In the boardgaming world, a game is represented by two separate yet equally important groups: the designers, who come up with a good idea, and the developers, who make those ideas work. These are their stories...
Well, actually, it's just my story – from the viewpoint of the developer. (Designer Ted Alspach has already written about the game design elsewhere.) For about a year now, I've been working on a great game, now called Suburbia. It is a tile-laying, city-building game that brings a lot of the feel of the computer-based city-building simulations onto your tabletop. I was asked to write up a little column about how the game changed while I worked on it. Rather than double the length of this piece, I'll first refer you to the rules of the game. (Editor's note: Alternatively, you can view my game preview video at the bottom of this diary. —WEM)
When I first heard about the game, it was called Meeptropolis. Seriously. I mean, you can't even say it without shaking your head in dismay. I was walking around the Messe halls in October 2011 at Spiel when Ted Alspach pulled me aside because he wanted to "show me something". We were in a huge and crowded public space, so I figured that it would be safe...whatever it was that I was being shown.
That something turned out to be Meeptropolis. Ted explained that he had been working on a city-building game for some time, but he felt that it needed a second set of eyes to look at it. Ted explained the basics of the game and asked whether I was interested in looking at it more as he needed a developer to make the game shine. He thought that there might be a lot of parallel with the development of Meeptropolis to my previous development work on Dominion because the big goal here was to look at the huge number of tiles that he had designed and figure out how to balance them and choose amongst them to make the game work.
I told him I was interested, and Ted emailed me the current draft of the rules. Just based on the initial elevator pitch, I knew I was interested in getting involved with this game. The city-building theme has always been one that I've been attracted to, going all the way back to Metropolis by Sid Sackson – which was one of the first European-style games that I played. There were even two releases with a similar theme at Spiel 2011: City Tycoon and Cité. Though there are many variations on the theme, I love the way that city-building games give you lots of pieces with which to build your city, and you have to figure out how to get all those bits to work together in the game. Despite my love of the theme, I'll admit that I still hadn't found one single game that worked in all regards; these city building games always ended up being too fiddly, too long, too AP prone, or something. The chance to work on a city-builder myself was definitely a challenge I wanted to take on.
Initially Ted felt development would involve just these four things...
-----1. Are the tiles priced and powered correctly? -----2. Are there tiles that don't work with the rest of the set? -----3. Some tiles in each stack are designed to work together, e.g., restaurants/farm/slaughterhouse and office building/supply store in A, airports in B, schools/university in C. Should any of the sets be removed? Should there be more of that? -----4. Goals: Are the goal tiles appropriate for their values? Are any too powerful/difficult when hidden?
While these four things were the most important issues to be tackled, we ended up working on dozens of other things along the way before we got to the finished product!
The first thing I did was read the rules, build my own copy of the game (cutting out over 160 f*%#ing hexes by hand), and get my group to play it about five times. Once I had some experience with the game, I sat down and made a list of things I thought about the game. I broke down the game into the different components:
-----• Overall thoughts -----• Rules issues -----• Component issues -----• Issues with city tiles -----• Issues with goal tiles -----• Review of all the existing city tiles and my thoughts/questions about them
My initial list was long, nine pages long. Rather then spew it all out here, you can see the actual memo – well, Manifesto – yourself. I know that a lot of the stuff there won't make complete sense because you don't have the original version of the game to refer to, but I think you can see that I was trying to approach the game from every direction imaginable.
The first version of the game was solid; all of the rules worked, and the tiles behaved with one another. However, it actually wasn't long enough. The first draft didn't actually give you enough time to build a city – only 12-14 turns. Since you add only one tile per turn, I felt like you had to get a little more out of the game for it to feel satisfying. That was an easy thing to fix; by simply adding more tiles to the initial set-up, the game lasted a few rounds longer, and it was much better as a result.
