The prediction of structure at the atomic level is one of the most fundamental challenges in condensed matter science. Here we survey the current status of the field and consider recent developments in methodology, paying particular attention to approaches for surveying energy landscapes. We illustrate the current state of the art in this field with topical applications to inorganic, especially microporous solids, and to molecular crystals; we also look at applications to nanoparticulate structures. Finally, we consider future directions and challenges in the field.