User blog comment:B1mb0w/Omega Count/@comment-5529393-20160712114533/@comment-10262436-20160712114955

This method gives me a short cut to calcuate equivalent T functions for various ordinals (up to a point). For example I can calculate the exact T function for epsilon_0 for n=2,3,4 etc.

Unfortunately, beyond this level the method is difficult to follow and I agree with you that it can lead to errors.