Analysing complex phenomena, such as the world we live in, or complex interactions, also requires methods that are suitable for considering both the individual aspects of these phenomena and the resulting overall system. As a method well suited for the consideration of complex phenomena, we consider agent-based models in this study. Using two programming languages (Netlogo and Julia) we simulate a simple bounded-rationality opinion formation model with and without backfire effect. We analyzed, which of the languages is better for the creation of agent-based models and found, that both languages have some advantages for the creation of simulations. While Julia is much faster in simulating a model, Netlogo has a nice Interface and is more intuitive to use for non-computer scientists. Thus the choice of the programming language remains always a trade-off and in future more complex models should be …