Dear Sigurd, sorry for late reply. The table in ATENA Theory manual Section 2.2.12.3 is not based on any code, but rather on having a best fit. So if you need to use this CC3DNonLinCementitious3 material and want to have values as much as possible to the code values, I would recommend to use the following table for the parameters that are not available in the code and for others use the the code values. The code gives values anyway just for fc, ft, eps_v,t.
Whether to start from mean of design depends on the purpose of the analysis. If you intend to simulate the average response then start from mean value. If you need to consider some safety, i.e. make an analysis for checking ultimate limit state then I would recommend to start from the design value of strength. Otherwise you need to consider some suitable safety format for nonlinear analysis. For more description about this I can refer you to fib model code 2010 or to some papers on our website:
http://www.cervenka.cz/assets/files/pap … roclaw.pdf
Let us know if you have further comments or questions.
With best regards
Jan