The University of Oregon had an IBM 360 mainframe available for all students to use. The drawback was that you needed to punch your program onto cards in FORTRAN, submit it into a queue, and wait 3 hours to see the results. You could submit 3 jobs at a time and got 6 seconds of cpu time per job. A beginning class in programming on this machine was my first exposure to real computers. A more advanced class in JCL taught me that mainframes were not for me (not that I didn't do well in the class, the IBM way of doing things just didn't fit well with my personality). About the only good thing I learned from this experience was how to write efficient programs that could compile and run in 6 seconds.
A class in numerical analysis gained me access to the other mainframe at the U of O, a DEC PDP-10 timesharing computer. What a contrast! Ever try typing faster than 10 characters per second? Most people hit rates 2-3 times that. A Teletype won't let you do that. You can only get the keys to go down at the 10 per second rate due to mechanical limitations. Teletypes were a mass of levers and springs just waiting to fall apart. They actually had a circular rotating switch wheel that sent the proper serial bits into the computer. Nothing electronic about Teletypes. I sure wish someone would create a program that would emit Teletype sounds from a modern computer when you type, including the whirring sound of the motor when nothing is happening.
I may have gotten lucky in my junior year (1972), but I think it was more a matter of the heavy computer class work I was taking. Professor Russ Donnelly asked me to do numerical calculations for some projects he was working on. Even better yet, he was going to pay me and let me have access to a mini-computer. Students were very lucky at the U of O in those days. They picked out undergraduates and allowed them to work on research. I didn't know it at the time, but I had just been asked to work for the #1 physics group in Oregon. With personal access to a mini-computer, how could a person turn down this offer?
The mini-computer in question was a Hewlett-Packard 2100. It had a 5 MB fixed disk, 5 MB removable disk, 9-track tape drive and a high speed paper tape reader and punch. It ran a simple DOS operating system which only supported single root directories. This wasn't too bad because the one disk was removable and could be used to store projects. It programmed in FORTRAN and assembly language. It was connected to a H-P mass spectrometer and other lab instruments, giving me plenty of fun things to play with.
The two main mini-computers used in colleges during these times were the H-P 2100 and DEC PDP-8 or 11 machines. I must commend both companies on their excellent technical documentation. The H-P came with a little hardware booklet that completely explained the instruction set and also how the I/O system worked. This book provided me with much of what I know about computers.
During my years at the Donnelly group I was involved in many projects. Numerical calculations were done on the superfluid characteristics of helium. Astronomical readings from Pine Mountain Observatory were analyzed. The U of O used undergraduates to operate telescopes at the observatory. While we didn't produce any ground breaking astronomical discoveries, several of the students went on to become well respected astronomers.