Quantitative agreement between experimental and calculated gain/absorption or refractive index spectra can only be found if processes that lead to the homogeneous broadening of the spectra (electron-electron and electron-phonon scattering) are treated on a fully microscopic level. Aproaches that use phenomenological dephasing times or lineshape broadenings can only fit experimental results. They lead to unphysical results like absorption energetically below the bandgap, wrong lineshapes and density-dependences and their ``reliability'' is limited to a small range of situations (densities, temperatures, material compositions, well widths, etc.) close to the ones for that experimental data already exists.

Likewise, wafer-level PL/PLE spectra cannot be matched without including these many-body ingredients, effects like the Coulomb-induced subband coupling and conduction band nonparabolicity and, a self-consistent inclusion of internal dopant fields due to p- and d-doped layers in the heterostructure.

Including all these ingrediences leads to a theory that can quantitatively predict the optical material characteristics for an almost unlimited range of situations: from zero density absorption to high density gain, for arbitrary temperatures, including internal or external electric fields, single or multi-quantum-wells, GRINSCH or SCH structures, electronically coupled or uncoupled wells, TE or TM polarization, internal strain or external pressure, etc., and all commonly used semiconductor materials. This also allows to determine quantitatively important operating characteristics like the high-density gain from low density PL as it is measured on an on-waver stage for a few excitation densities (without having to know the actual experimental densities).