fixed. thanks, russ. > In common.h, the results of the new dFMod() and dCopySign() > macros are getting cast to float even in the dDOUBLE case. > This makes my double-precision test suite very angry. :) -- Russell Smith http://www.q12.org