I'm trying to restrict the problem, but for now it seems that with newer numpy versions on x64 certain complex products return different results depending on whether the operands are wrapped in a ...

Abstract: We deal with accurate complex multiplication in binary floating-point arithmetic, with an emphasis on the case where one of the operands is a "double-word ...

Add a description, image, and links to the complex-multiplication topic page so that developers can more easily learn about it.