Hi all,
on the forum page http://www.groupsrv.com/science/about506648.html one can read the following (i cut out nonimportant parts):
Question: Does anyone know any condition (non trivial) that ensure that the global sections of the tensor product of two sheaves is the tensor product of the global sections?
Answer: If X is not affine then, under sufficiently strong finiteness assumptions on the (co)homological dimensions of everything involved, you can approach this problem using a composite functor spectral sequence (which unfortunately will involve the left derived functors of tensor product and the right derived functors of global sections, so you will need some kind of finite-dimensionality assumption on Tor and H^* for quasicoherent sheaves over this scheme to construct the spectral sequence at all).
Can anybody tell me which is the mentioned spectral sequence?
Thanks!