## #math - Thu 12 Apr 2007 between 01:55 and 01:59

### NY Lost Funds

 HiLander mot_: line 27 is wrong mot_ see how i set n to the return value of int() ?how is 27 wrong? thermoplyae Maybe 7 is too far out on the Taylor seriesTry something like .25 HiLander also, i'm not seeing anythign wrong with intermediate computations mot_ i know int i and inc() are the same, but like i said it's eventually going to be a recursive function HiLander thermoplyae's comment is probably dead on mot_ well i just have to be able to compute sin(x) thermoplyae I love being dead on HiLander then you better go farther than 5 mot_ i'm testing it against the cmath library sin() function and my numbers are wildly, wildy off thermoplyae Please try .25 :( HiLander you certainly need that each term is less than 1, so you need (2n+1)! > X^(2n+1) ailndx -1^n * X^(2n+1)/(2n+1)! is that some sine approximation formula? thermoplyae ailndx: Taylor series whaleofconfusion the problem is you are setting the result to Xas HiLander has been saying mot_ w00ti fixed it (by doing my iterations of the series) whaleofconfusion also that your inc function is stupid mot_ i was just doing 0-5, 0-10 gives a better (almost 99% accurate) approximation thermoplyae I love being dead on mot_ whaleofconfusion, yes, i know, but it will be useful when i re-write the function recursively. JabberWalkie yeah!...stupid. whaleofconfusion why would you want to write it recursively mot_ because i have to(because my professor is douche) JabberWalkie good reason HiLander no, it makes senseit allows for precision to be handled on the fly whaleofconfusion and you should not use the inc function HiLander instead of ahead of time whaleofconfusion you should just use i thermoplyae hmm whaleofconfusion C++ is not made for recursion thermoplyae Seems like it would be faster to just shift the argument over into [-pi, pi] and then use the method he has now mot_ amazing, it worksi just needed to do more iterations thermoplyae Rather than decide he needs to go out to the 101th term