Another aspect of the game which wasn't great was the fixed game end. Originally, the game ended when the final C tile was flipped over – and this ending always happened at the same time. The problem was that this made the endgame bog down and made some of the goals feel anticlimactic. So, in addition to adding a few rounds onto the game, I thought to make the game end be a little variable – ending in a three-round window now. The game length is currently at 15-19 turns, and there is definitely some tension now near the end of the game because you're never quite sure when the game will end. You have to pay a bit more attention to your status with the goal tiles. Additionally, there is a bit more of a risk/reward calculation now: Do you go for guaranteed points with a green tile, or do you risk the game lasting a round or two longer and go for a tile which increases your reputation?
Another change that came up was changing the tile from "End of Game" to "One More Round". In the initial version of the game, the "End of Game" tile, when drawn, finished the game at the end of that particular round. Thus, if it flipped over on the turn of the last player in turn order, the game ended immediately. After extensive playtesting, this proved to be unsatisfactory. The reason for this was that the quicker ending to the game gave the players later in turn order too large of an advantage in achieving the goal tiles because anyone who had a turn after the "End of Game" tile came up knew they could try to achieve a goal tile with those players earlier in turn order not having a chance to respond. Now, of course, functionally this still happens as those later in turn order take their final moves after the earlier players...but at least all players get at least some notice that the end of game is coming and can possibly do something about it.
So, to recap the changes at the end of the game through development:
-----• The game became longer, from a guaranteed 12 rounds to 13-16 for 4p, 16-19 for 3p, and 20-23 for 2p. -----• The end of the game is somewhat variable to prevent people from being able to fully plan for the endgame. -----• Despite that, there is a one round warning of the end of the game to give players a chance to wrap things up.
Of course, it wasn't just the end of the game that changed through the development process; there were plenty of other little tweaks that Ted and I discussed ad nauseum. In my Gmail inbox, I have no fewer that 415 threads (close to 2,000 emails) in which Ted, I and other playtesters discuss the game. Going through these emails, a few other important changes come to light.
There was a fairly robust discussion about how to discard tiles on turns that players did not buy from the Real Estate Market. The first version simply forced you to discard the rightmost tile. This had the major advantage of being easy to apply – but it seemed that we were losing out on a fair amount of strategic play by taking the choice away from the player. After a bunch of playtests and many email exchanges, we eventually decided that it was much better to allow players to choose which tile they wanted to discard. This change allows players to play a bit more defensively by being able to proactively remove a tile from the market that one of their opponents might benefit from. There is some risk that a really crummy tile might be left in the rightmost position forever, but since all players have the same choice of what to discard, it wasn't seen as that big of an issue.
Another interesting discussion was trying to figure out whether the initial player order needed any balancing. Many games offer players later in turn order a few extra dollars or some other material advantage to make up for being later in turn order. However, in Suburbia, there are advantages to being both early and late in turn order...well, at least that's what we decided after much discussion. At the start of the game, there is some value to going first; you have an earlier look at the tiles in the Real Estate Market and are able to make an earlier buy if there is a good combination available. How big of an advantage is this? It's honestly hard to quantify, and Lord knows that we tried to put a number on it. Luckily for us, there is also an advantage to going last – and it was also hard to quantify that. Namely, the later you are in turn order, the better positioned you are to lock up a goal tile bonus because you are able to make your final move knowing that some of your opponents are already done for the game. In the end, these two advantages generally cancel each other out, and this makes it easier for us because we don't need any specialized set-up rules, and everyone gets through the game on essentially level footing.
Another important change was to limit the number of "basic" tiles available to be purchased in each game. The initial version did not limit these (other than what was in the box). Now, only four available basic tiles of each type are available. There are a couple of reasons for this. First, there was a rare degenerate strategy that could be pulled off by buying nothing but inexpensive Suburbs tiles, then hoping to pick up a few High School tiles from the C stack. That wasn't fun, and by limiting the number available, it's impossible to pursue this strategy successfully. Also, Community Parks can be crucial in the endgame because of their +1 Reputation bonus for neighboring tiles, and limiting the number available creates a decent amount of time pressure in acquiring these tiles.
I guess that some of you may find it interesting to see the original version of the rules. (Well, at least the original version that I had to work with.)
