The information about the state of the ionosphere is important for various technological applications. This state is determined by two parameters: the critical frequency foF2, and the total electron content TEC, so their modeling and prediction is of great importance. One of the important points is the availability of data. Because coverage of the globe by ionosondes and GPS receivers varies greatly, data for only one parameter may be available in across the regions, and there is no choice as to which parameter to use. However, it is necessary to evaluate how different the accuracy of the foF2 and TEC forecast methods may be. This can be done at points where both foF2 and TEC measurements are present. Although the foF2 and TEC parameters are interrelated and interchangeable, methods for their prediction were developed in parallel and practically did not overlap. Therefore, in this work, neural network methods for predicting TEC, which provided the greatest accuracy, were used to predict foF2. The results are presented in two parts, the first of which compares the forecast accuracies of foF2 and TEC for the Juliusruh station, the second provides methodological details obtained by comparing the results for another region in direct comparison with literature data. The main conclusion is that the accuracy of the foF2 forecast is slightly higher, but the TEC forecast can be used to assess the state of the ionosphere.