This archive contains two procedures for calculating squareroots of 8 an 16 bit numbers. The result is rounded, so that the precision is +-0.5 ! These are named sqrt8 and sqrt16. They were written with CC65 version 2.10.0 and you should use this version, because i made extensively use of : labels. There is an example program, colcyc.s, which makes use of sqrt16. There is another routine included into the archive, which calculates squares iterately. It simply starts with one square number and then adds the right odd number to get to the next square. There is also a test program sqrt.c, which prints out different numbers and corresponding square roots and asserts, that the result is correct e.g. lies in the precision range. There are no changes planned, so version is 1.0. But who knows, if some optimizations can be done.