Ayar designers Fabio Lopiano & Mandela Fernández-Grandon deep-dive into the game's creation process...

Ayar: Children of The Sun is a follow up to Merv: The Heart of the Silk Road and Sankoré: The Pride of Mansa Musa. Like its predecessors, Ayar is a highly interactive Eurogame with design driven by its historical (in this case blended with mythological) setting, an interesting scoring system, and camelids. The game is all about timing, evaluating shared incentives, and exploiting the puzzle of your own player board to seize those opportunities.

And camelids.

Mandela: In the beginning, the creator, Viracocha, created the Sun, Inti, and the Moon, Mama Quilla. They arose from Lake Titicaca to illuminate the Earth, but the Moon shone so brightly that the Sun was jealous. He took a handful of ash and threw it across her face. Remember that, it will be important later. The Sun and Moon would bless the world with four sons and four daughters: the Ayar.

That is where the story starts...but we're going to skip ahead ~5527 years (to the year 2019 in the Gregorian calendar) and a fraction farther from the center of the universe over to a café in London where I was playtesting a game of Fabio Lopiano's that would go on to become Zapotec.

I liked the grid resource system in Fabio's prototype: select a row and get all the bonuses you'd added there. I suggested that if it worked across columns, too, it could be a wee bit more interesting. The idea of a grid activating on a row or column was on my mind as I'd already been playing around with a similar system but for action selection.

An early prototype of a game that eventually became Zapotec

After the test, I tried to explain my mechanism to Fabio, but realized it would be better to go ahead and make a prototype. After all, a prototype is worth a thousand ideas.

For the following playtest session, I had something ready. I hastily made icons to match components I had available, and since I needed a map, I copied the outline from 8 Minute Empire which I had to hand.

Despite appearances, the theme for the game was set in my mind. My family on one side are Aymara and something in Aymara mythology led me down the path of reading about how the Inca adopted stories from the existing Andean communities, which led me to the story of the Ayar. In this first version, you played as one of the sibling pairs of Ayar. The core idea was that you activated a row or column on your board and carried out the actions shown there in order, ending with you picking up a new action to add to your grid.

First test of the game

The game didn't actually have a way of ending at this point, but I glossed over that detail and got people playing. There was little incentive to use your full action-selection grid. After the play, Fabio and I debated this – him saying it should be just a single line rather than a grid, and me saying I'd change the rest of the game to create value in using the other spaces. It's not generally advised to argue with playtesters...but sometimes it can be fun and, as in this case, lead to fruitful discussion.

I continued to work on it for a couple more iterations, toying with some interesting scoring mechanisms and adding plenty of convoluted systems that were fun to play with but lacked focus.

Early iteration with fun ideas but too many jumbled together to be a good game

Then Fabio left the UK. In parting, I commented that it was a shame and that I felt this was a game we could have worked on together. There was a murmur of agreement, but in the knowledge it was not to be; distance was, of course, an insurmountable barrier.

I convinced Fabio to test the design again at SPIEL a few months later, where he provided useful feedback. However, at that point I shelved the design. Soon after that, I took a longer break from game design altogether before returning to work on Sankoré with Fabio.

After we'd finished Sankoré, I suggested we could take a look together at Ayar...but he wanted to do something lighter as a bit of a palate cleanser. With a great deal of effort, we designed a family-weight game, which is a really nice game but still to be signed, so maybe we'll write about that another time. Anyway, with that out the way, Fabio was willing to take a look at Ayar.

Fabio: I don't remember much of the first time I playtested Mandela's game in London, but I remember the second time we played it during SPIEL 19. I had just moved to Italy and met back with Mandela and other friends at the show. We then sat in the Atlantic (the big hotel just in front of the Messe) to try new designs.

Mandela set up his game, at the time called "Tiwanakuta", with an interesting action-selection mechanism on the player board and some "dudes on a map" going on the main board.

Fabio playtesting "Tiwanakuta" at SPIEL 19; the other tester, Mauro Lazzara (not pictured), would go on to become one of our main testers for Sankoré and Ayar

We didn't talk much about that game after the show until about three years later, when we threw around ideas about our next collaboration on a new game design.

