HP50G Bisection methode - Printable Version +- HP Forums ( https://www.hpmuseum.org/forum)+-- Forum: HP Calculators (and very old HP Computers) ( /forum-3.html)+--- Forum: General Forum ( /forum-4.html)+--- Thread: HP50G Bisection methode ( /thread-4422.html) |

HP50G Bisection methode - tigger - 07-24-2015 12:01 PM
I would like to program the Bisection Methode. Is there anybody who can help me with it. I am an absolute beginner. I have the function y=x^5-5. I assume the result between 1 and 2: We first define the function F(X). Now save 2 in a variable called B and 1 in a variable called A. We have to find the midpoint between A and B, our new approximation, C. Can anbody help me to go on. I would like to program other easy stuff. RE: HP50G Bisection methode - Gilles - 07-24-2015 04:35 PM
(07-24-2015 12:01 PM)tigger Wrote: I would like to program the Bisection Methode.Hi It seems there is a problem with your example as there is no root for this function between 1 and 2 here is an example (not optimised), in RPL, aprox mode with a modified function (X^5-5) : Code: `'F(X)=X^5-5' DEFINE` RE: HP50G Bisection methode - Namir - 07-29-2015 11:31 AM
(07-24-2015 12:01 PM)tigger Wrote: I would like to program the Bisection Methode. The root for y=x^5 (or for any y=x^n) is zero, not in the interval [1, 2]. You can start with an interval of, say, [-1, 1]. Namir RE: HP50G Bisection methode - Namir - 07-29-2015 11:37 AM
(07-24-2015 04:35 PM)Gilles Wrote:(07-24-2015 12:01 PM)tigger Wrote: I would like to program the Bisection Methode.Hi Your code assumes F(A)<0 and that's risky, because if F(A)>0 your implementation will not work. The traditional Bisection algorithm uses the test F(A)*F(C)>0 instead to determine if we set A=C (when that condition is true) and B=C otherwise. I assume the RPL code should be: Code: `'F(X)=X^5-5' DEFINE` Namir |