Jeff Warrender(jwarrend)United States
The exact origins of Sands of Time have, sadly, been buried by the sands of time; they are chronicled in the pages of an obscure notebook that I’ve since misplaced, but presumably someday will be unearthed from a box somewhere in my attic. Until then, my recollection is the best source that we have! What I can say conclusively is that I began work on the game in 2003, and the first year of Sands’ existence took place in Waltham, MA, at a now-defunct game store called Danger Planet. The regular players included Tony Farrand, Matt Farrand, Chris Garman, Vitas Povilaitis, Rich Gentile, Andrew Wright, Cindy Shettle, Preston Fuller, and Karl von Laudermann, and these are the unsung heroes of the game. Sands was looooong, and while no one ever felt that it was terrible, in retrospect, it was a cluttered and confused design in its early days, and it took a great deal of excavation to uncover the game that the core ideas were grasping for.
I’ve written elsewhere that Sands was inspired by Bruno Faidutti’s mention of the (at that time) new game Mare Nostrum – he billed it as a civilization game that didn’t take a whole day to play, and that it was the life’s work, 20 years in the making, of its designer, Serge Laget. As a new designer, I thought that this sounded like a neat challenge to attempt myself, and thought “20 years? No way it should take me that long!”; I haven’t crossed the 20 year mark but it’s been well over a decade so far, so it’s a much harder challenge than I perhaps realized at the outset!
A few ideas have been in there since the beginning, among them:
Unique map At the time I started work on the game, it seemed like a lot of games (not just civ games) had a board that was basically "plan-view of the Med". I wanted a map board that just looked different than that. I happened upon the "Orbis Terrarum", and latched onto it from the earliest stage of the design. It's come to have some nice gameplay benefits as well.
Few territories I judged, correctly or incorrectly, that a lot of Civ's length comes from manipulating your many pieces on the many territories of the map; I conjectured that having fewer territories to manage would lead to less bookkeeping and faster play times.
Production based on population I wanted the "number that goes with a territory" to be a limit on the territory's capacity for population, as opposed to the automatic number of resources that a player receives by virtue of owning that territory. It's a potential (and a limit), so bigger territories aren't necessarily "more valuable", although they can be.
Few resources One or two max. I didn't want a system a la settlers where it's "combine two of these and one of those to build this thing that seems like it might require those things as inputs." Nothing wrong with "resource combiners" of this sort but there are plenty of them and I didn't want this game to be one of them.
Few unit types Two max, peasants and warriors. We've stuck with this the whole way through.
The idea that the scoring system reflected historians chronicling your great deeds goes back to the very first iteration, but it’s been realized in a variety of ways. Originally, there were 4 scoring categories, and each scoring round VPs were paid out based on rank in these categories.
Two ideas didn’t survive, but certainly had an influence on the game’s later development.
The first was the way the board was laid out. Out of a desire to de-clutter the board, the map at the center would contain only player pieces, and each territory had an “information box” at the edge of the board, which contained three spaces: one for a tile showing the capacity, one for a tile showing the resource produced, and one for the owner’s marker. Structures AND advances were cards that were added to a territory by placing them at the side of the board adjacent to their territory’s box.
This led to a practical problem pretty quickly – calculating resource production took forever, because you had to keep looking back and forth from the territories to the information boxes to remember what each territory produced. The solution we came up with was to add “production tracks” which told you how many resources you produce at the start of each turn; and, you would count up your production only when a “Census” event happened.
This still didn’t help the fundamental problem, which was ease of play, so eventually we changed the board, removing the information boxes and placing resource and capacity information right onto the territories through tiles. But curiously, I didn’t get rid of the production rule and the tracks; they stayed. However, we realized that increasing production only on a random event was too unpredictable, and so instead, we added a rule that let you use one of your turn actions to re-calculate your resource production. Eventually, years later, one of the players asked “why is this production calculation rule in here, anyway?”, and the actual problem it solved was so many versions removed that I realized we had kept it all this time, and even built new rules around it, for absolutely no reason.
So if nothing else, this is a good cautionary note for designers!
The second that each territory you own gives you a “personality card". In the beginning there were nine action phases, resolved in order Avalon Hill-style. Each personality card had three abilities, each of which correlated with one of those action phases. Each ability had a number (3, 2, or 1), but the numbers were randomly distributed among the abilities on a given card, but always with the same sum. So on one card, maybe it was Strategist 3/Builder 2/General 1, and on another it was Strategist 2/Builder 1/General 3, and so on. You could use each card only once, so you had to decide whether to use the ability with the 3 that maximized the value of the card, or the ability that most corresponded to the actions you wanted to use, whatever its value. (At that time, each Structure also had a particular ability and level that it required, so you could instead use a card to enable you to build a particular structure; I think other games have since explored this idea). We didn't keep this system but it certainly influenced the way I thought about the player actions.
Game length and the first step toward an action system
That system went away shortly after the game underwent its first significant action mechanic overhaul. The game was consistently running 5 hours or more, just way too long, and the pace of play was too slow. I recall one session that we had scheduled early on a Sunday afternoon, and the game ran so long that it overlapped a Patriots playoff game. After taking their turn, the other players would rush out to the TV room to catch up on the football game, then come back to the table when their turn was up. "Sands of Time -- less fun than watching football!" did not sound like a good selling point, and I realized we needed to get the length way down.
I realized that having 9 actions in order, with players all deciding whether to play or pass on every single action gave too many decision points, and decisions are what slow the pace of play. It doesn’t matter how quickly an action can be resolved – any time you give a player a decision, it adds time (how much depends on the player). So what I instead did was to build those actions into a 4-by-2 display, with 8 actions available for players to choose, and on your turn, you simply picked an action from the display, BUT, originally, you couldn’t pick an action that was in the same column as an action you previously used – later, we relaxed it to where you could use both actions but had to take on unrest.
This was also, I think, a big step forward in the thematic solidification of the game. The idea was that each column represented a different “prefect”, and you were going to choose three different “prefects” to rule over your empire that year (or decade or whatever), and each could do two different things, so you were telling him which of those two things you wanted him to focus on.
The actions were thematically arranged; so, for example, “Conquer” and “Govern” were opposed – you could either look inward or outward, and so on. At this point, the game didn’t have Chronicles, caravans, heritage, emphasis, or many of the other systems it now has. It was a nice distillation of the complexity of the early design into a simple, intermediate design. But it lacked a lot of effects that I wanted to include, and gradually, over time, things were added in. This has been mostly for the better, I think, but certainly some of those have added some complexity.
As the next posts will further detail, much of the game's development has been struggling to find this "perfect" balance between complexity of rules and ease of play(*). I think the final design is the best compromise between these of any of the iterations; it's balances between strategic depth and a clean interface. It's not the 2-hour civ game I initially thought I was trying to create, and the funny thing is that most of the early playtesters mentioned above knew I was never going to hit that goal, and chuckled whenever I insisted "no guys, seriously, we can get through this in two hours!" They were right in the end!
(*) And this has been in parallel with trying to balance between the game being neither too loose nor too tight. That's a tricky thing indeed in a long game with a fair bit of complexity; small changes can really swing the pendulum one way or the other.
A blog about the development of The Sands of Time, from Spielworxx
- [+] Dice rolls