Although it might not seem like it, I am very much a "theme first" kind of designer. The first things I need to establish while designing a new game are the answers to the questions "Who am I?", "What am I doing?", and "Why?". These are also the questions I ask myself when playing other games, and I tend to like games where the answers to those questions can guide me to a winning strategy.

According to the myth, the Ayar traveled across the land looking for the center of the universe, Cusco, where they could found an empire. An important part of the journey is that only one Ayar pair remained at the end, Manco Capac and Mama Ocllo, who would become the first Inca and Coya respectively. The others were trapped in caves or turned to stone — temples, as we see them today.

I couldn't think of how a player could identify with one Ayar without some form of player elimination. Also, with players as the Ayar, the game wouldn't have worked well at any player count other than four, so we tried to change the perspective. The four Ayar were followed by a number of clans, or Ayllu, so maybe that's what the players should be. Each player is leading a clan and following the four Ayar, and the Ayar, one by one, retire until only one reaches Cusco at the end of the game. That scaled much better and allowed us to follow the story more closely.

Mandela: When I had started, I was pleased with the coincidence of the four Ayar pairs matching the potential four players of the game. The major shift that Fabio pushed for early was to make these neutral pieces. I had to recognize the elimination of the Ayar on their journey was such a central theme in the mythology that unless we wanted player elimination, they shouldn't be the players. When I relented, we reached the first milestone that allowed the game to take shape.

Fabio: While the four brothers were traveling through the Andes, their four sisters were teaching the basics of civilization to the Ayllu, activities like terrace farming, weaving, pottery, and reed bundling (which consisted of building artificial islands on lake Titicaca). We tried to keep close to the story and have the Ayllu doing these activities as they follow the four pairs of siblings.

Soon the action selection in the game involved picking one of the Ayar that were still active, advancing them on the board, then doing one of the four activities, depending on where that Ayar landed. Every round one of the Ayar would retire, triggering scoring opportunities.

Initially the path was a rondel, in this case made of cards, each with two action types and slots for buildings (which would provide benefits to the owner)

The grid on the player board determines which resources were available that turn, with the possibility of placing better tiles to upgrade it

Mandela: We played with various ways to track the strength of the Ayar during their progress along the journey, mainly through tracks up the side that would be adjusted after each turn. In one version, your overflow of resources were sacrificed to one Ayar, increasing their chances of survival.

Resource holding slots were occupied by tambos — the wee houses on your player board that you placed on the map. This made sense with the setting as tambos were often used as storehouses, so the more you built, the more goods you could hold. It also created an interesting dynamic of needing resources to do things, but also wanting an excess of resources beyond your storage at other times, which is easier with fewer tambos built.

This was all very interesting, but also very fiddly. It meant adjusting tracks, as well as taking resources only to immediately return some to the supply when really we wanted players lost in the decisions and anxiety of whether their strategy would reach fruition, not distracted by admin.

The solution came in one of those rare and pleasing moments of unison when we both realized that we could integrate the tracking of the Ayar onto their physical movement across the map and make the systems work around that. Why abstract it out and have players imagine the Ayar's progress across the land when we could have it all laid out in front of them? While co-designers often say they can't remember which idea came from which designer, although we both bounce ideas around and refine those from one another, I tend to have a good recollection of the origin of each idea — and that's why these scarce moments of collective realization can feel like a true epiphany.

First version of the board with distinct paths for the four Ayar

Fabio: The rondel was replaced by four distinct paths along the board. At first, the actions were on the player board. Each player had a token for each Ayar color, placing one on a row or a column to move that Ayar by a certain number of steps and activate all the actions on the row or column it pointed at.

At the end of each round after all players had used all of their discs, the Ayar who was left further behind on their path would retire. Players lost the matching disc and would have one fewer turn next round, but the strength of the actions kept increasing as the game proceeded so that you could still have a powerful last round with just one Ayar still alive.

The next step in the design was to move the actions onto the main board. Each Ayar now has a sequence of action slots along their path. Each player starts with their 4x4 grid covered by tambos. On their turn, they pick one of their discs, place it along one row or column, move the matching Ayar by the number of steps indicated on the grid, then take one tambo off the grid (from the same row or column pointed at by the disc) and cover one of the action slots on the main path of the Ayar, taking that action with a "strength" equal to the number of empty slots on that row or column. (When you take off the first house, you get a strength of 1, and when you take off the last house, you get a strength of 4.)

