i second this. this is probably my favorite text book behind Mackay's "Information Theory, Inference, and Learning Algorithms".

if you take the time to honestly follow the text you come away with a great understanding of convex optimization... both the theory and the methods.