This investigation has studied the six most-used macroscopic models to compute hardness from the elastic constants, using an experimental database of 143 materials. "The Hardness Calculator" is proposed as a solution to estimate hardness in an easy, fast and confident manner. This study divides into two stages. The first approach, referred to as "The Classic Calculator", is a selection model based on simple properties of a solid such as crystal system, bandgap, and density. The second phase is machine learning (ML) based, and it is referred to as "The Machine Learning Calculator". We used two different methods to compute hardness in the ML approach. The first ML method is a classifier that targets the best model to calculate hardness using the mechanical properties of a solid (bulk modulus, shear modulus, Young's modulus, and Poisson's ratio) as input variables. The second ML method is a regressor that directly predicts the value of hardness using the same input variables as the classifier. Using the Materials Project's database, the classic and ML schemes were compared and tested to predict new hard and superhard materials. All methods are available in a free-access online application for users to discriminate between the different available results.