryption, you need to parse the blockchain, build a weighted connection-graph of all addresses, compute the probability of all connections which go from and to the address you want, and decide which is most probably the true transaction path.

For this, you need huge amounts of memory, and that's something quantum computer are REALLY bad at.

It's by far more practical to do this with a traditional computer architecture. But you would still need a lot of resources, and your result would (in almost all cases) still be just a probability, not a proof.

 

[moli]

I think we're still able to trace DS transactions with 4-6 rounds and less but with 16/17 rounds now ... Darksend is doing a good job.

 

[Sub-Ether]

With quantum computers the last thing I would do is deanonymizing darksend transactions.
They break all known encryption, so why bother scanning blockchains when you can have all coins at once?

I beg to differ on this Vertoe, quantum computers are only good at solving specific types of encyption based on shor's algorithm for example,

''In contrast, most current symmetric cryptographic systems (symmetric ciphers and hash functions) are secure from quantum computers''
http://en.wikipedia.org/wiki/Post-quantum_cryptography
''The quantum Grover's algorithm can speed up attacks against symmetric ciphers, but this can be counteracted by increasing key size.[7] Thus post-quantum symmetric cryptography does not differ significantly from conventional symmetric cryptography.''


Hash functions are a one way function making it harder, as well, the encyption techniques can be modified to make the hash function highly resistant to quantum computing attacks when they come along(and we're years away yet), so in a quantum mechnical sense this is similar to X11 being resistant to parallel ASIC attacks(in the normal standard classical physics model)

Quantum computers are very specific machines that have been somewhat of an anti climax in the real world, as the data coming in/out is bottle necked, as it is in the standard logical binary on/off form, and this must be converted into the qubit equivalent that the quantum machine can understand, I heard they slower than a normal computer for most application
What you have is 2 totally different types of physics and generations of computers (classical and quantum mechanical) with a very bad interface effectively sandwiched in the middle.

 

[JuanSGalt]

ok. That sounds good, certainly more resilient to this theoretical attack than the question implies.

Now, assuming quantums did come, and some crazy person wants to de-anonymize darksend users, Is there any data available on how many rounds the average user does? or what the cost of going to 16-17 rounds, per tx would be?

Thank you