Castles of Burgundy Review

A picturesque castle sits atop a hill in an idyllic French countryside. It commands a sizable domain; buildings and small towns sit in close proximity, and farmers tend to their livestock while ships move goods to and from the estate on the nearby river. This is Castles of Burgundy, a peaceful eurogame that tasks players with building the most successful estate.
Castles of Burgundy, designed by Stefan Feld and published by Alea/Ravensburger in 2011, is a lovely little tile placement game, in which players take turns taking hexes off of a main board, and placing them on their own player board. Of course, each hex grants the players different bonuses, and choosing what and how and when to place them will be the main determinant of victory. The game supports two to four players and a listed playtime of 30-90 minutes.
Gameplay
How do you play the game? Is it a fun experience? How much interaction is there between the players? Is there very much luck involved? Will it take forever to to learn and teach?
Each action is represented by a die – in three out of the four actions, the roll of the die influences the scope of what a player can do for that action. Once both dice are “used,” the player is out of actions for that turn.
1 – Take a tile from the main board

The main board. The hexes that come out are drawn at random. This is for a two player game – in a game with more players, more spots would be covered (hence the “3” spots for three players, and “4” for four).
The main board, every round, is replenished with hexes that can be taken by players during their turns. As an action, a player may roll a die, and choose one hex to pull from the board, which immediately gets moved to a storage area on their own player board. The tiles available for choosing are limited by the roll of the die. There are six “depots,” all of which contain their own tiles, and each depot corresponds to a number on the die. So, if a player rolls a 3, he can choose any tile within the 3 depot.
2 – Place a tile on the player board

The player board. The brown tiles in the corner can only be placed if the player rolls a 3. With the current roll, only one can be placed. With the help of a worker, the other die’s 4 could be changed to a 3, allowing the other tile to be placed as well.
When a tile is taken, it moves to the storage area on the player board. On a placement action, any one tile in storage is moved from storage onto the main region of the player board. Upon being placed, the tile immediately yields its benefits to the player. There are stipulations to where the tile can be placed; once again, the player rolls a die, and this determines placement options. A player who rolls a 5 cannot place on a 2. Secondly, all tiles must be placed adjacent to a tile currently on the board. Players begin the game with one tile, so all placement grows out from that spot. Finally, the spot that where the tile is placed must match the tile’s color. There are dark green, light green, yellow, brown, blue, and gray tiles.
3 – Sell goods

The good tiles. Players can hold up to three different colors at one time. If this player rolled a four, he could choose to sell his red goods, flipping them over and getting two points for each one as well as a silverling.
“Goods” can be obtained throughout the game, and they accumulate at the top of your board. A player can take an action to sell these goods, which yields victory points and one silverling (which can be used to purchase tiles as a special action) per transaction. The goods tiles also have die numbers on them, so in order to sell a good, a player must roll the corresponding number.
4 – Obtain workers

You can use a worker to adjust the result of the die plus or minus 1. Multiple workers can be used at once.
Workers are the game’s biggest luck mitigation component. By using a worker during any given action, the player immediately gains the ability to modify their die result by plus or minus one. So, if a player really needs to place a brown tile on a 6 spot but rolls a 5, he can choose to cash in a worker, which will add a point to his roll, allowing him to place the tile. Multiple workers can be used at once, for example, a 2 can turn into a 5 with the help of three workers. If a player runs out of workers or simply wants more, they can exchange a die for two workers.
Bonus Action – Buy black tiles

