Sankore: back to school in ancient North Africa [boardgame]

Games
1

For a guy who talks such a big game about being utterly uninterested in new boardgames, I am ashamed to admit that I just caved and bought a new game called Sankore. It grabbed my attention for two (2!) reasons. The first…

Sankore
Sankore1600×671 258 KB

I mean, look at that!

If you look at that picture and think “uhhh…heavy Euro?” then you’re right. But if you’re like me, you look at that picture and are immediately mesmerized by the palette, the detail, and the promise of all those icons and pieces. That’s Ian O’Toole for you, who’s pretty famous for boardgame artwork. I look at that and want to be drawn into the exotic sunny embrace of, well, whatever is going on there.

Sankore is one of those games that’s impossible to make sense of at first glance, but with a little thematic assist, you start to see through the detail, into the design. And once you learn the rules, it’s one of those boards that tells you everything you need to know, so just throw the rulesbook back in the box. I learned Sankore by reading through the rules, then setting up a solitaire game, at which point…well, first let me tell you the second reason I bought Sankore: for its solitaire.

The solitaire game isn’t just some automaton deck of cards, as is fashionable with most games that boast solitaire play (I have very strong opinions about the state of solitaire gaming these days, and they are NOT kind!). Instead, it’s a more holistic approach to model the arc of a typical game. It divides the automaton player’s actions into three discrete stages, each with different rules for how to proceed, each modeling the specifics of a typical early-, mid-, and late-game move.

I have no idea how well it works, but I’m fascinated by this approach, and it’s the main reason I decided to buy this game! If it’s doing something unique with its solitaire bot – this is NOT a solitaire game, to be sure, but a multiplayer game with an included solitaire bot – then I want to find out. Also, this is a game by a relative newcomer named Fabio Lopiano who’s got a number of credits under his belt already, so let’s see what kind of work he does.

Anyway, entranced by the artwork, lured by the promise of a unique solitaire bot, I got it set up and I sat down to make my first move. This is a game about running a madrasa in the 1300s in the city of Timbuktu, in the kingdom of Mali, under the fabulously wealthy king Mansa Musa. Your player board represents a curriculum of classes you run students through, and as the students learn and advance through your madrasa, they put into practice their lessons to put your pieces on the board and change the board state in your favor. Your astronomy students learn to finance trade caravans to secure salt from oases and trade with port cities. Your theology students assist distant madrasas to bring manuscripts to the Timbuktu library, which will determine scoring. Your mathematics students erect walls that are the game’s timer. And your law students improve all your other actions by adorning them with various bonuses, basically threading political power into the other disciplines.

What I probably most like about the design is how it begins completely open-ended: the game clock and scoring are completely up in the air. You have no idea how long the game will go, and you have no idea what will score points (it reminds me of one of my favorite boardgames, Archipelago). Instead, the game end and scoring criteria are determined as the board progresses and changes, as students take classes and graduate, as the great library at Timbuktu collects manuscripts, as the caravans fan out across Africa, as walls spring up around the city, and as legal scholars build up a rich tradition of special privileges. It’s all heady stuff, meticulously interconnected, and believe it or not, it bubbles up clear as a mountain spring from all that lovely detailed artwork and iconography.

So anyway, I’m sitting down to take my first move and I realize that I have no idea what to do. I mean, I know the rules, and I know how to make things happen, and I know why they’ll happen. But Sankore is one of those games that’s just so vast and sandboxy, even though it gives each player four very specific objectives to chase. I guess I should do one of those. But which one???

This is how many heavy Euros go, and in that regard, Sankore might feel familiar. It can seem utterly inscrutable at first, even when you know the rules, even when the board is clear as day, even when you’ve wrapped your head around its economy. If you’ve ever bellied up to a Vital Lacerda game, you’ll know the feeling. So…where to start?

I was paralyzed.

I needed someone else to be paralyzed with me. So my friend with whom I’ve been playing Apocrypha offered to play Sankore instead (i.e. I talked her into it). She’s as entranced as I am by boards like this, and I wanted to flex my voice to see if I could manage a teach. I lost my voice completely for a few weeks recently and to my immense relief, it’s come back. So we got through a teach, which was a lot easier than I imagined! Sankore might look complicated, but it’s not that hard to teach thanks to the excellent iconography and “interface”.

So we got a game underway, each of us kind of feeling our way blindly into this exotic sunny sandbox, bumping into each other, stumbling over one another’s caravans, interrupting each other’s legal scholarship, sometimes stealing books out from under one another, but also doing plenty of “multiplayer solitaire”, and generally giving the game shape, gradually putting the clock and scoring into focus. We didn’t finish the game because near the end it was obvious that one of us had managed a blow-out in terms of scoring. But it was a great way to see how the machine worked, to push buttons and pull levers and watch our madrasas process students and effect changes on the board.

