We describe a general strategy for sampling configurations from a givendistribution, NOT based on the standard Metropolis (Markov chain) strategy. Ituses the fact that nontrivial problems in statistical physics are highdimensional and often close to Markovian. Therefore, configurations are builtup in many, usually biased, steps. Due to the bias, each configuration carriesits weight which changes at every step. If the bias is close to optimal, allweights are similar and importance sampling is perfect. If not, ``populationcontrol" is applied by cloning/killing partial configurations with too high/lowweight. This is done such that the final (weighted) distribution is unbiased.We apply this method (which is also closely related to diffusion type quantumMonte Carlo) to several problems of polymer statistics, reaction-diffusionmodels, sequence alignment, and percolation.