The black tiles on the middle can represent any type of tile. These can’t be pulled off the board normally; silverlings must be paid to acquire them.
There is one last action that a player can take; this is a bonus action that can be played in addition to their two actions. If a player has two silverlings, they can exchange them for a black tile, which tiles rest in the middle of the board. Black tiles are kind of like wild cards, they can represent any of the other colors. These come out at random every round, meaning that every round there will be more tiles than others in certain categories. Unlike the rest of the tiles on the board, these can only be bought by silverlings; taking them with a regular action is not an option.
Everything that the player does is dictated by these four actions. It’s also worth noting that the same action can be played twice in one turn. When it comes down to it, a player is doing one or two of these five things every turn until the game ends. However, the functions of the tiles combined with the various scoring methods provide an enormous amount of strategic variability. In order to understand the Castles of Burgundy, it’s important to understand how each type of tile works:
Dark green tiles represent castles, and placing them grants an immediate free action, with a die result of the player’s choosing. Light green tiles represent animals, which grant victory points. The more animals of one type that are concentrated in one patch, the more points they grant. Blue tiles allow players to collect goods – they can select goods from a depot of choice (which accumulate randomly throughout the rounds), and then they can move their player token forward on the turn track, allowing them to influence turn order. Brown tiles represent buildings. There are eight buildings, and each one of them gives the player an immediate benefit, allowing them to either grab or place more tiles than usual, or giving immediate bonuses, such as silverlings, workers, or victory points. Gray tiles represent mines, and give players an extra silverling at the beginning of each round (the players usually each get one). Yellow tiles represent “knowledge” and each one grants a special ability. These are unique from buildings in that they grant passive benefits instead of immediate ones. Many yellow tiles grant bonus points at the end of the game, or grant permanent bonuses, like making workers more powerful.
It’s all about victory points in CoB, and so it’s also important to know how the game scores. Understanding this, and playing your tiles to accommodate for the most efficient scoring is the key to success. The player boards are made up of colored regions, where tiles of said color can be clumped together. Upon filling up an entire region, a player gains points based on the size of the region, and how early in the game it was finished. The sooner a player is able to complete regions, the more points they’re worth. For example, a region finished in the first round will grant ten extra points, whereas in the final round it will grant only two. So, for example, finishing the smaller regions early on will grant a higher point advantage than aiming for the large ones, which usually can’t be filled in one round.

This table shows the amount of points you get for filling up one colored region on your board. A five hex region, like the light green one to the side, would yield 15 points, while the one-hex brown region next to it would only give 1.
By laying tiles, players can gain benefits that give them more action economy, which can then be used to fill regions in the most efficient manner possible. This is the crux of Castles of Burgundy: It’s all about placing one tile at a time to get the most points possible. It’s hard to capture the puzzle-like essence of Castles of Burgundy in words, but in practice, it’s a delightful mental exercise.
One thing that stands out to me with Castles of Burgundy is that it is, for lack of better words, relaxing. I love to play Castles of Burgundy when I feel like unwinding; finding the right pieces and planning how you’re going to chain together your abilities is weirdly therapeutic. Like most euros, this is a “decision” game. You have a limited number of decisions to make, and every single one counts. There is an opportunity cost with every action, and optimizing your moves will yield the most points.
I think what makes Castles so relaxing is that you’re not punished hard for your mistakes like you often are in action economy games. In Castles of Burgundy, you reap the benefits if you make wise decisions, but the game doesn’t feel more difficult if you don’t. I can think of plenty of games where this isn’t the case. Tzolk’in comes to mind – in that game, using your limited actions unwisely makes you wildly inefficient, and opponents who chose better will soar ahead while you try to stumble out of your mire of bad choices. The lack of punishment for less-than-ideal decision making is what makes Castles soothing for me. Sometimes you want to play a hard game that unabashedly requires your acute mental presence, but other times it’s better to just play a game that’s not so…demanding.
I suppose something that contributes to the feel-good nature of Castles is its brilliant scoring system. Due to the nature of how points are earned (through filling regions and the timeliness thereof), most players usually end up in the same ballpark range, points-wise. There are only so many spots on each player board, and there is always a fixed number of turns; roughly the same amount of board will be filled at the end of the day no matter what the players do, so the points are always relatively close. The small things matter, of course, and they’re always the difference between a win and a loss. However, the closeness of the score every game makes the game a lot less punishing to new, casual, or downright unskilled players. Even if the other player completely outplays you, it’s psychologically rewarding to be able to look at the board and say, “hey, I was only that far behind you!”
Now, I may be exaggerating a little bit; there have been several games where there’s been a significant point disparity between players, but in the majority of games, this isn’t the case. The winner is rewarded for superior tactics, but the losers don’t feel beaten down and dragged through the mud. Overall, Castles of Burgundy is a wonderful puzzle-esque thinking experience that doesn’t stress you out when you make poor decisions. If this is the kind of board game experience that’s appealing to you, Castles is a real winner and deserves a spot in your collection.
Our sessions typically average between one and two hours. I’m not sure if I’ve ever played a session longer than two hours, but I’ve definitely never played one that wasn’t over one hour. I would plan to set aside at least an hour for any session. One thing that also contributes to this is setup time. Castles isn’t a nightmare to set up, but it’s not something that’ll just take a couple minutes either. There is a truckload of little miniature hexes in the box, and the game’s setup mandates that you put them all face down before you start the game, and It can take forever turning all of those over. To expedite this, we cleaned out some medicine capsules, in which we store stacks of hexes. Aside from that, there’s just a bunch of stuff that you have to put on the main board. A lot of little hexes to place, as well as goods tiles and point bonus tiles and so on. Setup might take ten minutes or so, so account for that when you feel like playing a game.
For anyone who’s read my 7 Wonders Review, you’ll know I have a bone to pick with “iconography games,” or in other words, games that resort to little icons and pictures to explain every feasible action in the game. Castles of Burgundy is one such game. Iconography isn’t such a bad thing when you’ve got it down, but it’s a pain to teach, and in my opinion, not a whole lot more efficient than simply having a text reference. The one that gains the most from iconography is the publisher, who can save money not having to print components in multiple languages for international editions. Castles of Burgundy has this silly glossary on each player board that details 13 different actions that can be performed by certain tiles, and I still can’t believe how often during any given game that I still get questions about them. Each yellow tile is also a unique action, also represented by icons. Even after plying many games, I still find myself looking in the rulebook sometimes to decipher them.

