This study aimed to find a significant difference between the learning outcomes of physics students taught through problem-based learning supported by a variety of visual media and those taught through conventional learning. This is to determine whether The method used in this study is quasi-experimental, using only a post-test control group design. The study population consisted of all 7th graders from public schools in South Sulawesi, and the sample was divided into two classes, an experimental class and a control class. The test vehicle for learning cognitive outcomes in the multiple-choice format of the extended material serves as a test vehicle. The sig values were obtained as a result of data analysis, i.e. based on the independent-samples t-test. (two-sided) is 0.000, which is less than the significance of 0.05, so Ho is rejected, and H1 is accepted. Thus, there is a significant difference between the physics learning outcomes of students taught through problem-based learning supported by a variety of visual media and the physics learning outcomes of students taught through conventional learning.