The general problem of obtaining reliable results from gauge-fixing and projection is discussed. It is shown that the usual form of the maximal abelian gauge gives poor results for the string tension in SU(3) lattice gauge theory. A generalized form is suggested. Evidence is presented that monopoles in SU(3) are associated with SU(2) subgroups, and that P-vortices pass through monopoles, similar to what happens in SU(2).
Categories: hep-lat