At that point we also added a "wild" disc that lets you move any Ayar, so that in the first round each player has five actions and in the last round two actions (for a total of fourteen).

This early version of the prototype, from October 2021, was getting close to where the game would eventually end up

Since the grid holds sixteen tambos, we decided to place one during set-up — each player picks one of the four starting slots and takes its action with strength 1 — so that when the game ends, only one tambo remains on the board. Because we had bonuses for completing rows and columns, choosing which tambo to leave on the board at the end of the game also has implications for which bonuses to skip.

Moreover, the wild disc gave more control to players about which Ayar retires at the end of each round. We wanted the retirement of the Ayar to be meaningful, i.e., players should care and have reasons to root for one particular Ayar to survive — or to retire.

An Ayar pair reunited in the night phase, ending their journey. This may mean a plan perfectly executed for some players and a change of strategy for others.

Mandela: So we designed the scoring system around that. We have two scores: Sun points and Moon points. Sun points are scored during the day, while the Ayar move. Moon points are scored during the night, when one of the Ayar retires. Remember we said the Sun, Inti, was jealous of the Moon, Mama Quilla? Well, to represent the need to balance devotion between the two and avoid that jealousy, only the lower of your two scores will count in the end.

We also wanted these to feel different, so Sun points are gained in small chunks across turns, while Moon points are scored in bulk at the end of each round. This created a nice delineation of scoring that happens directly through your actions during the main day phase (Sun points) and those that are inevitable — though their timing may vary — in the night phase (Moon points).

This provides a bit more texture to each type of scoring as more tactical play is required to optimize Sun points and more strategic considerations are needed to gain Moon points. It sometimes makes sense to sacrifice the chance to gain a glut of points of one type for a measly few in the area you are lacking — or do you play with optimism (cf. hubris), hoping that you can catch up on the other type later? This brings to life with clearer delineation than usual something we've striven for in our all our designs: that a player must balance strategic play with tactical moves to achieve victory.

Initially we had Sun scoring tied to a temple type; after all, there are both temples to the Moon and temples to the Sun. However, your potential to score points escalates as the game goes on, so it was felt there should be a trade off in scoring later to keep the tension throughout the game. Initially our solution was early spots offering multipliers that diminished as play went on. How late can you leave it to take those high-value spots before someone else does? But at higher player counts, it could be too frustrating to see several get taken before your turn and being multipliers the swing in points was too dramatic.

We tried other ways in which players gave up their turn doing an action to acquire them, ways that were smoothly integrated into the existing systems, but players were too absorbed in completing their plans and having fun to want to give up a turn just to score silly game winning points. This is where having a round structure, something we've avoided in all our other designs, had its advantage. It allowed us to hide some of that math and further build a sense of progression. We moved Sun points acquired onto a devotion track that would never go down. At the dawn of each round, you add this devotion to your Sun score, without ever reducing your devotion level. This created the kind of satisfying multiplier trade-off that we sought, i.e., a Sun point earned in round 1 will score four times. It kept the tension of early vs late scoring alive and leads to a rich sense of progression in the game as you see it pay off in an ever-climbing score.

Player board midgame: One of the most notable component differences from our prototype is that devotion existed as a track around the player board but is now represented on tactile and beautifully illustrated dials. Small though it may seem, earning those numbers on the dial is particularly pleasing, especially ticking into double digits

Knowing players wouldn't break from their enjoyment, even if it meant winning, we needed to integrate the acquisition of the Sun points into an area on which players were already focused. Ideally this would be informed by the setting and work within existing systems rather than introduce more complexity.

It made sense to tie the scoring with the progress of the Ayar, which we did by introducing scoring tiles at various points along their paths. These represent Inti Raymi, an important festival of the Sun which is still celebrated to this day. As soon as an Ayar reaches one of these tiles, all players score Sun points for their progress in the activity shown on the tile. This adds tension to the game as some players might want to slow down an Ayar to prepare for the scoring — or speed them up if they are ahead in that activity. Others might even want the Ayar to retire before reaching a certain Inti Raymi tile so that no player benefits from it.

