My own article gave me an idea on how to optimize the 6502 16-bit multiplication with 28% speed-up, and the calling code speed is enhanced as well. Available in the CVBasic repository https://github.com/nanochess/CVBasic