At first, before I looking carefully at this tile game, I thought it was a game of random dungeon exploration. Actually, though, it is more of a strategic tile placement game.
Each turn, you either place a tile, or you move. You can't normally move backwards, so often you or someone else will have to place a tile in order for movement to be possible. Each tile can contain traps, special actions to take, or treasures. You win the game when you gain 5 gold pieces.
Consider a 2-player game. Both players start together. Whenever you extend the dungeon from your current location, the other player has the option to move first. But if one player moves, the other can follow. The two players could do this until they reach a portion of the dungeon which has no exit tiles placed yet. At this point the person in the lead loses the initiative, as he has to place tiles rather than move. So the person who moved first really hasn't gained anything over his rival. These rules mean that, from a theoretical standpoint, there is no benefit to moving until enough tiles have been placed to give you a winning path, so that by moving first, you can either pick up the 5 gold one turn before the other player can if he follows you, or you can "shake" the other player by being the first to take a special action from one of the rooms. But if you put down the tile that creates a winning path, the other player can take the path first. This symmetry is broken by the fact that you have hidden tiles in your hand, and some of those tiles could break up the path the other player takes. So you want to trick the other player into starting on what looks like a winning path, then play your own path which disrupts it or is better. Or you want to trick him into not realizing that you just played the winning path – although this is still a pretty good chance he can do something to mess you up. Anyway, until the players actually split up, the complexity of deciding which play is actually the best one is just way too much for me. Some people might like this sort of deep thinking, but for me, it just hurts my head.
I tried playing with the rule that the players couldn't follow each other, and the game was much more to my liking. Basically, you end up trying to create a profitable path in your part of the dungeon, while messing up the other player's path in his part of the dungeon. The fact that you have to place your tile on one turn, and move into it on the next turn, creates a funny slow pace that is different from what would I expect from an exploration game. But it works OK, and I like building up the dungeon and wandering around. And it doesn't take too long to build up enough money for someone to win the game. I wouldn't mind playing this variant of the game some more, though I'm not sure I'd go out of my way to do so.