Cosmology Notebooks

CAMBvsCLASS

This is a quick and dirty Jupyter notebook that compares reference spectra output for CLASS and CAMB. At the highest precision settings, they agree at a 0.1% level up to $\ell \sim 3500$. The linear matter power spectra agree to ~0.01%, and HALOFIT nonlinear P(k) agree up to ~0.1%.

I'm particularly interested in the TT spectrum for $3000 < \ell < 5000$. For these purposes, the key parameter appears to be k_max_tau0_over_l_max.