This study demonstrates that the advanced theoretical basis and the consequential numerical complexity do not always guarantee the success of EOS models in predicting the experimental thermodynamic property data. Although one of the best versions of SAFT, namely SAFT-VR-Mie might have doubtless advantages in predicting the data of nonspherical molecules, once again it is shown that there is a price to pay for the excessive model's complexity. In particular, the present study reveals a previously unnoticed kind of numerical pitfalls, yet generated by the chain term of the SAFT-VR-Mie EOS. A possible way of avoiding the numerical pitfall under consideration is proposed.