Behold, the copious iconography of Castles of Burgundy. The icons on the right explain what all of the tiles do, yellow ones not included, which are in the middle. The ones on the left show what actions are available to play every round.
Don’t get me wrong – Castles of Burgundy is pretty easy to learn, and it’s pretty easy to teach. It won’t take long to start playing, BUT, expect to be answering a lot of questions about the finer details while you play. I would estimate that it takes between fifteen and thirty minutes to explain the game and start playing.
That being said don’t expect there to be no interaction – if you want to get more competitive, the game certainly allows for it. While players can focus on their own boards, seeing what other players are doing is an important component of deeper strategy in this game. I recall one time where my opponent was one animal hex away from getting a massive bonus, and on my turn, I took the last tile she needed to get that boost. I didn’t even need it, and I felt dirty, but it got me the win. Let me repeat that though: I felt dirty. You can get sort of aggressive in this game, but I would make sure you’re playing with the right people. This isn’t by nature an aggressive game, so make sure you know who you’re playing with. Looking back, I’m not sure my opponent would have done the same to me, and it made me feel pretty mean for a game where you mostly keep to yourself. So keep that in mind – the game has limited interaction, but you can decide to take it to another level if you want to get competitive; just make sure everyone’s on the same page if you do.
Depth
How much strategy is involved? Is there a sense of variety and balance? Does the game play well no matter how many players there are? How long does it take to play?
I will say that your decisions matter, and that the most efficient player is the one that always wins. I mentioned before that the game is really good at getting most of the players to score semi-equally, but this doesn’t mean that making good decisions doesn’t matter. I suppose the easiest way to demonstrate effective strategy for this game would be to compare two players.
Jane wants to fill her animal pasture. There are two cow hexes on the board, so she decides to go with cows. For animals, you score a point for every animal on the tile. However, if you play another animal tile of the same type in the same region, you’ll get the points again for the tiles you already had in there. So, placing a tile with two cows will give two points. Placing a hex in the same area with three cows will give you three cows plus two cows again, equaling five points. So, back to Jane. Her goal is to fill every spot in her animal pasture with cows. She goes for it, and it takes until the last round to complete the region (cows aren’t guaranteed to come out every round, so focusing on one animal may take awhile). We’ll say that her cow tiles had 2 cows, 2 cows, 3, 3 and 4. The cows would score her 39 points (2 + 4 + 7 + 11 + 16) plus 15 points for completing a five tile region, plus two points for finishing on the last round, totaling 56 points for her pasture.

Mixing and matching animals will allow you to complete the region sooner, but won’t give as many points.
Jack, on his pasture, decides to just get whatever animals come out. If we assume that cows didn’t come for the next few rounds, Jack would have the lion’s share of animals, and could fill up his hexes pretty quickly. We’ll say he finishes his pasture on round 2, but with random animals. If he had 2 chicken tiles, worth 4 and 3, two sheep tiles (also 4 and 3) and one pig tile (3) his animal score would be 25 (4 + 7 + 4 + 7 + 3). Completing the region would be 15 points, and completing it on round two would be worth 8 points, instead of Jane’s 2. His animal score comes out to 48 total, eight points less than his opponent.

