I remember those mainframe days. When your cards got mangled by a reader, you had to know how to disassemble it and clear the pathway. The 360 had large consoles.
The first microcomputers that I used in the 70's ran CP/M. Some of the hardware was on S100 cards plugged into a bus. Floppies were 8". The ones that I used were rack based. Altair and IMSAI were two popular products. You could key in an address and data from the front panel to get the computer to boot, like the DEC PDP computers. In the 80's I designed computers, Basic Input Output Systems (BIOS) that were loaded onto EPROMS to access the hardware, disk drivers cards, and video displays. We focused on video and imaging research. This was all before Microsoft and the IBM PC. For many years those machines were considered toys, not able to do anything useful other than to play with. Computer games started, there was pong and chess.
It was an exciting time. New hardware revolutionized the industry. Big companies would take years of incremental research advancements and shrink it onto a single board or chip. Before there was autofocus cameras that we take for granted, people worked for years coming up with algorithms, trying all sorts of approaches. At one point, I worked on constructing 3D images from narrow focus 2D slices. Not as easy as you'd think. Very common in medical imaging today. It took a lot of work to get there. For the most part, advancements in hardware and processing speed drove the advancements. It made things that worked only in the lab practical for others to use. Now things like that can run on a laptop instead of a multi-million dollar machine.
View attachment 4005961