Well, the biggest project here was playing with the tiles. Ted had already created a humongous number of tiles, and like Dominion, the task at hand was choosing from among the vast selection of tiles to come up with a set that worked together well in the game. I honestly wish there was a sexy story of how I came up with the tile set in use, but in reality – and I think this is a reality of game development – it is the result of nothing but a lot of hard repetitive playtesting.
For about six weeks, most of my freetime was spent setting up the game on my dining room table and playing through mock 4p games. Sometimes I would play the game with no specific strategy, and other times I would try to "break the game" with off-the-wall strategies and tile combinations. With each game, I would take notes about the tiles, and sometimes pull out my trusty Sharpie marker to doctor a tile or two to see how it would work if the cost or tile effect were changed. For the most part, the tiles you will find in the final version of the game were all amongst the initial set that I received, though many of them have slight alterations to the cost or effect. Suffice it to say that hundreds of games went into tweaking the tiles in oh-so-small ways.
What did change a lot with the tiles is their appearance.
The original tiles that I had were spartan, with lots of empty space that could be used for better purpose. You can see them in the top row of the illustration below. They were functional, but we ended up changing plenty of things. To compare, here are some of the final tiles below:
The changes that we ended up making included:
-----1. The title was moved to the top and is in larger print. -----2. The cost is easier to read in the left-hand corner. -----3. Related tiles (such as the restaurants) now have a restaurant icon in the right corner. -----4. Each of the color of tiles also has an icon associated with it (seen below the cost) – which helps with color blind issues. -----5. The banners at the top and bottom were placed there to make it easier to print out tiles in different languages at a minimum of work/cost to the printer.
In the end, I think that the changes to the format of the tiles will have as large an effect on the playability of the game as the actual choice of the tile. The tiles are much easier to read now, and they are more easily identifiable from across the table. The addition of the icons to the right side is a crucial change so that all the players can see which tiles interact with each other at a glance.
You can also see some of the subtle changes made in the details of the tile. The Fast Food Restaurant's long term effect increased from +2 population to +3 population per adjacent green tile. You can also see that Freeway increased in cost from $3 to $5. There were dozens of little changes like this along the way, but trust me, it would be a complete snoozefest for me to list all those changes for you.
I feel that component quality and good ergonomics go a long way in determining someone's happiness with a game. There were lots of things that I found through repeated playtesting that improved upon the initial version of the game that I was given. As I showed above, the tiles went through some significant changes – but that's not all that changed!
One of the other things that changed was the borough board itself. The original version was a simple rectangle with a green bar in the top center to show you where to put your original tile. It worked...sort of.
Almost all of my initial games of Meeptropolis had at least one city catastrophe in which the board got jostled and all of the hexes moved around. I made a simple suggestion to add hex cut-outs to this board which helped lock the hexes in place.
Another change you can see here is a streamlining of the scoring tracks; now there is a single number between the two tracks. Additionally, the iconography has changed a bit with squares representing the reputation and circles noting income – the same shapes that we use on the tiles for these two things.
The scoreboard also went through a number of changes. The initial scoreboard tried to put everything in one place. The scoring track went around the outside, while the tile market, stacks of tiles, and goals were all found in the center.
This arrangement ended up not working on a number of levels. First, it was way too easy for things (especially scoring markers) to get jostled. Having to reach over the scoring track each time you wanted to take a tile was treacherous. Additionally, the A/B/C stacks were always at risk of falling over and disrupting the upper left corner of scoring.
We went through a number of different trials before realizing that it would be better to have two different boards; by putting the scoring track on its own board, this reduced the chance of it being jostled. Additionally, by moving the Real Estate Market to the bottom of a board, it made it much easier to remove tiles from the Market as well as easier to slide all the tiles to the right when adding a new tile. You might also note that the original game had only two goals in play, and this has since been changed to one community goal per player in the game. The two boards in the final product are below.
Finally, there was a lot of time spent on the rules. We had lots of mundane discussions about word choices, appropriate example situations to be put in the rules, and the use of the Oxford Comma. (This is a humorous explanation of the Oxford Comma.) A few notable changes that I suggested include:
-----• Separating the rules from the tile explanations. By putting the tile explanations on a separate sheet, this can be passed around the table for all players to reference as needed. -----• Adding in as many rules examples as possible to help illustrate difficult concepts and/or non-intuitive situations. -----• Most of the other rules changes are graphic in nature, and I think Klemens Franz will talk about those later.
