--- Day changed Fri Aug 28 2015 11:51 <@andytoshi> so, i was able to write a jacobi symbol impl from the wiki page which was correct the first time in ten minutes ... unfortunately it takes 60x the time that gmp does :) 11:52 <@andytoshi> even though i cheated and assumed my inputs were coprime 12:02 <@andytoshi> ah, the cheat gained me almost nothing. (wiki claims i had to check whether the inputs were coprime; cohen says just check if one of the inputs ever hits zero) 12:23 < gmaxwell> andytoshi: hah. 12:24 < gmaxwell> andytoshi: so sipa was telling me he's made tremendos progress with the group law verification. 12:26 < sipa> yeah, i hope to do a few of the hard ones today :) 12:27 < sipa> basically, the system allows verifying that a ratio of polynomials jacobian formula behaves the same as an affine one, including infinities, but excludes points where either of the sides results in a 0/0 12:31 <@andytoshi> sipa: really awesome 12:32 <@andytoshi> i have always been a tiny bit nervous that more bugs like that (A - lambda*A) one are somewhere 12:33 < sipa> now, these are polynomials in Q 12:33 < sipa> not in Zp 12:33 < sipa> maybe i need to explicitly add a rule that a number lambda exists with certain properties 12:34 <@andytoshi> i think Q is fine 12:34 <@andytoshi> both Zp and Q are fields so 0/0 is 0/0 regardless 12:34 <@andytoshi> oh, i see, there is e.g. no nontrivial cube root of 1 in Q 17:03 -!- jtimon [~quassel@c-24-4-96-213.hsd1.ca.comcast.net] has joined #secp256k1 20:30 -!- jtimon [~quassel@c-24-4-96-213.hsd1.ca.comcast.net] has quit [Ping timeout: 245 seconds] 22:15 -!- jtimon [~quassel@c-24-4-96-213.hsd1.ca.comcast.net] has joined #secp256k1 23:41 -!- jtimon [~quassel@c-24-4-96-213.hsd1.ca.comcast.net] has quit [Ping timeout: 250 seconds]