There are a number of facets to this problem. My own personal experience has been that learning existing math is not very hard and quite fun (YMMV of course an

There are a number of facets to this problem. My own personal experience has been that learning existing math is not very hard and quite fun (YMMV of course and it also depends on how much pre-requisite knowledge is required to even approach the topic) Once you've acquired the tools, applying them to solve actual engineering problems, also relatively easy (depending on the problem of course) and very fun. However, solving math problems is a completely different game, and this is where (again for me), the discipline is the most frustrating. Solving a math problem is like finding a path out of a dense forest, and some people seem to have a "natural compass" guiding them towards it. For me (born w/o much of a compass), it's always felt like I have to recursively try all possible paths until I find the one that gets me there. Needless to say, if the forest is dense and thick enough, that's a completely hopeless endeavor. For example, reading the proof to a theorem, assuming it uses tools, concepts and facts you're familiar enough with and does not take giant leaps (the infamous "from here it obviously follow that ...") is easy and can be fun. But when I get to the QED, I'm always left wondering how the guy who first proved it effing found the path in the first place. It's borderline disheartening.