2
Lets assume I keep generating random public keys with their matching private keys and never use them. Severals years later I am checking for positive match against the blockchain for all my generated public keys. If a match is found that means someone is actually using a public key I generated in the past and for which I know the private key. I should then be able to gain control over someone else wallet even in the case of a paper or brain wallet. Best regards,
2
Thats right, if you got the private key you got access to the bitcoins on it. You dont even need to wait until someone starts using this key, you could also randomly find a key which is already in use by someone.
BUT:
Bitcoin addresses are 160 bit hashes of the 256 bit private keys, so there are 2^160 or about 1.460.000.000.000.000.000.000.000.000.000.000.000.000.000.000.000 possible addresses. You will notice thats the possibility is pretty much nil.
For more information you can read this article: http://www.miguelmoreno.net/bitcoin-address-collision/
Targeting a single address would be a waste of lives but in the case of a global multiple random targets. I was wondering how much time, computational power, and recording space would be needed to find all possible addresses. – isaac – 2014-07-02T16:09:48.323
The keys have to be generated randomly for those odds to apply. – None – 2014-07-02T16:10:34.643
i'm not sure, didn't benchmarked it, but as satoshi posted once, its always more efficient to mine blocks than create random address – Dennis Kriechel – 2014-07-02T16:25:22.637
For how long would this be true goes back to my first comment, right? – isaac – 2014-07-02T16:48:40.373
not exactly sure what you mean, but its should be true forever, after the blockreward has gone its replaced by transactions fees which should increase over time and so this should be always more efficient. And mining is done with ASICs which cant be used for creation keys so they cant just swap their power for that. – Dennis Kriechel – 2014-07-02T16:58:43.987