For EE make sure you study at least some analog. At UCSC I took the analog class that was taught by the Physics department, with the aim of teaching us to buil

For EE make sure you study at least some analog. At UCSC I took the analog class that was taught by the Physics department, with the aim of teaching us to build our own instrumentation for physics experiments. Dr. Dorfan pointed out that analog wasn't required for a EE degree. He said "I can see how they can design digital chips, but I can't understand how they get them to actually work." Consider problems like capacitive coupling, and hot spots due to frequently-used parts of your digital circuit being close together. Learn the assembly code of at least one instruction set architecture. Pointers will make a lot more sense then. These days I would advise that you learn ARM; Steve Furber's ARM System-on-Chip Architecture is quite good. Intel assembly may have more widespread application however it is far harder to make sense of than ARM.