William Jockusch

This fixes your issues about accuracy and also about GammaLn with negative inputs. I'm choosing to return NaN for GammaLn in cases where the Gamma function is negative. But it...

Simple -- probably. Easy -- probably not. For example, opensource.apple.com has an implementation that looks quite simple, but it appears to have gone through several generations of tweaking, and it...

It seems that Apple took their algorithm from netlib originally, and it has since been used by people all over the world. A Google search on the number 6.64561438202405440627855 is...

OK, it looks to me like it would be worth using. I made some changes to the public domain code, which appears to have been written when the precision of...

I took the John D. Cook one, then tested it at a bunch of values where I thought there was a significant risk of being wrong. It was pretty far...

I just downloaded Math.Net numerics, and asked it to calculate SpecialFunctions.Gamma(5). The correct answer is 24, but it returned 24.000000000000114. Can this be fixed? (My proposed changes would fix this....

I've got a fix for the accuracy issue, by simply using more terms of the power series and moving the lower breakpoint. I'll post once I've had a look at...

I've just posted a version that addresses both of your issues. I'm returning NaN for GammaLn in cases where Gamma is negative. But it's unclear if that's the right choice;...

I am glad to see this issue. I had never noticed WPFMath until this issue came along. About different front ends, yes, my intention is that it should be possible...

Also, I don't think different languages in the core and front end projects should be a problem.