Re: Bitcoin challenge transaction: ~100 BTC total bounty to solvers!
By the way, isn't it kinda scary that a few keys in 80 and 90 bit ranges were already found? Many users have passwords with less bits of security, and passwords also aren't random, so bruteforce can be optimized. Or am I missing something, and those keys were cracked with some algorithm rather than brute force?
you are missing the difference between Asymmetric (elliptic curve) cryptography and Encryption (like Block ciphers). i am not an expert but the algorithms are very different. for example in this topic's case the algorithm (when starting from a certain private key like 2^90) can be simplified to a very simple single point addition but in breaking something like AES you still need to brute force the whole thing.Quote
if you want to have fun, you can program something. i did that a while ago in c# for small ones as a "challenge" to my skills knowing there is no other reward.
So, I did it and it takes 2.5 ms for me to calculate an address from a raw private key
doesn't seem that bad. my normal point multiplication takes 42 ms using BigIntegers...
that's not the whole thing though, it is all about the LOOP where you have to repeat the same thing, and you shouldn't work with raw private key and addresses. you start with the bytes and end with hash bytes and skip any unnecessary conversions.
It is unfortunate that challenges like these are so complicated and that only a few people with specialized hardware and knowledge can take part in this. We need challenges like this that can give open participation to a lot more people, without specialized knowledge to garner a lot more publicity for Bitcoin.
I know it is difficult to create something that would be more accessible for the average guy on the street, but it can be something that can be done by a group of people with different knowledge on several topics and something where groups can all benefit from the rewards. We had something like this before, but the prize was not this big.
this is not a giveaway. i believe it is more like a test of the security of a 256 bit elliptic curve, and the incentive is the way to keep an eye on it. the conclusion so far is that 256 is very safe since so far after years of work only ~90 bits of it is searched and the "rewards" were claimed. because of that it is not for the "average guy" while that average guy can still start learning more about ECC and gets his hands dirty...I know it is difficult to create something that would be more accessible for the average guy on the street, but it can be something that can be done by a group of people with different knowledge on several topics and something where groups can all benefit from the rewards. We had something like this before, but the prize was not this big.