/pic1026447.jpg)
/pic1003636.png)
(Editor's note: Asmadi's Chris Cieslik has announced that the retail price of Fealty will be $30 instead of the previously anticipated $40, and the funding levels on Kickstarter have been adjusted accordingly. Also, the two $1,250 pledges – which include hand-delivery almost anywhere in the U.S. – have been claimed, which surprises the heck out of me; aim high, future Kickstarterers! —WEM)
I've broken this post into two pieces: discussion and timeline.
Part 1: The Easy and the Hard
Easy: The core concept. I generate way more game ideas than I'll ever have time to work on. Sometimes they start with a theme; sometimes with a mechanism; sometimes with a vision of game-flow. Fealty was born from the marriage of a mechanical concept (pieces radiating territorial influence) with a game-flow concept (fewer but high-impact decisions). Many of the central ideas came together the first afternoon, and playtesting quickly revealed what people liked most: placing pieces in the face of interesting constraints. Some constraints are created by your own plays (as you cannot play into a row or column where you already have a piece), some by the opponents (as each Duchy can be played into only once per turn); the player who can best work within those constraints to position themselves against their foes will win.
Developing that central kernel into a balanced, engrossing game took a great deal more time and effort than discovering the heart of it.
/pic1056460.jpg)
Each set has just nine pieces. How hard could it be?
It turned out to be one of the trickiest balance problems I've tackled in a long time. Changes to one piece altered the usefulness of every other piece in the set, based on the likely board interactions between the pair of them; then those secondary utility changes reverberated and collided further, like ripples in a pond. For the base set, my use of special powers to tweak balance was somewhat constrained: there should not be too many different types of powers, they should all be fairly straightforward, and they should include the core concepts of "place Conflict" and "move another of your pieces" at least twice each; this made fine-tuning rather tricky.
Some prototype iterations,
leading to the Inquisitor
/pic1055743.png)
leading to the Inquisitor
In addition, I found that I had a fair margin of error when balance-testing against myself: at least once, I went down a long, painful dead-end because of a balance issue I'd found during self-testing that didn't manifest in play with other people. After that, I made preliminary conclusions when playing solo, but never took them as final until I verified them in real games.
We eventually found that an eight-turn game added a lot more meat for not too much more time. Plus, with eight turns, each player will eventually have access to all of their pieces, and play all but one. This made the balancing act less critical, though I still did much balance-testing using shorter six-move games to try to keep things tight. The breathing-room feels good, though; it's almost inevitable that hundreds or thousands of players will find patterns and nuances which mere dozens of playtesters might miss.
Easy: Iterating piece designs. This was easy because I deliberately made it easy - I find the surest obstacle to working on a game is getting too fancy too quickly. If creating a new iteration of components requires any notable graphic design work, it takes much longer to happen.
When Fealty became my top priority, I put in the up-front effort to make templates for cards and pieces that were dirt-easy to edit. The hardest part of making a new card was replacing the slips of paper in four stiff-backed card sleeves.
Eventually, when doing first-pass solo testing, I didn't even bother making cards - I'd just scrawl out pieces by hand, with cryptic abbreviations for special powers.
Hard: Coming to grips with final-move analysis. Through most of a game, optimal play is genuinely ambiguous - the move which would claim the most territory is often not the best choice. But the very last placement of the final turn of the game is full-information: the board state will not change further after it is done. Playtesters felt they should check every single legal play to see which was the best. We tossed around a number of ideas for fixing this, but everything which made the expected outcome of that final play uncertain enough to short-circuit the analysis also undermined the ability to make meaningful decisions about placement.
So we switched gears, and instead explored reducing the scope of that final-final-move analysis rather than making it ambiguous. This led to the "one play per Duchy" rule, which narrowed the scope of that decision - now the final play usually had 5-6 options. Checking to see which was best doesn't take long.
We didn't want a final-turn special case, so we tried the new rule throughout the game. It changed the feel substantially - slower pieces now had an extra disadvantage, and placement as a whole was much less efficient - but everything still worked, and the underlying dynamics still felt good. The last play of each turn did feel a touch too constrained... but the process of fixing that led to the deeper eight-turn game. Win!