And I think now I might – might! – be ready to give this a try solitaire.

By the way, a third thing I really like about Sankore is the subject matter. Running a 14th century North African madrasa, furthering human knowledge and advancing civilization while a diminished Europe waits out the Dark Ages? Sign me up! Garphill’s South Tigris trilogy about Baghdad is about similar subject matter, and this reminds me a lot of Wayfarers of the South Tigris and especially Scholars of the South Tigris. Sankore is considerably heavier than Wayfarers, and it actually overlaps a lot with Scholars, at least thematically, with its emphasis on collecting manuscripts. Sankore also feels a lot more serious to me, mainly because I really dislike Miho Dimitrievski’s webcomic artwork in the Garphill games. Sankore, on the other hand, seems to take seriously the Islamic prohibition against human likenesses with the single exception of a drawing of Mansa Musa:

Mansa himself
Mansa himself485×647 162 KB

I was curious how such a drawing even exists, so I looked it up. That image is from a Jewish cartographer’s map!

Another nifty touch that isn’t documented is the student meeples have their turbans untied and hanging down to their shoulders. Here are a couple of students loitering around class:

students hanging around classes

What look like neckties hanging down from their shoulders are actually untied turbans. Which explains why this is the icon for when you graduate a student:

graduation

Not sure if you can see, but it’s a multicolored background – any type of student – with a turban in front:

graduation zoom

Because when students graduate, they get to finally tie up their turbans! I love little touches like these in a boardgame.

Anyway, now to try solitaire and see how this weird new three-act structure works. I’ll report back later!

2

I love everything about this game and your write up. What a gorgeous game board! And I love the premise.

3

I got a three-player game of Sankore underway (although I still haven’t sampled the solitaire yet, but for reasons that will shortly become clear, I really need to!). My two opponents are both very good at wrapping their heads around Euros like this, whereas I’d say I’m a child of Ameritrash who can struggle with abstraction.

Also, I’m realizing that one of my issues with multiplayer games is that I feel compelled to play host, so I’m constantly puttering around during other players’ moves. Also, since I eat differently from other people, I try to set up my dumb feeding robot before fixing lunch for folks, and that takes a bit of busy work, which actually isn’t that much of a bother, but…well, I appear to have trouble tracking a heavy Euro unless it’s the only thing I’m doing.

So as our three-player game unfurled, I think I kind of broke my ability to process what was happening? By constantly disengaging and then trying to re-engage, I pretty handily lost track of not only what I was trying to do, but also how certain things worked. I was getting frustrated at how often I was being corrected by two people who I had just taught the rules and by the time it was over and I was clearly trailing, I got all grumpy and even kind of shushed my friend as he was trying to suggest a possible move while I was obviously paralyzed. But at least I was aware I was being a putz, so I apologized after I’d locked in my suboptimal move. :)

So even though I know the rules well enough to teach them, I’m not sure I know the rules well enough to play unless I lock my attention exclusively onto the game. At least that’s my theory, and I hope to mess with this solitaire some more before trying to share it with folks.

Also, the three of us are a bit concerned that there might be some (inadvertently?) hard-coded “meta” in Sankore. The game features four “disciplines”. Three of the disciplines (astronomy, math, and theology) are part of an economic circle of resources. A fourth discipline (law) instead boosts your actions. So it seems to us that you always want to pursue law first so that your successive actions will more productive. Seems like Game Strategy 101. I’ll be sorely disappointed if it’s that simple, but that’s the state of our current and very inchoate “meta” for Sankore: a game where astronomy, math, and theology have to take a back seat to law during the early game, so you better hope you score some law students.

Anyway, Sankore is boxed up for now until I dive into solitaire, which is the whole reason I bought it in the first place! But if it’s taught me anything, it’s that I might have a personal ceiling for Euro complexity.

4

I can’t help but wonder if that’s commentary on contemporary society or 14th century Mali.

5

I finally got through a solitaire game of Sankore!

It took me a while, because I’ll readily restart the entire game when I realize I’ve gotten a rule wrong. I restarted several of my solitaire Sankores because I kept forgetting to score regions at certain thresholds, so a few turns later, I’ll realize, “Oops, I should have tallied Theology two turns ago” or “Oops, all the books in Astronomy were supposed to already be assigned!” Minor stuff, but enough to mess up the flow, to alter the outcome, so I just reset the pieces and start again from the beginning. I never mind. The results of any game are far less important than the process. Eventually, usually around the second or third game, depending on complexity, I’ll get all the way through. I believe this was my fourth attempt at Sankore. She’s nothing if not dense and occasionally inscrutable!

