Hi, I'm quite a beginner when it comes to writing code, but I thought I start learning C++ by writing some code that approximated polynomial roots with the Newton-Raphson method. The program asks for the number of terms in the polynomial, and then stores coefficients and exponents in an array (a bad idea, but I have no idea how to do it in a better/ more efficient way). The program then calculates f(x) and f'(x) (x being a guess), and then evaluates x-(f(x)/f'(x)). This evaluation becomes the new x, and after the number of iterations prompted for are finished, the program pauses. The program seemed to work well enough for the first two iterations, but then it never got another digit of accuracy. I assumed (incorrectly?) that this was because the precision of the numbers were too low to produce a more accurate estimate. I decided to see the results of trying to do bignum arithmetic (and I figured that it would be worth it, because without a bignum arithmetic library it would be worthless to do more than 10 iterations of most polynomial estimates), and then spent the next three hours figuring out how to compile, build, and place the libraries in their correct spots. I then read the documentation, and changed all most of my calculations into functions with mpf numbers. There are a couple of things that are giving me some trouble though. The first would probably be" void mpf_set_default_prec"; if I wanted to set the default precision to 256 bits, what should the argument be? (I know that the documentation did cover this, but no matter how I change the argument, there's always a syntax error) The second is that mpf_mul does not allow me to multiply an mpf number by a long double. The long double I must multiply the mpf number is inside a for loop and changes every time the loop is goes through once. How should I work around this? Finally, I was considering storing the mpf numbers in an array, but I'm not completely sure how to do this. Could someone please explain to me (or just link me) on how to manage custom arrays? Thanks
-- You received this message because you are subscribed to the Google Groups "mpir-devel" group. To post to this group, send email to mpir-devel@googlegroups.com. To unsubscribe from this group, send email to mpir-devel+unsubscr...@googlegroups.com. For more options, visit this group at http://groups.google.com/group/mpir-devel?hl=en.
