This study aims to describe whether there are differences in learning mathematics-based problem-solving in terms of results and student motivation of mathematics. This type of research is a quasi-experimental research and the design carried out on this study is a pretest-postest nonequivalent group design. In this research, the observation was done twice, before and after treatment. Observations made earlier are called pre-test and later observations are called post-test. The result of the calculation of the implementation of mathematics learning with the problem-solving based approach to student achievement is significant (p = 0,000 <0.05). It showed a Fcount of 1.002 <Ftabel of 1.76 with a significance level of 5%. Thus H0 is accepted and Ha is rejected, meaning there is a significant difference in mathematics learning achievement between problem-solving approach and conventional. The result of the calculation of the implementation of mathematics learning with the problem-solving-based approach to student's learning motivation is significant (p-count = 0.03 <0.05) or refer to Fcount of 1.07 <Ftable of 1.76 with a significance level of 5%. So that Ho accepted, it means there are differences in student learning motivation that is taught between using problem-solving based learning and conventional learning. Learning with problem-solving approaches will be effective if the teacher as an educator prepares the problems according to the context. From the results of this study, the researchers recommend being used as a reference for similar research with different subject materials. This research can be developed by adding other attribute variables such as student self-confidence, verbal ability, motivation, nor critical thinking ability.