I'll try to keep that in mind, Ralf, but then I'll ask for a bit more explanation in the Trac ticket since "fix Hermite / Gegenbauer evalf" doesn't really say what issue was fixed.
And thank you for fixing it! On Wednesday, June 22, 2016 at 10:45:44 PM UTC-7, Ralf Stephan wrote: > > This is fixed in the upcoming beta5. Maybe you better search trac > before posting about already fixed bugs. > > sage: gegenbauer(2,1,1.) > 3.00000000000000 > sage: gegenbauer(2,1,2.) > 15.0000000000000 > sage: gegenbauer(2,1,0.) > -1.00000000000000 > > > On Thursday, June 23, 2016 at 1:39:43 AM UTC+2, Paul Masson wrote: >> >> This series of values is correct for integer arguments: >> >> >> sage: gegenbauer(2,1,x) >> >> 4*x^2 - 1 >> >> sage: gegenbauer(2,1,0) >> >> -1 >> >> sage: gegenbauer(2,1,1) >> >> 3 >> >> sage: gegenbauer(2,1,2) >> >> 15 >> >> >> But nonzero floats give wrong answers: >> >> >> sage: gegenbauer(2,1,1.) >> >> 0.500000000000000 >> >> sage: gegenbauer(2,1,2.) >> >> 0.875000000000000 >> >> And even worse, gegenbauer(2,1,0.) gives the error: >> >> OverflowError: numeric::inverse(): division by zero >> >> >> Ralf? >> >> >> -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To post to this group, send email to sage-devel@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
