Abstract
The basic principles of integral field spectroscopy are reviewed using a figure of merit, the specific information density, as an objective means of comparison. This shows that the best method in theory is image slicing, especially as implemented in the Advanced Image Slicer design used in the Integral Field Unit of the Gemini Near-Infrared Spectrograph. However, an alternative, a hybrid of lenslet array and slicer techniques, may offer a better compromise between theoretical performance and practicality for panoramic spectroscopic surveys requiring millions of spatial elements such as the San Pedro Mártir twin telescope project. Conceptual designs of this type already exist, including one aimed at the IMACS spectrograph on Magellan.