The Solo Game
The last big thing that I had a chance to work on was the Solo Game. I've always been a fan of games that include a solo play option, so I knew that I was going to try to sneak this into the rules as I worked on it. From the start, there were two main options to consider: 1) allowing players to play the game on their own while trying to maximize their score OR 2) creating some sort of robot player for the player to compete against. In the end, we decided to do both!
For the first version, the Solo Architect, the game is set up like a two-player game though no goal tiles are available. (After going over the goals, most of them didn't make any sense without competition, so they were left out of this version.) Given that you set up stacks of only 15 tiles each – less than half of the total available – I wanted there to be some way to still be able to work with the tiles. Luck would still have a big role to play given the inherent randomness of using less than half the tiles, but I needed to give the player some interesting choices. In the original version, the solo player would buy a tile, then discard the rightmost tile that remained. This proved to be a little dull, so we decided to let the solo player choose which tile he wanted to remove from the Real Estate Market. This gives the solo player a fair amount of control in deciding which tiles make it to the free area at the right of the Market. There should still be times when you need to pay for tiles (to get their effect at a more beneficial situation in the game), but there's room for skillful play as you choose which half of the randomly drawn tile set you will add to your city.
However, even changing the tile choice rules, the game still dragged a bit – mostly because players were able to too easily maximize their Population and Reputation without other human competition. Thus, the penalty for crossing over a red line on the scoreboard was increased to -2 on each track. This change causes players to continually improve their city at a rate that feels like the multiplayer game.
The other solo game is a game against Dale the Bot. What I tried to do here was create a framework of rules to allow you to play against another player. By adding a dummy player into the mix, this gives you another city to interact with, which changes the ways that you approach the tiles as some of them definitely work better with other cities on the table. Also, by providing you with an "opponent", it allowed me to re-introduce the goal tiles to the solo game.
Okay, so I'll be the first to admit that my namesake Bot is really kinda dumb. Well, that's maybe a harsh thing to say since the solo player will be the one actually controlling the Bot. The Bot follows a simple rubric that helps it decide how it will buy a tile, then four simple rules to help it decide where to play that tile. When the Bot has to buy a tile, it buys the most expensive tile it can find; when calculating the costs, it combines both the base price on the tile as well as the cost above the tile on the Real Estate Market. The costs from the Market had to be added in to prevent the Bot from buying the same super-expensive tiles in EVERY game as well as making it a bit more diverse in its purchases overall. The assumption here is that higher valued tiles generally are better tiles to have – though this is clearly not true in all cases because cost alone does not take into account how the tile would interact with the other tiles already existing in your city. In order to level things out, Dale the Bot spends only $3 regardless of which tile he buys.
Once the tile has been acquired at a low discount price, then the Bot follows a simple progression of rules to place it. The new tile must be placed:
-----• in a spot that increases both Reputation and Population if possible; if not... -----• in a spot that increases Reputation only; if not, -----• in a spot that avoids a decrease to Reputation, if possible; if not, -----• in a spot that increases Income.
As long as a tile can be placed within a category, it can be placed in any valid location that meets the criteria for that category. This amount of latitude gives the solo player a challenge in being able to choose among several different locations for most tiles.
Since the Bot generally makes plenty of money in this game – mostly due to the low bargain costs for its tiles – any deficiencies in tile choice and placement are made up for with a huge endgame bonus of VP when money is converted to points. Clearly, Dale the Bot isn't going to win any Suburbia tournaments, but it does give you the feel of a two-player game with someone to compete against. I'd say that veteran players will likely beat the bot four out of five games, but beginners will be closer to 50-50 with it.
Well, that was a pretty sweeping survey of the different sorts of changes that Suburbia went through over the past year. Unbelievably, this list is in no way comprehensive, but I better stop here because the deadline to get this turned in to BGG is close, and I've got some Spiel Preview pieces to write for The Opinionated Gamers. I hope that you've found it interesting, though, to see how much a game can change through development.