I was playing the lowest of the four difficulty levels and I’m happy to say I lost by four points! In my earlier games, I was pretty sure I was going to steamroller the basic AI, and I probably would have if I hadn’t restarted. However, for this game, I opted to ignore the early warning signs – more on this in a moment – to see how I would do if I just plowed ahead with my initial strategy. But shortly after the halfway point, I was in danger of losing egregiously. Ah, so that’s how I would do. I frantically tried to change course, and I almost pulled it off. I lost 88 to 84.

Four Points
Four Points1200×900 199 KB

So a couple of things I’ve realized about Sankore. Firstly, I don’t think it’s a very good game for just two human players. Sankore is about fighting for control of limited resources, while simultaneously establishing the value of those resources for scoring. Not only do you start with nothing, but everything on the board is worth nothing! As players earn “prestige” in the game’s four disciplines – Math, Astronomy, Law, and Theology – some of their actions will put books representing these disciplines into the library. The shelves begin empty…

Bare Shelves

…but by the end of the game, it fills up like so.

Full Shelves

The majority discipline on each row is worth two points per prestige, and the second place discipline on each row is worth one point per prestige. A Sankore player will recognize in the top row a pitched battle between Math (teal) and Theology (purple). In the end, Math won the top row because ties are resolved by first book to hit the shelf, and that means each Math prestige will be worth two points. But since Theology placed second, each Theology prestige will be worth one point.

But then check out the second row, where I managed to get Theology into first place for another two points per prestige! That means each Theology prestige is worth three points, whereas each Math prestige is only worth two. Still, the AI had amassed an absurd amount of Math prestige and I couldn’t overcome its lead, despite my frantic attempts to keep up, and eventually to switch over to Theology. I almost succeeded! Curses, foiled by those pesky Math students!

But this is why I think Sankore isn’t quite as rich with only two players: they’re jockeying for position in a zero sum game between only themselves. They can either play against each other or past each other. There will be none of the chaos and opportunity of a third player similarly playing against or past either of the other two, each of who might also be playing against or past either of the other two. With three players, no one will enjoy the luxury of playing past both opponents. And with three players contributing to the library and therefore the endgame scoring, Sankore will be an exercise in triangulation instead of mere sums.

And here’s where I realize that playing Sankore against an AI is just a way to watch the design wiggle and juke and writhe with something resembling intent. Which – heavy sigh – is true of most solitaire modes for multiplayer games. Sankore is, of course, not a solitaire design by any stretch of the imagination! But neither is it just another inane bot flipping through a deck of autamoton cards to arbitrarily Make Things Happen.

Instead, the AI in Sankore is a relatively simple three-act progression that requires only the instructions on the reverse side of one of the player boards (the solitaire mode doesn’t even use any additional components). For the first four turns, the AI will draft a student, building up its placement and scoring preferences. Then, for turns five through nine, it will then build up its madrasa the same way a human player would, drafting classes to build a “tree” of available actions. Finally, for turns ten through sixteen, it will cash in on its student/class combos. And all the while, on every turn, it will place a piece on the board to vye for one of the disciplines, occasionally seeding the library with books to score its favored discipline. After sixteen turns, go to scoring.

The bot doesn’t have to adhere to the rules, of course. It gets the usual break for not having to do any resource management, but it’s only doing things that would be mostly plausible for a human player. It reacts to the board state, more than willing to play against or past you at your discretion. But of course, it can easily be manipulated into doing counterproductive things, into chasing your lead in areas where it’s never going to matter, into wasting its actions, into blindly following its instructions instead of trying to win. The AI rolls along the deeply furrowed path laid by its script with only minimal help, stepping easily into the gaps between your turns, never demanding much time or even attention, eager to get you back to your own situation.

This is the point of the solitaire game: watching and reacting, remaining agile, learning to read the inscrutable boards, and appreciating the nuances of each of the four disciplines. Practicing, I guess?

Of course, this is only after a game against the most basic of the four AIs. Which I lost. I’m going to give it another shot, presumably because I figure I should be able to win every time against the basic AI once I appreciate the rules. For instance, getting out in front of a math vs. theology showdown on the winning side. The harder AIs seems to just get more powerful actions earlier, with greater influence on the library composition. I don’t know how much this will change the basic style of a solitaire game beyond racking up higher scores, but I’ll find out eventually.