Historically, during festivals participants may carry aloft a llama statuette or present it in appreciation; it can be made of gold, which is the sweat of Inti, or silver which are the tears of Mama Quilla

And this is where our camelids come in. During one of these Inti Raymi scoring events, if you participate by having the prerequisite devotion to the activity (that is, revealed suns in that action), you will also gain a llama statuette. These can be used at the end of the game as either gold or silver. While they may not be worth a huge number of points alone, the flexibility they offer can be invaluable. We wanted them to be as important to the game as the real llamas are to Andean life.

Mandela: My cousin (left) pictured with my dad's pet alpaca, Moroca (right)

They may also determine how wide or narrow you want to go: ignore activities and you will lose out on llamas, but spread too thin and you may never score big.

At the same time, we also introduced temples that players may build at certain times, with each temple dedicated to one of the four activities and to one of the Ayar pairs that are still alive. When an Ayar retires, all temples dedicated to them will score Moon points for the players who built them. Therefore, by building a temple you are investing in an Ayar, trying to keep them alive for as long as you need to maximize your score.

In the myth, a couple of Ayar turned into temples, as did many of the stone people that existed before. (The Ayar were Viracocha’s, that is, the creator's, second attempt to begin life.) This conveniently explains why there were already ruins at the time of the Inca. Matching the myth, at the end of a round you will retire the avatar of the Ayar from your board, turning them into a temple. You choose which activity they will score for, giving you a reliable source of Moon points, but again the timing and order remains critical if you want to make sure to get the most from them. At this point, each player has 15 actions, each one involving several choices at once: which Ayar to move, by how many steps, which tambo to take off the board, and where to put it on the path of the current Ayar, thus determining their activity for this turn.

Fabio: Each of these choices has several implications, not just on which Ayar advances and what action to take, but also because clearing rows or columns from the grid could trigger additional bonuses and scorings, and clearing a quarter of the grid lets you place a temple.

We had to keep the actual activities basic and straightforward to not burden the player with additional cognitive load, so we decided to eliminate all the resources in the game. You are not managing resources; instead you are managing your priorities and your tempo, constantly deciding what to do and when. The pottery activity became a simple set collection, terrace farming a simple area majority, textile weaving simply meant covering an area with tiles with basic matching, and the reed bundling on the lake became a race to the farther islands.

Reed bundling is less core to the Inca idea of a good way of life, but we wanted to include interaction with Lake Titicaca, which is prominent in Andean mythology and our game board; it was also a nice way to honor the Uros, the clan who moved out onto the lake, bundling reed to make their boats and the floating islands on which they still live today

As you progress with an activity, you unlock a number of Suns and Moons, but these will be counted only when the activity scores. For example, each island you build is worth one additional Sun, and when a festival is triggered, each player will gain one Sun point for each island they have built.

On the other hand, each island is also worth an increasing number of Moon points; the closest island to the starting place has 4 Moons, whereas the one farthest away is worth 32 Moon points. When you score moons, you take them only from the highest value island you made, so you want to have reached a few islands when the first Sun scoring is triggered, but if you dedicated a temple to a certain Ayar, you want to reach the other end of the lake before that Ayar retires.

At the same time, we wanted players to engage with all of the activities and not just pick one or two and ignore the others, so the game is all about timing. Players should try to focus on the next activities to score, then quickly move to a new activity once the scoring tile has been reached.

We achieved this in a couple of ways, one in the placement of the scoring tiles along the paths. Each path crosses all the other paths, and we placed a scoring tile at every crossing so that each tile can potentially be triggered twice, once for each Ayar that crosses it. There are two tiles per activity, so an activity could score up to four times if both Ayar survive for long enough to reach the second scoring tile for that activity.

The teal Ayar is three steps away from triggering the Raymi scoring tile rewarding weavings. Will he make it?

At the same time, we added an additional scoring trigger for each activity on the tambo grid so that when a row of tambos is cleared, the player can score that activity. (This gives them more control so that if they go heavy in a certain activity, they can still score it even if the Ayar retire before reaching the scoring tile.)

