2009 Game Award evaluation
This game has promise, but it has some fundamental flaws.
First, the dice are completely hidden. There is no need to go through all the trouble of hiding them for the whole game: just roll a die at the end of the game for each tower, and add its value to its score.
Now, there is potential fun by letting each player see some of the dice, but not all of them. Then there could be a mind-game element by building up on some towers, and wondering if the opponent is doing it because it's valuable, or because he's trying to get me to waste my resources. Perhaps the players can see some of the dice at the start of the game, and/or peek them one at a time during the game, for a price (sacrificing a piece, say).
Second, the player who goes second can guarantee a draw by always playing a piece of identical size on top of the piece his opponent just played. If there is partially know information as discussed above, that will mitigate that a lot, but there should still be some sort of limitation to avoid duplication: the heights of the towers, or playing on top of previous move, or a variety of other possibilities.
My assessment for the award: this doesn't make the cut. But I'd like to see more done with this.