Keep in mind that the SAM results are quite different than the MTMF results. SAM can tell you which of your input training classes a given pixel is most like. But with MTMF, the results tell you how much of each of the input materials exists within each pixel. So there is not a single best answer for which material represents each pixel. You could probably make an image that color codes each pixel according to which material has the highest abundance in each pixel. But that defeats the purpose of using the MTMF method, which is to return abundance values. That is why ENVI does not create a "classification" output for MTMF. If you just want one material identified for each pixel, then you are probably better off using the SAM method, which is designed for that type of analysis. Regards, Peg