The encoding of vector quantization (VQ) needs expensive computation for searching the closet codeword to the input vectors. In order to reduce computational burden, many researchers have developed some efficient full-search-equivalent algorithm by using the characteristics of the means and variances of a vector. However, some computational redundancies still exist in them. In this paper, we introduce a technique to efficiently partition a vector into two dynamic subvectors according to the patterns inside the block (or vector) and then further improve the computational load. Experimental results show that the proposed method provides a higher codeword searching efficiency than other existing algorithms.