Re: [PEPr] Call for votes on Math::Math_Polynomial
| From: | Keith Palmer Jr. | Date: | Sat, 19 Nov 2005 18:13:19 +0000 |
| Subject: | Re: [PEPr] Call for votes on Math::Math_Polynomial | ||
| References: | 1 2 3 4 | Groups: | php.pear.dev |
| Request: | Send a blank email to pear-dev+get-40432@lists.php.net to get a copy of this message | ||
I didn't mean that I would be releasing a semi-working package. I said I'd be adding features in the future. All packages have had features added to them beyond the feature set found in the first release.
Also, important to note that I *believe* its actually not possible to always find numerical roots for polynomial of degree 5+. There isn't any known algorithm to always find the roots, and I think its actually been proven that there *can't* be an algorithm to do this.
Wikipedia mentions this in the History section here:
http://en.wikipedia.org/wiki/Polynomial
- Keith
colder.ch wrote:
Well, PEAR users are not looking for semi-working packages ;) Implementing PEAR::Math_Numerical_RootFinding for 5+ degree polynomials would also be a nice idea. Keith Palmer Jr. wrote:Yeah I'll be looking into adding features with future releases. I have a semi-working factoring/finding the roots for 1st, 2nd, and 3rd degree polynomials, and 4th degree isn't too much worse than the 3rd degree one. Finding critical points/minimums/maximums is pretty easy too using the derivative, so I'll do this in the future. - Keith Etienne Kneuss wrote:Keith Palmer wrote:I'm also looking into algorithms to find the GCD of the polynomial, and polynomial factoring.I don't think it will me possible without too much overhead. Anyway, that might help you: http://mathworld.wolfram.com/PolynomialFactorization.html http://documents.wolfram.com/v4/MainBook/A.9.5.html http://wseas.org/mastorakis/IEEE_CAS1_Jan.pdf