Jeff Warrender(jwarrend)United States
After a few years bringing the game to our designers' gathering, Spielbany, I realized that the game’s length and complexity just weren’t appropriate for that setting – the game hogged too much table time. So, I began hosting dedicated sessions for the game, and it was in these that we really hammered the game into its final form. Principal players in this stage included P.D. Magnus, Dan Purdy, Doug Hoover, TauCeti Deichmann, Mike Pearsall, John Hornberger, Quentin Hudspeth and Dean Howard.
Getting out of the players' way
The most significant change that this group presided over was the reworking of the action system. As I mentioned, by the end of the Spielbany era, there were 8 actions arranged in a 4x2 grid. Your turn consisted of taking one action. There was an achievement token reward associated with the column you chose, but you incurred Unrest if in a subsequent action you re-used an action in the same column (originally, this was outright forbidden). As a player, I found this frustrating; many of the actions benefit from or require combinations with other actions – for example, you want to move and then attack, or produce and then build, etc. In pursuit of short, punchy, single-action turns, I was actually getting in the way of what players wanted to do, and I came to see that I had to let players take two actions per turn (although, initially, I couldn’t see a clean way to accomplish this with the 4x2 board.)
I can’t emphasize how difficult it was for me to see this, and yet how important. It can be incredibly hard as a designer to take off the “designer” hat and see things through the eyes of an unbiased player. In all of my games, not just this one, there have been many playtester suggestions that I resisted before eventually accepting, because it’s just so difficult to see what the experience for the player is like when you’re committed to seeing what you think it ought to be like – what you’re sure it will be like when the game is done and everyone who’s playing understands the rules and so on.
Anyway, as a result of this realization, I changed the action mat to a 3x3 grid, with each box representing one action. On your turn, you placed a marker on an edge between two actions, and then used both of those actions (in either order). Each edge was marked with one of the three civilization categories or an Unrest symbol, so when you used a particular pair of actions, you received the reward or penalty appropriate to that particular pair. I still think this was pretty clever, and it worked well. Thematically, it mostly preserved the idea of different “prefects” ruling over your empire. I envisioned the action board looking perhaps a bit like the council board in BattleLore: it would depict the ruler's throne room and twelve different prefects waiting to do the king's bidding, each prefect bordering the two different roles that that prefect specialized in.
During this phase, length was still an issue; my wife and (at the time) young kids endured quite a few sessions that ran well into the night despite my assurances that, surely this time we'll wrap up by bed time! I had hoped that this change to a 2-action-per-turn system would make the decisions quicker but it wasn't to be. I thought that if we stripped out all of the extraneous bureaucracy, all of the status phases and state checking and such that similar games rely on, that surely we could get the game shorter. All of those things helped (I think that Sands has a higher "up-time ratio" than just about any other game in this genre) but the game length remained stubbornly long. Two minute turns were typical, and that added up to a lot of minutes.
Simultaneous action selection could, we hoped, speed things up, but that couldn’t be done with the action board. (*) This drove the change to the card-based selection system that is in the game now. Of course, thematically, it's a small step back. (I guess we could think of the action cards as “edicts” that the player is issuing to his empire), but in the most important way -- controlling the game length -- they've proven to be a big win.
In the action system, players simultaneously select two action cards, one of which can instead be an “emphasis” card (see the next post). Each indicates one fairly short and punchy action that you get to execute. You can take the actions in either order. Additionally, you can use a single “bonus” action (see next post). The game turn or “generation” lasts 2-4 turns, and in each turn after the first, you have the option to re-use one action card that you previously used, but must increase your Unrest by 1 at the turn’s end if you do.
Perception is an interesting thing. To me, and probably to most of the core playtesters, the card-based action selection mechanism is simply the culmination of various iterations over possible ways that actions can be selected. It never seemed, to me, to be the beating heart of the game, it’s just the mechanism I stumbled upon that gives the players the best experience in selecting their actions. However, to Uli, and to many of the people who have played the game, it’s the aspect of the game that has resonated most strongly. Players seem to really enjoy it. And I guess, of all the systems in the game, it’s the one that’s probably the most easily ported to another game design. I have no specific plans to do this, but game designers who are reading this, take note: you have my blessing (not that you need it) to use the action system in your own game!
(*) Grr…I actually realized while writing this that, actually, yes it could have: in a 3x3 grid, there are 12 internal edges between boxes. So, number each box, and give each player a 12-sided die, and they all simultaneously pick a side of the die, then resolve which box they have chosen. That would have worked pretty well. Ugh! But the card system does have some real advantages – more flexibility, and makes the Emphasis concept possible; and anyway, there wouldn’t be space for the 10 actions in a 3x3 grid so something would have to go, which might be tough at this point.
Trade routes, or exactly how many different things can we cram into this game without it feeling overwrought?
One other important change in this era pertained to the trade routes. I originally wanted the game to have some cooperative interaction, and in fact, have it be important enough that it could be a scoring category unto itself. In the very first proto one of the scoring categories was “who has the most population currently residing in foreign territories?” This wasn’t motivated well by the rest of the design, though; why did you want your population in foreign territories? It was just something you were doing because the game’s scoring system was telling you that you should be doing it, but it wasn't actually particularly interesting.
This morphed gradually into the idea of trade "connections": if you built a city that was adjacent to a foreign city (and remember the funny effect that Roads can have on this), you were eligible to receive an achievement token from that adjacency when you took a certain action. Also, one of the scoring categories pertained to how many such adjacencies you had. This system was unsuccessful in two ways. First, the pendulum almost always swung too far in the "easy" direction: you would often have many adjacencies, so you could easily get a lot of tokens. Second, players’ positions ended up very homogeneous: because of the board layout, most players all ended up with about the same number of adjacencies; it wasn’t something you could (or even needed to) target, so it was weird to get points for it.
The solution to this came from a suggestion John made about having trade routes physically appear on the board. I expanded on this idea to where they’re a network that you have to grow through deliberate placement. It solved a problem the action board-based action system presented – what do you do when you don’t wish to use two actions? You place a trade route! But it also enabled, for a time, a distinction between the “resource” aspect of trade routes (getting tokens or, later, heritage) and the “goal” aspect (getting VPs for your trade routes). There were different colors of trade goods and you counted variety of colors for one of those but number of a single color for the other -- it was really too much complexity, and even my most hard-core playtesters found it super confusing. The distinction is somewhat less in the final form – a trade route connecting you to a trade good gives you a heritage boost AND counts towards a scoring category. But the advantage of this approach is the asymmetry it creates between players: your trade route network is your own.
These two changes were nearly sufficient to get the game into its finished form. It just needed one small-ish tweak to the action system and one big tweak to the Advance system, which are what the next two posts will cover.
A blog about the development of The Sands of Time, from Spielworxx
- [+] Dice rolls