I haven't looked at Form before. I seem to recall the name from somewhere, but I might be mistaken. Perhaps it's that the theoretical physicists I know use Mathematica almost exclusively.

Indeed Form received some attention at my department. I wasn't urged nor required to look into it, but did so out of curiosity. It's a very powerful piece of software that IMHO may deserve being known also outside of high energy physics circles. I remember experimenting with it, and finding it has many interesting features. But seriously, since you mentioned Mathematica, speaking about the latter, or similar "high level" software, the tutorial (or was it the reference manual?) of Form is clear: they're like swiss army knives that "do it all". Form is quite different: it is extremely specialized, a knife that will only do one thing. To do more complex stuff you will have to know how to do so, and that won't be just as intuitive as with those other programs. For example, Form doesn't know what a derivative is: it can indeed do calculations that involve derivatives, but you have to instruct it to do so! The advantage is that it will happily and effciently handle computations on which those other things may choke...