Approximately 12 months of data from the TOPEX/POSEIDON satellite altimeter mission are analyzed for the major short-period oceanic tides. A harmonic analysis is performed on data captured within bins defined on a deep-ocean grid, which, owing to tidal aliasing considerations, must have a relatively coarse spatial resolution. Our analysis is in terms of corrections to the Schwiderski and Cartwright-Ray models, and it confirms many of the Schwiderski differences previously reported by Cartwright and Ray. Our differences with respect to the Geosat-based Cartwright and Ray model form a sectorial pattern in M2 with high/low differences separated roughly 180¿ in longitude. We suggest that these sectorial errors were most likely induced by Geosat's relatively large orbit error. Comparisons to independent data validate the improved TOPEX/POSEIDON solutions; in situ ''ground truth'' shows M2 RMS differences of 4.10 cm. (Schwiderski), 3.86 cm (Cartwright and Ray), 2.63 cm (this paper). Global rates of energy dissipation confirm earlier estimate for M2, and show improved agreement with satellite tracking studies for K1 and S2. These preliminary exercises confirm that TOPEX/POSEIDON should result in a new generation of improved global tidal models. ¿ American Geophysical Union 1994