Dear Eigen Team and Users,


I'm new here and I have a newbie question. I have implemented a version of Block Conjugate Gradient, to resolve AX=B where A(nxn), X(nxp) and B(nxp). Then, I compare the walltime of this block version (using p=1) agains Classics CG using just VectorXd. Result: the block version took more time.

Specifically I see that the operation A*V, with V defined as MatrixXd V(n,1) took more time than the operations A*v, with v defined as VectorXd v(n). For example, with a sparse matrix with n=14800, the kernel A*V took 5.4 seconds and A*v took 3.8 seconds, both resolve the problem in 2400 iterations.

My questions, is this an expected result?, if yes why?... can I do something to get approximately more closer results?.

Thanks in advance.

Best Regards

Pedro Torres

Can you show some piece of code, because there should not be much difference between VectorXd and a column major MatrixXd for CG.