Smaller areas give diminishing returns the longer the game goes on; filling them earlier allows them to score more points. Because she didn’t focus exclusively on animals, Jane was able to fill three small regions by the end of round 2.
However, if we were to carry out the scenario farther, we’d see other implications from their strategies. For instance, if Jane were to wait for cow tiles to come out, she’d have other things to do in the meantime when they don’t appear. Perhaps Jane, in her cow deficit, would decide to target the smaller regions in the early game to take advantage of early region completion points. On the default board there are two regions just one tile strong; bagging both of those in round one gets you 20 extra points (10 points each for finishing a “region” in the first round).

By focusing on animals for the first two rounds, Jack was able to fill his patch early, which would shoot him way ahead of Jane in points. However, not taking advantage of small regions earlier, as well as not stacking identical animals will come back to bite him later.
Jack, on the other hand, wanted to fill in his big animal region early. The time it would take him to buy all the animal tiles off the board would cost him time he could have used to get better point bonuses early on. Filling small regions in the late game is much less of a payoff. The one tile regions, for example, are worth 1 point for completion, plus their time bonus, meaning they’re worth eleven points in the first round and three in the last round. That’s a big difference for areas that are so easy to secure. Timing is everything in this game, and the wiser players will find more ways to maximize their point potential.

These three buildings allow you to immediately pull another tile off the board (of the specified color) when they’re placed. Taking advantage of free actions such as this, as well as combining them with others, can give you a huge advantage.
The other strategic element I love in this game is finding ways to get more actions each turn. Filling the board earns you points, and taking advantage of free actions to collect more tiles is key. One thing I love doing is playing multiple tiles at once that allow me to chain abilities together, allowing me to get twice as much done in one turn than what I normally could have. For example, one building, when played, allows you to immediately pull another building off the board and into your storage. If I started the round with that building, I could place it with one action, allowing me to pull another one down. We’ll imagine that I pull down a city hall, which allows me to immediately place a tile for free. I decide to place it, now I can place another tile in my storage immediately, with no die roll required. If we imagine that I have a castle sitting in my storage, I can place that, which grants me a free action. Maybe I use that action to pull down a valuable animal tile. I could have just pulled that tile for one of my turns, but by finding the more efficient way, I was able to do the same thing, but with two more tiles now on my board than I would have had otherwise.
All in all, if you’re looking for a game with tactical depth, Castles of Burgundy should be able to scratch that itch. The gameplay itself is fairly simple, but there is so much marginal benefit to be had from wise decisions that you’ll have a ball trying to find them every time.

This player needs one more tile to get an enormous point bonus, and…it didn’t come out on the last round. Too bad!
As far as the dice go, they’re here and there in terms of luck. Generally, there’s not much of a problem. You can almost always do something meaningful with the results you roll, and if not, you can always use workers to change your outcome. The workers are a brilliant mechanic that drastically minimize the amount of luck, good or bad, that come from the dice. As is the case with most luck factors, there are exceptions. For example, sometimes you just really can’t do anything with what you roll, or the numbers force you to use your actions in a way that isn’t utilizing their max potential. Sometimes you have to turn in an action for workers. While workers help, it sucks if one player rolls well enough to never need workers, and the other has to take one or two actions to get some. That’s one or two actions they lost that the other player didn’t for no reason other than luck.
Aside from that, it basically comes down to personal skill. The impact of luck in this game is minimal, and I’ve not been bothered much by it. At the end of every game it always feels like the person who made the best decisions was the one who won. So yes, luck will come here and there, but it has rarely ever made so much of an impact that it changed the course of the game.
One element of the game that deserves scrutiny is the collection of player boards that are included. In most of the pictures thus far, I’ve used the default player board, which is the one that the game recommends playing with if you’re a new player. However, for players who crave variety, there are other boards with different layouts, and each board demands its own strategy. Some boards have are packed with big regions, areas that take a long time to fill but yield higher points, whereas another has a lot of smaller areas that incentivize the player to branch out as much as possible to clear them quickly.

