piercebeatz said:
Does it still have a sense of Euclid-style geometry-are there still cubes and spheres, so to speak?
While it's certainly possible for cubes and spheres to show up, they are not central to differential geometry. It certainly has a very different flavor than usual Euclidean geometry.
Is it mostly about 1D curves/2D surfaces, or does it consider higher dimensions?
Higher dimensions are extremely important in differential geometry. However, when you first start out, you will usually only learn about curves and surfaces. Only later do they really consider higher dimensions.
Are the surfaces which the field concerns mostly graphs of several variables, e.g. ## x^3+y^3+z^3=1 ##, or are they more abstract, like in topology?
They are defined very asbtractly, much like in topology. In fact, the things considered in differential geometry are called manifold and they are topological spaces with a certain smooth structure.
Graphs do show up and are very important. In fact, we can prove that every manifold is actually (locally) the same as a graph of a good function.
What prerequisites does it have?
An introductory course would be about curves and surfaces. The prerequisites are calculus and linear algebra. A rigorous analysis course wouldn't hurt either.
More advanced courses require a very good knowledge of topology
Are complex numbers/complex analysis used at all?
Yes, they are important in many ways. In fact, an entire branch of differential geometry is focused on complex geometry.