The Gaussian β-ensemble is a real n-point configuration {x j } n 1 picked randomly with respect to the Boltzmann factor e − β 2 Hn ,It is well known that the point process {x j } n 1 tends to follow the semicircle law σ(x) = 1 2π (4 − x 2 ) + in certain average senses.A Fekete configuration (minimizer of H n ) is spread out in a much more uniform way in the interval [−2, 2] with respect to the regularization σ n (x) = max{σ(x), n − 1 3 } of the semicircle law. In particular, Fekete configurations are "equidistributed" with respect to σ n (x), in a certain technical sense of Beurling-Landau densities.We consider the problem of characterizing sequences β n of inverse temperatures, which guarantee almost sure equidistribution as n → ∞. We find that a necessary and sufficient condition is that β n grows at least logarithmically in n:We call this growth rate the perfect freezing regime. Along the way, we give several further results on the distribution of particles when β n log n, for example on minimal spacing and discrepancies, and that with high probability a random sample solves certain sampling and interpolation problems for weighted polynomials. (In this context, Fekete sets correspond to β ≡ ∞.)The condition β n log n was introduced in earlier works due to some of the authors in the context of two-dimensional Coulomb gas ensembles, where it is shown to be sufficient for equidistribution. Interestingly, although the technical implementation requires some considerable modifications, the strategy from dimension two adapts well to prove sufficiency also for one-dimensional Gaussian ensembles. On a technical level, we use estimates for weighted polynomials due to Levin, Lubinsky, Gustavsson and others. The other direction (necessity) involves estimates due to Ledoux and Rider on the distribution of particles which fall near or outside of the boundary.