no, you can "add" Huffman coding to this algorithm to get better compression (but it will be much slower). What I used is a modified version of the LZ algorithm

no, you can "add" Huffman coding to this algorithm to get better compression (but it will be much slower). What I used is a modified version of the LZ algorithm family, but instead of a table assembled from the data I'm compressing I use a fixed one. It's basically a simple table-lookup, with two special bytes needed to emit verbatim bytes.