Finally, a note about the overall resource balance. It might sound weird, but the basic resource cycle is that Astronomy lets you spend gold to sends caravans out into the desert to find salt. Theology lets you spend salt to visit distant madrasas and bring back books. Math lets you spend books to build up the madrasa at Sankore for gold. Which then sends caravans out for salt, which is spent for books, which earns for gold, and so on.

But then there’s Law, which exists independent of the resource cycle. Instead, Law powers up your other actions. At least when you do it early in the game. When you pursue law late in the game, it gobbles up the overflow from the other disciplines, the leftovers that no one won, the unclaimed spoils. So Law seems to be the discipline that early on gives you a minor boost, and later on lets you clean up the leftover scraps. Whereas my friends in the multiplayer game were concerned that law was overpowered, I’m not really seeing it now. It’s certainly got its benefit, but as with each of the disciplines, I think it all comes down to how the library shakes out.

In fact, that could actually be the tagline for Sankore: it’s all in the library.

6

Welcome back to my Sankore blog! We began with this at the top of the thread:

One reason being how the design and layout are so adoringly realized with Ian O’Toole’s beautiful art. The second reason being the solitaire mode, which scripts a three-act structure that simulates what a human player would do. I thought this sounded like a different approach to solitaire play than the usual deck of “automa” cards.

Which it might be, but I’ve come to the conclusion it’s not a better way. All it does is more-or-less blindly jostle the game state. It’s the equivalent of someone randomly changing the numbers while you try to do a math problem. It’s not AI so much as AA: artificial activity.

Having just finished another solitaire game that I handily won 115-59, I think I’m officially done with Sankore as a solitaire game. Once again, I’ve discovered how poorly certain designs fit to solitaire play, how some games simply can’t be managed with a bot or a script or a deck of cards. How multiplayer designs are so often, well, multiplayer designs.

The main issue for me is that Sankore is about intentionally building up a scoring system based on what actions you take and how you think the board is going to look at the end of the game. It’s very much about looking ahead, anticipating, setting up, carefully ensuring that your actions help you and hurt your opponent. Yet the AI scripting has no compunction about blindly performing actions that help me considerably more than it helps itself. In fact, this seems to be the cornerstone of the solitaire game: pushing the AI into making moves contrary to its own interest. Do that for 16 turns and you win.

Furthermore, many of the moves the AI makes violate the rules and impact your ability to play the game as intended (i.e. the way it works with human players). For instance, several actions require players to commit books to the library, which establishes how points will be scored. As the game progresses and the library fills, a “score economy” starts to take shape: which disciplines are worth the most victory points. Yet before books enter the library, you can see the possible directions this might go based on which books are in circulation! You can somewhat anticipate the “score economy”. Books have to be earned and then shelved, a multi-step process that you can anticipate and react to as it happens. But since the AI doesn’t care about resources, it just plucks books from the supply and instantly drops them into the library. The whole scheme of watching the economy take shape and acting accordingly goes out the window when you play Sankore solitaire. Scripting trumps design, yet again.

I’m still very glad to have this in my collection, although I don’t anticipate it’ll get played much more, if ever. It’s a very very heavy Euro design that will take an extensive teach and at least three players. And even then, this is the kind of game you probably have to play to learn it, and you’ll probably play wrong the first time. On the upside, I do feel that my and my friends’ concerns about the Law discipline being overpowered were probably unwarranted, and I’m eager to play with humans to find out whether that’s the case. And I’m also pretty sure it doesn’t commit the sin of many heavy Euros of wearing out its welcome; this seems to be one of those games that – hey, what happened??? – is over too soon. So, perhaps I’ll convince a couple of friends in the future.

Otherwise, maybe in a few years I’ll forget that the solitaire mode was bad and I’ll unwittingly break it out again.

7

My feeling is that the best you’re likely to get playing most games solitaire that were designed as competitive multiplayer titles is a bot that gets in the way of your plans sometimes and provides some randomness and pushback while you explore the game’s systems on your own. It’s difficult to really create the sense of human opponents playing the same game, especially when for expediency’s sake you tend to need to abstract away hugely important parts of the game like resources, and far as I can tell pretty rare for designers to care enough to even make a serious stab at doing so, still less succeed. There are times when I’m willing to deal with these failings simply to get to take a game for a spin when I don’t have a realistic prospect of playing it anytime soon with friends…but I try not to buy those games in the assumption that I will enjoy them primarily solo. Especially not when there are so many really good games actually designed for solo play.

8

Can you PM me this post every time I’m going to buy a new multiplayer boardgame based on my hopes for the solitaire mode? :)

One of these days, I’ll assemble a list of multiplayer designs that I think also work as solo games. It should be a pretty short list.

