Abstract We argue here that models of SN Ia based on gravitational collapse with the ejection of only a small amount of matter and nickel will be subluminous, and are likely to have problems with the light curve and especially with the spectral evolution. Canonical thermonuclear explosion models can account for the light curves not only in the optical but in the IR and for the spectral evolution. Some classes of thermonuclear models can give adequate kinetic energy of expansion with a range in nickel mass, thus accounting for both normal and subluminous SN Ia. The subluminous models may form interesting amounts of CO in the ejecta. We discuss the comparison of SN Ia,b, and c. SN Ia have an ejecta rich in high opacity iron. SN Ic have comparable ejecta mass, and somewhat lower opacity C O dominated ejecta. SN Ib probably have the highest mass, but are composed largely of low opacity He. We argue that SN Ic are very deficient in both H and He, but that SN Ib may have small, but detectable, amounts of H in their outer layers. There is general agreement that some form of convection is critical for proper understanding of the explosion mechanism of gravitational collapse. There is still a healthy debate about how the convection proceeds and whether it is the universal cure.