Dominion created an entirely new genre of game: in-game deck-building. As the second major release in the genre, Thunderstone is flawed, but oddly compelling. Players build decks representing fantasy adventuring companies, then use those decks to slay monsters lurking the dungeon.
Each player begins the game with a small deck of cards of militia members, daggers, torches, and rations. From a hand drawn from their own deck, each player takes a turn either shopping in town or raiding the dungeon.
Town is a set of 12 commodities and 4 heroes, chosen from many more options. Each chosen commodity is a set of identical cards. The heroes come in 3 levels. Some of the cards in a player's hand provide coins which can be used to purchase these commodities and heroes to add to their personal deck.
The dungeon is composed of 3 different categories of monsters, again chosen from a larger number of sets. 3 are visible at a time, available for attacking. When a player chooses to raid the dungeon, he picks a monster to attack, presumably defeating it with his hand. A player must muster a suitable attack, typically by equipping heroes in his hand with gear from his hand, but also bringing enough light and coping with special rules for each monster.
A defeated monster is added to the player's private deck and a new monster is drawn to ensure 3 are still visible. Monsters are the primary source of victory points, but they generally provide no mechanical bonus, clogging up hands with useless cards. Slain monsters also provide experience points, which can be spent in town to turn low level heroes into higher level heroes.
For a game so obviously derived from Dominion, they are dark reflections of each other. Dominion shines in carefully tuned balance, the reflection the result of years of polishing. Unfortunately Dominion has been polished so aggressively that the game feels abstract. In Dominion the connection between the mechanics and the theme is almost nonexistent; I never feel like a lord expanding his domain.
Thunderstone, on the other hand, is rough and colorful. The theme is pervasive and the game is dedicated to reproducing many tropes from fantasy dungeon raiding games. Instead of describing mechanics, players tend to buy in enough to talk about slaying monsters and equipping heroes. That it was rushed out, at least compared to Dominion, shows in the flawed graphic design and erratic gameplay.
The cards are covered with numbers, identified only by position and by color. Early games will be dominated by questions as people confirm what the numbers mean. Small labels would have helped.
To track experience points, there is a custom deck of cards, all identical. These XP cards are never mixed with the other cards. But inexplicably, they have the same backs as the rest of the cards, making it easy to accidentally mix them in. Small tokens, glass beads, or half-sized cards would have served the task better.
The game makes a big deal about light, the visible monsters are subject to -1, -2, or -3 light. If the heroes lack matching light, each point missing makes the attack 2 worse (so -2, -4, or -6). The multiplier is confusing and raises problems for new players. Simply multiplying all of the light levels by 2 before printing the cards would have simplified teaching and playing.
Many of the cards refer to "heroes." The weakest heroes, Militia, are labeled as being Heroes, but the more powerful heroes are not so labeled, despite being heroes. Some of the monster cards provide benefits when they appear in a player's hand, but it's easy to overlook them there or to erroneously apply them when they are still acting as monsters on the table.
Actual play is similarly rough. The monsters vary widely in power. The monsters are shuffled together, so it is common to hit a glut of tough monsters in the early game. At that point play stalls while everyone grinds town repeatedly trying to build up a powerful enough deck. Once players hit that sweet, frequently the remaining monsters are exceptionally easy. Both modes of play can be dull. Unsurprisingly, players have proposed a number of variants to fix this, including roughly sorting the monsters by difficulty.
With groups of 4 or 5 players, monsters can cycle so quickly through the visible spots that it can be hard to predict and plan. It can leave an already very random game feeling chaotic and out of control. This problem isn't evident with 2 or 3 players.
The overall game balance leads to lopsided games. Dominion truly supports multiple effective strategies. Even poor strategies can lead to a good run and a low but not embarrassingly low score. With a mix of skilled and new players, it's not uncommon to see everyone finish within 5 points of each other with final scores in the area of 20.
Thunderstone, on the other hand, tends to have obviously superior strategies each game. A player with an inferior strategy, by bad choice or bad luck, can end up with 10 points while the winner might have 40. This can be dispiriting for a new player and frustrating for a skilled player blocked by back luck.
So where do strong theme and pervasive flaws add up to? A weird mix. After many games with a dozen or so players, the consensus appears to be that it's not a very good game, but they'd like to play again. Something about the play is compelling. The game generates amusing little stories as you play. When you pull a lucky hand and stomp a monster it's extremely satisfying. The three militia and two feasts becomes the worlds most dangerous army, marching on its stomach. When you draw a bad hand, you can commiserate with your friends about how the heroes you hired must all be too busy getting drunk to bother showing up for work.
Dominion clearly has superior gameplay and remains a favorite. but if you're looking for a change, Thunderstone's unexpected draw is probably worth your money.