A video coder (encoder or decoder) determines a search region (SR) relative to a current block and determines, based on a sampling interval, a plurality of sampling positions in a plurality of sampling rows of the search region. Sampling positions of the plurality of sampling positions in two adjacent sampling rows of the plurality of sampling rows are in different columns of the search region. The coder further determines a plurality of template matching costs for a plurality of respective candidate reference blocks (RBs) at the plurality of respective sampling positions, determines an RB based on the plurality of template matching costs, and determines a prediction block of the current block based on the RB.