/pic1056456.jpg)
Hard: Researching production possibilities. This is my first published board game, and I wanted to put a good foot forward, so while Chris was concentrating on the Innovation expansion, I looked into a variety of production possibilities to see if any might work well for the limited run of Fealty. Let's just say that I now have even more respect for game publishers!
Easy: Playtesting. It's amazing how much easier it is to playtest a 15-30 minute game than a 75-100 minute game – not just because you can fit in more plays per unit time, but because people are far more willing to commit themselves to half an hour than an hour-and-a-half, especially for something that may not be polished yet. They're also happier to play it at one weekly gamenight after another, which allows for more rapid iterations. And because I'd signed with Asmadi before I ever got to the point of blind-testing, Chris ran blind tests rather than me having to set them up myself.
There are a thousand and one things I've omitted - some because they were neither easy nor hard, but somewhere in between (e.g.: getting the modular maps working; discovering types of boards that work well/don't work well; working out how "place an influence immediately" powers should interact with cities) and some because they were so trivial that there's little to talk about (e.g.: cities being worth an extra point - it was one of a large list of tweaks to try, and in just a turn or two it became obvious that it was a keeper).
/pic1056547.png)
/pic1056548.png)
Part 2: The Diary
I started developing Fealty in October 2010 - one of several designs I was working on at the time. Around December 2010 or January 2011, Chris from Asmadi Games said that he'd like to publish it, and it took over pretty much all of my design cycles. Here's a (fictional but mostly-accurate) diary of the process.
-3 days: A notion has floated through my head: "Chess-like pieces which control territory instead of capturing each other."
The First Day
+0 minutes: Time to take an afternoon and work on one of these game ideas. What sort of game? I'd quite like a thinky game that plays quickly. Perhaps something with a lower decision count but high decision impact, like King of Siam?
+15 minutes: That territory-control idea! Many possible placements x many possible pieces yields greatly branching decision trees. I have scribbled down some tentative notions, based on a contested royal succession, because I am still thinking about Chess.
/pic1056473.jpg)
+35 minutes: ...what happens if Piece A neutralizes Piece B which commands piece C to neutralize piece A? Ah, crud.
+1 hour: Timing rules prove tractable, but the resulting game seems un-fun. The "commanding" dynamic means one minor misplay can result in several excellently-placed pieces having no game effect. That is a recipe for Annoyance Stew, not Tasty Gamer Nirvana.
+2 hours: Dropping the "commander" notion but keeping some timing rules yields interesting results: early-acting pieces can block mid-acting pieces, running interference for late-acting pieces. Vaguely reminiscent of the "My 1 blows up your 2 saving my 4 which blows up your 7 so you don't mine" dynamic of Light Speed, only less extreme and more carefully developed. I like it! And it plays well into the theme: minor players in the succession struggle will have very limited spread (purely local appeal), but can block the major players in their own small bailiwick by claiming first. For further differentiation, each piece will have an action it can use after placement - moving some other type of piece, perhaps, or replacing one piece with another?
+2.5 hours: To further keep game-time down, players choose moves simultaneously: each has nine cards for pieces, nine cards for rows, and nine cards for columns. I have cunningly convinced myself that two players placing in the same location shouldn't be a difficult edge case to handle. Ha ha ha ha ha.
+3 hours: Let's make the row/column cards a finite resource, so as the game goes on players are more constrained in where they can play. This adds the consideration of "not precluding future desirable plays" to the more greedy short-term planning.
+5 hours: I have a notional set of pieces and actions! Some of the archetypes seem like they should interact with specific terrain features - forests, or roads, or cities - so I add these to the board.
Components from early-ish prototype,
late prototype, and limited printing
/pic1056457.jpg)
/pic1056458.jpg)
/pic1056459.jpg)
late prototype, and limited printing
The First Month
+1 day: Plays OK with just me, and the core dynamic comes through about as I'd hoped. Time to throw it to the wolves.
+5 days: Playtesting accomplished! Shockingly, the players didn't like juggling 27 cards at once - should have seen that one coming. However, the constraint of not playing again in the same row/column was seen as very interesting, as was the core territory-contention dynamic, and playtesters had a number of excellent (if sometimes contradictory) suggestions on tweaks, re-workings, and ways to elimimate start-of-game information overload.
