A detailed study of two commercial turbocharger compressor impellers with a parallel wall vaneless diffuser is reported. The study shows that present turbocharger compressor maps from different sources cannot be reliably compared as random errors can amount to ±0.05 (5 points) on efficiency (20:1 odds). Vaneless diffuser traverses reveal a flow field with backflow and separated regions. The quantitative traverse results reveal the level of uncertainty existent today in such basic deduced parameters as slip factor, backflow loss, rotor efficiency and diffuser recovery. A fully bladed and cutback bladed impeller are compared. Further work is outlined.