This paper reports the results of an experiment in a Computer Architecture Laboratory course classroom session, in which students were divided into two groups for interaction both with a hinting e-learning system and with human teachers generating hints. The results show that there were high learning gains for both groups, demonstrating the effectiveness of the human teachers as well as of the computer-based hinting e-learning system even without the use of adaptive and personalization capabilities. In addition, in the worst case, the difference in favor of human teachers (with a low student-to-teacher ratio of 13.5 students per teacher) would not be significant with respect to the e-learning system, so the computer-based system can replace teachers without a significant loss of effectiveness. (Contains 1 table and 3 figures.)