Floating point arithmetic -- mantissa and exponent

I have been trying to understand floating point numbers 开发者_开发百科in Assembly, but mention keeps being made of the mantissa and exponent.

I really have no idea what these two words mean or are referring to...could someone explain it to me in relation to floating point arithmetic, or point to an easy to understand explanation?

---

http://en.wikipedia.org/wiki/Floating_point

At the assembly language level, you usually don't need to worry about any of this unless you're actually doing bit-tricks with floating-point. The only case that could come up is performance degradation from denormals - which is also explained on wikipedia.