Moreover, we also added a Moon scoring at the end of each round in which each player will cover with their retiring disc one of the four activity scoring slots on their player board, immediately scoring moons for that activity. This adds some pressure as each round you should focus on a different activity, score it with your retiring disc, then move on to another activity for the next round.

Mandela: With all of the core elements in place, we had a complete and satisfying game — but one that could handle just a little more spice. From all the time messing around with a system for the Law action in Sankoré, we'd designed a few systems we really liked but which ultimately didn't fit that game. Ideas, like our Inca, don't truly die. One had still been on my mind, and we realized it could fit perfectly here to give it that extra flavor. This came in the form of Viracocha tokens — blessings from the creator. They can be gained from each of the activities, though with varying degrees of reliability and abundance, and applied to unlock upgrades on your player board, letting you customize your bonuses each play.

Everything was in place, and players seemed engaged in the game, very engaged. In fact, "brain melting" was a phrase that came up often. A comment that stuck with me was from Daniele Tascini, who tested it and said, "It is impossible to play this game well and still be sane at the end." He suggested that we could split it into two or three games.

We played around with simplifying the design. Could we keep some bits back? We even replaced the action-selection system entirely with something new and ran tests. It all worked well, but ultimately we decided that we should make the game we were excited about and reverted to our grid. We were happy for it to be challenging but recognized the puzzly elements could be toned down a little to let the other elements shine through.

Player board with a lot...maybe too much to think about

Fabio: Yes, we kept the original action-selection mechanism, but we stripped down considerably the side effects of choosing which tambo to remove. In the initial grid, we had a Sun scoring trigger for a specific discipline when completing each row and column and a specific temple on each quadrant, so as you were choosing which house to place on the board, you were trying to solve several puzzles at once: not only how many steps to move this turn (and possibly next turn since the crossing row or column will have one extra power) but also which Sun scoring to trigger once the row is empty (while making sure you have enough Suns when it triggers) and which temple to work for.

We decided to simplify a lot of the decision space. We replaced half of the scoring triggers on the grid with bonus llamas or Viracocha tokens, and we left to the player the choice of which temple to place when clearing a quadrant. This reduced considerably the amount of brains melted during playtests.

The design of the original grid (left) was "evil" as the temple category never matched the row and column triggers; the final grid (right) has less entangled outcomes

Mandela: It might seem strange that we didn't design Ayar with Osprey Games in mind, but when Anthony Howgego, at the time a developer at Osprey, suggested they would be interested we knew that it made sense. The game continues the confluence of design ideas that Fabio had begun to explore with Merv and that we'd continued with Sankoré. We knew that Osprey would handle the game and our vision with respect, and find the best way to make it shine.

In addition creating the resplendent and setting-appropriate art, Ian O'Toole was up to the challenge of making the paths of the Ayar as clear as possible, creating lavish artistic detail but letting it sink away during play so that critical information is clearly read. We know this was not an easy task, and it took a lot of back and forth to get right. Osprey brought on board a cultural consultant, Antti Korpisaari, who knows a great deal about Andean pottery, to allow us to add those extra levels of details that will mean nothing to most but hopefully a lot to some.

I'm very proud of the game. My first game, Glasgow, was about the city I come from. This design felt like I was exploring that fertile ground further, driving the staff a little deeper and taking the setting of where my family came from both in geography and in the kind of stories that are told there.

Although the design feels quite personal, I certainly don't want to be the one to throw ash in anyone's face. It was Fabio who took it on the journey from a surfeit of loosely linked ideas, letting darlings fall to the wayside and new ones emerge, until only the leanest game — and one true to the story — remained.

Although this is my seventh game out in as many years designing games, it will be my first to have a proper release at the center of the board gaming universe: SPIEL Essen. Whether there or wherever you may get a chance, I hope you will raise your llama and join us in celebrating what the whole team has created by reliving the tale of the Ayar.

Jallalla!

From left to right: Ian O'Toole (illustration and graphic design), Jordan Wheeler (development), Fabio (design), and Mandela (design)

***********************************************************

Ayar: Children of the Sun is out now. Order today.

Read the rulebook on BoardGameGeek.

Explore the other games in our eurogame trilogy from Fabio & Mandela here.

• Board Game