+8 days: Dropping the row/column cards permits use of modular maps. This is so patently the right way to go that I wonder how I missed it before.
+12 days: Updates accomplished, and self-testing looks good. But is this really the theme the game wants? "Medieval succession" is a very dead horse. Perhaps it should just be an abstract positional?
+13 days: Verdict: NO. Way too many concepts which want to be hung off of a theme. After brainstorming, a likely alternative emerges: a clash of different protest groups, all demonstrating in the same city. The street/avenue (row/column) restrictions are local crowd control ordanances; instead of terrains, there are different types of residents (veterans, union, etc) who may be be swayed by different types of protesters.
+14 days: Hmm, the information presentation on this city-block prototype is proving a bit tricky.
+15 days: More than a bit tricky. *twitch*
+16 days: After several days of work, this map is still so confusing that I can barely play a game against myself. Icons just aren't cutting it, and texturing makes busy patterns more suitable for hypnotizing the nation's youth than conveying useful information. Heck with it: I'll spend the effort on gameplay instead, and bring the originally-themed prototype to BGG.CON. Perhaps playtesters there will have some helpful thoughts on the matter.
+30 days: BGG.CON was awesome, and playtesting went well: reactions were positive, game-time averaged 20-30 minutes, and even people who lost terribly wanted to try it again. Players thought the theme worked fine, and one pointed out that it's a classic theme for a reason: lots of people like it.
Onwards
+45 days: Playtesters are optimizing the final move of the final turn in mind-numbing detail because it's a solvable problem. Brainstorming leads to a drastic solution: once a sub-board is played upon, no further pieces can be played there that turn. It's a huge change, but doesn't break the game – just shifts how it plays. It also adds new tactics: playing early purely to block off a board, or choosing which board you play onto based on which pieces your opponents have chosen. It also brings the number of legal final-moves down to a reasonable "analyze all of these" count.
+50 days: Enheartened, I have roughed out several sets of pieces. I don't expect they'll all be good - in fact, I strongly suspect some of them are hideously broken - but trying them out should be interesting, teach me more about how different types of pieces and powers work, and perhaps give me further ideas down the road. Asmadi has said that they'd like to include two sets of pieces with the base game.
+2-3 months: Extensive self-testing with the main piece set has revealed a critical gameplay flaw: certain stupidly greedy choices are competitive with more strategic play. Crud. I regroup and come up with a new set of pieces designed to handle the problem.
+3 months, 1 day: Extensive testing with other people has revealed a critical development flaw: when playing against myself, I am susceptible to groupthink. The problem I just spent a month fixing doesn't manifest in actual play, and instead of fixing it, I've developed a slightly dysfunctional piece-set and needlessly tweaked the rules. *facepalm*
+3.5 months: Testing continues. Pieces in base set almost balanced.
+4 months: Testing continues. Pieces in base set almost balanced.
+4.5 months: Testing continues. Pieces in base set almost balanced? This is proving trickier than nearly any other balance work I've done. Happily, it turns out that playing a slightly longer game (eight turns instead of six) is both much meatier and smooths out balance considerably; but we're still shooting to be well-balanced in a six-turn game.
+6 months: Tuning Fealty sets becomes easier with practice! Being able to use more complex powers also helps. The alternate set has come together nicely, and it looks like we'll be able to use it. Fantastic!
+6.5 months: Design and development has pretty much concluded, as Asmadi is moving the game towards production. Still want to test some of the variants I've had in mind which draw upon both sets of pieces.
+7 months: One obvious variant I did not directly design for - because I would have gone stark raving mad - is to have different players using different sets of pieces: some Missives, some Suns. But testing seems to indicate that it works! (So long as there are equal numbers of each, at least.) This is probably a side-effect of keeping power-levels comparable in order for other variants to work; but regardless of the reason, it's a lovely discovery to make.
+8 months: Published! Demoing at Origins 2011 gets very positive responses. Hurrah!
R. Eric Reuss
:strip_icc()/pic1059811.jpg)
/pic859250.jpg)
/pic1055546.jpg)
:strip_icc()/pic216929.jpg)
/pic1056877.jpg)
