[Octopus-users] bug in spherical harmonics?

[Eugene Kadantsev]

Hi, 

I'm using Octopus-1.1. 

I think there is a bug in the routine which computes 
real spherical harmonics for l > 2: ylm.c in liboct.  
For l > 2, the normalization 
factor \sqrt{2} is missing. I verified it by integrating spherical 
harmonics over the spherical box of R=10 a.u. with weigth 1: for l <=2 
I get approximately 10^3/3 = 333 but for l > 2 I get only 166...


Regards,
Eugene Kadantsev