Thanks for the help :)
Yeah, the computer analysis just got back to me and I really missed some chances in the middlegame, but there was no reason to lose that endgame =/. I've always assumed creating passed pawns with your majority was the idea, but things seem more complicated now, lol
On whether ...Kd6 in the ending was drawing or not, I haven't looked at it in detail yet but black has to be very careful to avoid zugzwang. Imagine white's king on d4, black's on d6, and white pawns on c4 and b5. A king move by black will allow c5 which I'm guessing gives white a winning pawn ending. If white can triangulate to always give black the move in that position then black will eventually run out of pawn moves on the kingside which do not give an isolated pawn.
Other advice on the endgame: Nc5+ was not a bad move at all, but I'm thinking that Nxd3 was a bad move. The pawn structure favors white, generally speaking. This advantage goes away though when black can recapture ...bxc5.
Another note on this sort of typical Caro position. Black often wants to play ...c5 in these positions, but it's not like he _has_ to. I've faced several times when white has not played energetically, and I've thought about ...c5 and then decided against it. Unless you're getting a real advantage, (or unless it's a book accepted move) then why play it? Perhaps black can delay it, or play for the ...e5 break instead.
I lost this game as black, but I'm not quite sure why. I've been playing the Caro-Kann for just over a month now, and from everything I've heard, reaching anendgame favors black. However, I just blew this endgame, and if I don't understand exactly what was wrong about my approach I'll probably do it again. Any advice?