assert() says: this ought to hold when I do testing
assume() says: I have proven this to hold
Just as another example of how these two are different:
You might want to assert() that you have enough headroom in a
statically allocated buffer when testing. Say you assert that the
buffer is never more than 50% full for normal loads. You
obviously don't want the -release build to make such an
assumption!
So the content of an assert() is in this regard different from a
prove(). It can sometimes be desirable to put too strong
constraints on the program when testing, then relax it on release
to get that extra headroom for "odd situations".
You can probably make the same argument for tolerances in
computations that are approximate/randomized/iterative. Testing
for a tight tolerance does not mean that you want the optimizer
to assume that the tolerance is always within that bound. You
want some extra headroom in release to be sure that it behaves
correctly under more difficult circumstances.
