We describe several projects addressing the growth of galaxies and massive black holes, for which adaptive optics is mandatory to reach high spatial resolution but is also a challenge due to the lack of guide stars and long integrations. In each case
kinematics of the stars and gas, derived from integral field spectroscopy, plays a key role. We explain why deconvolution is not an option, and that instead the PSF is used to convolve a physical model to the required resolution. We discuss the level of detail with which the PSF needs to be known, and the ways available to derive it. We explain how signal-to-noise can limit the resolution achievable and show there are many science cases that require high, but not necessarily diffraction limited, resolution. Finally, we consider what requirements astrometry and photometry place on adaptive optics performance and design.
