The rapid growth of 3-D-sound technology has created a need for 3-D audio tools which can be used by musicians and composers in much the same manner that visual artists use 3-D graphics tools. Experience with such audio tools reveals some issues of special concern for music. For example, the rapid movement of musical sound sources can create Doppler shifts that produce harsh detunings of pitch, and the realistic rendering of intensity loss with changing distance can cause some musical elements to be buried in the mix. Consider too that individual instruments are best spatialized in different environments—high strings in large reverberant halls, electric basses in small dry rooms—and that some musically useful spatial effects, like stereo decorrelation, are not conveniently produced with environmental models. Spatial sound rendering software needs to provide numerous exceptions to accurate physical modeling in order to adapt to the musical context and it must support a heterogeneous collection of musically useful tools. In fact, the true goal in the design of 3D software for music should be to provide a creative tool that captures the complexity and richness of spatial sound without slavishly adhering to models of acoustic reality.