Computers were rather more primitive than that in my day
As far as Computer Science is concerned, computers haven't really changed much since the 60s.
"Computer Science is no more about computers than astronomy is about telescopes."
The shiny stuff is more Software Engineering, Computer Systems Engineering and the more applied subjects. The trick is knowing which to study as an ostensibly bright A-level student who reckons that mucking about with computers is a reasonable career path. I'd suggest that, as with the hard sciences rarely being a bad move, a thorough grounding in either the electronics or computer science side of things is unlikely to be a bad thing. The elephant in the room is knowing where your talents actually lie. When I was doing A-levels people thought I was good at maths, and that came back to bite me in short order.
Indeedy.
The company I used to work for - 3dlabs/ziilabs before it (or at least, the software-y people) was sold off to Intel included driver/sw, hardware (rtl/asic), and validation (tests to verify the hw during design). In more recent years the two guys recruiting for validation - where understanding low level stuff is kinda helpful (drivers too, for that matter) had real trouble recruiting from the uk. The favourite whinge from one was wrt matlab and java IIRC...but there certainly seemed to he some sort of skills disconnect or focus away from low-level. We picked up the occasional brit electronics engineer that was c/c++ literate, but mainly europeans - Spain, Italy, Portugal (possibly aided by those home job markets.
I'd be very surprised if the level of maths on engineering courses has gone down, at least on those where it was rraditionally high. In particular much of our first year ('83ish) was common to electrical, electronic, I think civil engineering , and maths degrees, and it's stuff you'd probably not be able to shy away from, being relevant to design/analysis, but that level wouldnt have been so directly relevant to computing/sw engineering (unless it involved dsp/signal processing)
(Fwiw I actually did a bridging course after TEC rather than a-levels - the useful bits of maths/mechanics/physics/elec eng in 12 weeks
)
One thing that I feel is worth pointing out wrt sw/computing is that being academically good is not the same as being good at doing it - ive been a hw design engineer (mostly digital - board, fpga, asic/rtl), hw validation engineer (low level c++ to test design blocks), but met my match in moving to the (linux) driver team.
I came to the conclusion that my head likes diagrams - fine with hw (tho' some pipeline stuff occasionally screwed with my head) - but very often with drivers/sw the nitty gritty that you need to hold in your head doesnt fall out neatly diagrammatically - particularly when inheriting someone's code, which is actually much of one's time. Whereas I think I remember stuff when it dovetails together - eg I found my degree not that hard, but failed english lit o-level, my memory's rubbish wrt loosed-related information. In moving to Intel the bite-sized chunk I was trying to digest to fit my head diagrammatically was much of the graphics driver :/ (spittoon joke anyone ?)
That said, one mate often says quite how poor -relatively- engineers are elsewhere in industry by comparison to our old company, so maybe im not as bad as I think
