Is matrix exponential function (expm) problem solved or how to solve?

As mentioned in Cleve's blog A Balancing Act for the Matrix Exponential, Is this kind of problem solved for expm funcation? I tested the case in MATLAB 2012 and 2013, they still generated the same wrong results written in Cleve's blog.

Answer 1

[Walter Roberson]

R2015b has improvements in expm() and related routines such as logm()