Each player board offers its own strategic possibilities – experimenting with all of them makes the game highly replyable.
Of course, as is always the case with asymmetric mechanics, there’s a question begging to be asked: Does it work? After playing the game pretty extensively, I’m inclined to say yes; no individual board seems particularly stronger than another. The truth is that the scoring system is just so well designed that it’s hard to find imbalance.
One other thing that comes to mind is the built in score balancing that I mentioned earlier. In most games, players end up pretty close to each other. On one hand, this is excellent and allows everybody to feel like they did a pretty good job, and on the other, it begs the question, “do my decisions really matter?” The answer here is a definite yes. When I was learning the game, I conducted an experiment. I played one board as “myself,” making the best decisions possible, and another board completely randomly, with no kind of strategy. My board ended up winning the game with a 100 point advantage. This goes to show that, despite the game’s remarkable ability to balance everyone’s scores, strategy does matter.
In the above scenario, it’s important to remember that the “AI” played completely randomly. Unless you’re playing with someone who just has no grasp on strategy, most people have a good general idea of what to do, and when all players are pursuing the same types of goals, the scores end up much closer. Even though they all come in close proximity, it’s the small, tactical choices that will push you past your other players in points.

This is what the board looks like with four players – this provides a much greater variety of tiles.
So, what about other player counts? With three or four players, it’s just peachy. Honestly, I can’t really criticize the game at either count. This is one of those games where adding a player hardly changes the core experience. There are some differences though. For example, more tiles come out with each player count, so playing with more people will give you access to a greater variety of tiles, which can certainly affect your strategy – if one person, for example, goes for one specific color and nobody else does, they could fill its respective region quicker than they would in a two player game. Another difference is length – four player games stretch on significantly longer than two player games. The game encourages you to roll your dice before your turn so that you can plan ahead while other people take their actions. Make sure this gets enforced, because not doing so can lead to a lot of analysis paralysis, and may keep people waiting quite often.

These are bonus tiles that you obtain if you’re the first to fill an entire color on your board. The points gained depend on the number of players; 5, 6, and 7 for 2, 3, and 4 players respectively.
Certain point values also change with different player counts. Selling goods is an example. When any given goods tile is sold, it yields points equal to the number of players. So in a two player game, every goods tile is worth 2 points, whereas they’re worth 4 in a four player game. The other scoring mechanisms, however, do not change; region completion both in size and timeliness is the same across all counts. This means that selling goods is a much more powerful tool with more players. For a while, my lovely fiancée was the only person with whom I could play this game – I still remember clearly when we were finally joined by another player, and he nearly won his first game just by selling a lot. We were taken aback by the effectiveness of his method; neither of us had ever given much credence to the goods tiles. Being worth just one more point each, they made much more of a ripple in our game than we expected.
The moral of the story here is that Castles works very well with each player count, and it does a good job scaling to each. If you’re looking for a two player experience, this also fits the bill very nicely.
Look and Feel
Is the game aesthetically pleasing? Are the components made out of quality material, or do they feel cheap? Is the rulebook well-designed and easy to read? How well is theme integrated into the game?
Castles of Burgundy looks, well, alright. It is, by no means, a bad looking game. There are games that aren’t super pretty, and then there are games that are straight up ugly, and Castles is in the former category. It’s really not bad, but it does feel kind of bland. Known to some as “50 Shades of Green,” Castles of Burgundy has a muted palette doused with greens and grays and browns. I do have to say that I enjoy the earthy feel of this game; with no dramatic artwork or flashy colors, combined with a soft earth tone, the game is very easy on the eyes. That being said, this can be an issue under the wrong lighting, where the yellow tiles can be easily mistaken for the light green ones, and so on. In more than a couple of occasions, I’ve accidentally misread the colors represented in the game’s iconography, leading me to use tiles that I shouldn’t have. Whoops.
The artwork isn’t bad, but it doesn’t really stand out. In fact, it has sort of a rough quality about it that makes it seem like everything was drawn with colored pencils. This may appeal to some, but to me there was nothing really remarkable about it. I suppose that’s how I feel about the whole thing from an aesthetic standpoint. It’s just, “eh.” It’s not bad, and it’s not that great either. It’s just kind of…there, and I don’t have strong opinions about it one way or the other.
The player boards are made out of cardstock. While it’s nice to see cardboard player boards, even games with premium quality components (Eclipse comes to mind) tend to go with cardstock. It works, and you already know if you’re the type of person that would want to laminate or not if you’re worried about wear and tear. Aside from the chits, which I’ll get to in a moment, the game’s point/turn markers are wooden circles, which is always nice to have when plastic is the alternative. Each player also gets two dice of their color, all pretty standard, nothing too interesting to say about them.
There are a lot of chits in this game, like, a lot. Between the tiles, silverlings, goods, and additional point tiles, there are 292 little pieces of cardboard in this box. 164 of these are the little hexes you’ll be playing with. They’re made out of thinner, smoother cardboard, not the thicker textured stuff that you’ll find in, say, Fantasy Flight games. And that’s okay. There are so many chits here that them being smaller and thinner isn’t really an issue. They do their job well, and that’s all we can ask of them. Overall, this game is right in the middle for me with components. They’re not the top quality in the market, but there’s also nothing to complain about.