9

Not including co-ops? Are there any? Solitaire modes, in my experience, are mostly good for learning the multiplayer game.

10

Gah, you snipped off the “one of these days” from my comment! Now I have to do it! :)

I have a weird approach to coop, as you may know. To me, cooperative games are almost always an iteration of solitaire, which is why I specified multiplayer designs. I should probably specify “multiplayer competitive” to be more clear, but that just sounds so wonky to me.

But, yes, I believe there are “multiplayer competitive” designs that work well solo! John Company, for instance, and I had a massive thread where I wrestled with that one before concluding that it’s very much its own game when played solitaire. A few others that I can think of off the top of my head that I probably haven’t talked about here:

Nigel Buckle’s Imperium games have amazing David Turczi bots that, IMO, make them better solitaire games than multiplayer. The multiplayer is too long and drawn-out, with too much potential for players to play past each other. But I love these games as solitaire matches against Turczi’s bots!

Games like Gaia Project that lean into an asymmetrical faction’s asymmetry with their automa decks provide interesting faction-based challenges (part of the appeal of Turczi’s Imperium bots, by the way!). In fact, this is probably why I impulse-bought Sankore; Gaia Project is a very heavy Euro with a hearty solitaire mode, so why can’t other heavy Euros work just as well?

Many Euros are “multiplayer solitaire” designs – ugh, now I’m really sounding wonky – that work nicely as score chases (i.e. you’re basically playing “multiplayer competitive” against your own past games). Pretty much any Uwe Rosenberg design fits the bill here, some better than others (Feast for Odin and Fields of Arle are my two personal favorites for Euro/Rosenberg solitaire).

I’m not sure I can articulate why yet – stand by for the next game I plan to write up, which is thematically related to Sankore! – but Garphill’s games are fascinating to me for many reasons, one of which is how well I feel they work solitaire. Unsurprisingly, they finally released a dedicated solitaire game (Legacy of Yu) and to their immense credit, it plays very similar to their “multiplayer competitive” designs. That might not sound like high praise, but that is my intent with that comment: Garphill really “gets” solitaire play, and it shows even in their multiplayer designs.

Otherwise, there aren’t many multipalyer designs I can think of that work well solitaire. My shelves are littered with examples of multiplayer designs that aren’t good solitaire games! It’s my feeling that multiplayer designs are multiplayer designs, and attempts to appeal to solitaire players are almost always crass marketing and/or half-assed design work. But there are diamonds in the rough out there, and I intend to keep looking!

11

Yeah, I’ve pretty much come to thinking of those games as solitaire experiences. Not unlike how I feel like the Super-Skill Pinball roll-and-writes, while obviously very different games, have no real impetus to multiplayer. You’re just rolling dice and making choices about how to use them to score the most points…I guess you could do that together but I don’t really see why. Great fun solo though.

12

My thoughts:

I think the solitaire mode here is a work of staggering genius by Ricky Royal. I wouldn’t have thought you could do a solo mode in this game. It’s truly an impressive feat! But it is also cumbersome to play. You have to refer to a book to find out what the Crown is going to do. I’ve done a couple of playthroughs with it, but it’s too much work to compel me to do it more.

I have played these and they are impressive. Again, though, they’re real fiddly. And due to the enormous numbers of combinations of empires, they don’t always work well. Too much work for me.

I’ve never played Gaia Project and didn’t know it had a solo mode. I looooved Terra Mystica back in the day and played dozens of games on @jsnell’s site so I really need to check this out.

I love many of these games (including most Rosenberg designs) but I have a hard time enjoying the score chasers as anything but ways to learn the game. That’s probably just me though.

I actually have enjoyed a few Eklund solo modes quite a bit, e.g. Bios: Megafauna and Greenland/Neanderthal. Bios: Genesis is, I think, best solo; the multiplayer is a mess.

I really like what you wrote about how Sankore has a more holistic approach for the solo mode. I’d really like to find a game that isn’t a score chaser and has a solo mode that is both simple to implement and creates surprising challenges for the player.

13

Oh, I wasn’t necessarily recommending those games! I was just rattling off a few multiplayer designs that – in my opinion – also work as solitaire designs. I didn’t mean for you to take it as a list of recommendations, otherwise I would have given it more thought and probably included caveats.

Well, there are tons out there, but we were talking specifically about multiplayer designs that also work as solitaire games. But if you remove the multiplayer requirement, you’re going to have a much easier time finding a solitaire/coop game that fits the bill. In fact, that’s going to be a pretty long list!