<>To solve chess, a computer the size of the universe is needed. There is a very famous >paper on this, I forget where it is. There are more chess positions than atoms in the >universe, so with "information density" such that a solution to one position could fit on >one atom (an impossibility) you would need a universe to solve chess.

This is wrong for 2 reasons:

A)You don't have to store the solution somewhere. You just have to find it and then play it.

B)It is possible that a set of strategic rules can be found which ensure a win for one of the players or a draw for both. If the correctness of these rules can be proven, it is not necessary to evaluate a large number of positions, to obtain the result of the game if both players play correctly. (Victor Allis)
For example at Connect 4 7x6, there are about 1.6·10^13 position possible so to store all these positions we would need 4 TB, but the game has been solved and the program that plays perfect Connect 4 7x6 is only 2.5 MB.