The rules are well written, and are designed to be easy for newcomers, as well as veterans who might need to quickly reference something.
One thing I love about this rulebook is the quick reference bar. The game explains the rules in detail, but has a sidebar on every page that gives a brief rundown all the important points. This makes it easy for veteran players to quickly solve a rule discrepancy without having to dig all over the rulebook. It’s just the worst when you have to scramble through a million paragraphs to find some tiny rule clarification amidst a mountain of text, and Castles avoids that problem with its rulebook design.
Oh, and did I mention that it also comes with a rulebook in French? If you were on the fence about this game up until this point, fear not, it does have the French rulebook. What more do you need?
Value
Is it a game you can play over and over? Are there expansions available for the game, and if so, are they necessary? Does the amount of the content in the box justify the price?
The great thing about Castles is that it’s easy to play, but teeming with layers of depth. For two people who want to play casually, the game is perfectly viable, and on the other hand, there are plenty of ways to hone your strategies and learn from what the game has to offer. The more you play, the more you’ll find clever ways to maximize your points in ways that you hadn’t thought of before. This adds longevity for competitive players, but still allows it to be perfectly playable for newer audiences. If you have doubts about how much mileage you’ll get out of this game, know that it’s a game that doesn’t get old quickly. I know people that have played the game for years and still enjoy it, and, although I have not owned the game for so long, it shows no signs of getting old.
Conclusion
Castles of Burgundy represents the best of what eurogames have to offer – elegant simplicity riddled with tactical depth. You have options upon options upon options, and will have fun discovering ways to make your progression ever more efficient with each play. Castles scales remarkably well and is one of the best two player games you can find. The game is certainly non-aggressive; you mostly focus on your own board, only able to influence your opponents in minimal ways, so offensive, interactive players might find this to be their cup of tea. Furthermore, if you’re looking for a “no luck” euro, you could do a little better than Castles. Tile availability is here and there, and while the dice are well-balanced, there are occasional flukes. The aesthetics and theme aren’t a strong point, but the quality of this game is in its gameplay, and in that it doesn’t disappoint. If you don’t mind a cake that’s not well decorated, this is a tasty one indeed. Overall, I highly recommend Castles of Burgundy, and for the cheap price at which it can be found, the game is a gem.

Check it out
Verdict
YOU WILL LIKE THIS GAME IF…
- You don’t like aggressive games
- You like thinking games
- You want a good two player game
- You don’t like games that feel punishing
- You enjoy having lots of different decisions to make
- You want a game with high replayability
- You’re looking for a cheap game
- You like balanced gamese
YOU WON’T LIKE THIS GAME IF…
- You like strong player interaction
- You like highly thematic games
- You don’t like excessive iconography
- You want a “hard” game
- You don’t like dealing with dice
- You don’t like dealing with lots of components and pieces
About the Author
Zach Hillegas
Zach is an avid tabletop gamer, and he created Board Game Resource out of his love for the hobby, and his desire to see more people come into it. When he's not writing for or managing BGR, Zach might be hanging out with cats, hiking a mountain, spending time with his lovely wife, or writing about video game stuff for Insert Gamer. Zach has also enjoys creating digital character art. You can check out his (long neglected) gallery here, or follow him on Instagram at @artworkbyzach!
Comments
A set of bags for the hex components VASTLY improves both initial setup and inter-round setup. Having the hexes hidden means you don’t have to turn them all face down and mix them up—just draw from the bag. It probably cuts 15 minutes or so from total play time. BGG has a color-coded set at a very reasonable price.
Great review, very comprehensive!