HP Prime: Dirichlet's eta function recognized but not numerically evaluated

[Helge Gabert]

I know that the Prime's help screen says that zeta(a) should only be evaluated for real a, but I couldn't help trying out some complex values, like zeta(1-i), which returns the expression in terms of the Dirichlet eta function

eta(1-i)/(1-2^i),

which is great, (recognizing this relationship), but the eta function apparently cannot be approximated numerically (or can it?), and does not show up in the catalog. Any insights? Thanks.

P.S. Other complex arguments can be approximated by zeta(), e.g., zeta(0.5-i).