As a general guideline, while building a model on HANA, Calc view should be first choice in case the model can be built-in without any major SQL scripting to derive logic is required.( HANA takes care of Optimum sql's generated internally)
A SQL script view will be required when Calc view cannot provide required model as an example a recursive or loop through logic.
e.g.- Deriving data by joining Material movements + material history table. In this scenario, if a material movements from header tables needs to be searched in History table with backtracking of date ( like searching value in Current period, if not found then previous period and so on till value is find for any period) , it has be done in SQL view.
Regarding performance, it can be fine tuned for both to an extend.