Recent content by pjg

There's was a mistake in my previous post. The fix should have been for(r=1;r<=nstate;r++) { V_r = V[r]; Ax = V_r[r].r; Ay = V_r[r].i; norm = sqrt(Ax*Ax + Ay*Ay); Ax = Ax/norm; Ay = Ay/norm; for(s=1;s<=nstate;s++) { element.r = V_r[s].r*Ax +...

Check out http://www.netlib.org/eispack/ All the code is there. And free!

No, I just translated the EISPACK routines used in CH from fortran into C, and use that. It works well, and I don't have to worry about someone struggling to get the code compiled and linked to libraries. Here's the snippet of code I wrote to solve the problem...

Yes, you understand my problem. Your previous post, however, made me realize that the solution is quite simple. I just adjust the phase of each eigenvector column in V so its diagonal elements are real. This doesn't affect the diagonalization of H (since it's an unobservable phase), and...

In my case, I do not have degenerate eigenvalues. So, I guess that means my eigenvalues in H are distinct? My problem is if H is very close to diagonal before I numerically diagonalize it, I expect the numerical diagonalization to give me a V that is close to the identity matrix. More...

I'm wondering if anyone here might have a solution to a problem I've having. This is a Quantum Mechanics problem I'm doing. I calculate a 4 by 4 complex Hermitian matrix (H = Hamiltonian) in a basis where it is not diagonal. I diagonalize it numerically (using eispack) and get eigenvalues...