Designer Diary: Shadows: Amsterdam

This is the story of a nice story, one of those that should never end, such as a Ken Follett book or an HBO episode.

I work at Libellud with the best job in the world, awesome colleagues, nice projects, and a workplace located one hundred meters from my flat in the center of a welcoming city. My game has now been published by Libellud, a great French game publisher, as far as I know, even if I had never stepped around to the other side where designers and illustrators are interacting with us.

Episode I: Frustration

The Incident happened on the 24th of May in 2017. Wikipedia notes that May 24 is the birthday of Daniel Fahrenheit, Charlie Taylor, Bob Dylan, and Kristin Scott Thomas and that... Funny how the time can be wasted on Wikipedia!

This 24th of May was quite an event for me because it was the first time a publisher played one of my games — and this playing was involuntary.

I have made a lot of game prototypes, starting when I was ten with my bro Clément and my friend Camille, but the only purpose was to entertain our friends. Each time, it began with "Man, it would be awesome if we could do this", then we proposed some ideas. These ideas became games that existed in no more than a single sample, which gave us some bad/good times for an afternoon or evening. As we were far from the gaming world, we never thought one of our creations could be published. We also never used the word "prototype".

It is weird, but I did not have more expectations with this "Spies" prototype.

At that time, Libellud — especially the game designer team — was thinking about new image interpretation games that would function differently. This story begins with a Dixit-like prototype from Alexandre, who has done a great job of various experimentations with this game mechanism. I LOVE his One Key, and I like Illusions. Regarding the third prototype, I'm frustrated. It's May 15th, nine days before the Incident.

Do ideas come from frustration? In my case, often. Sometimes by writing, regularly by playing. A kind of internal "Man, it would be awesome if we could do this" that plants in the head and waits and grows. Then — sometimes — it meets some other seeds and Wham! something is born.

A light bulb turns on in my head: "Man, it would be awesome if we could plan several Dixit/Mysterium enigmas", that is puzzles that the guesser must solve. I talk about this idea to Alexandre and Valentin since both work as game designers for Libellud. One of them gives me the right answer: "That's a good idea. Guys from the game design team will think about it." The other one asks me why I don't do it myself. Funny question. Why not? The idea goes through my mind for several hours, and I decide to try that very evening.

In my view, the game plays out like a line. I know I have several consecutive enigmas to solve with my deck of cards; I must think about the best way to manage them all in a row, as a road — which will mean accepting that I cannot play the best card if it matches with an upcoming enigma. Hmm, a line could quickly become boring. I want to choose. I need to choose. Other roads? Some intersections? A map! As if we would move on a Mysterium/Codenames board card by card.

Then comes another flash. I have already had a similar "pictures-map" idea: The game takes place in 1944 in a Paris made from square cards viewed from above. Players are members of the French Resistance or German invaders. Moves are done from card to card, based on image interpretation. The Resistance leader gives several cards to their team, and some of these cards can be intercepted by the invaders' team. There's no real-time gameplay in this idea. Only a draft of this idea exists among everything else in my online sandbox on Google Docs. It has never been materialized.

Playing something similar, but in a competitive way fits the new idea. However, going from a square card to an adjacent one gives only four possible replies to an enigma. This is not enough for a Dixit player. By allowing for diagonal movement, each enigma would have eight possible replies, but this would be too complex while playing in real time. In the middle, there is the number six, as in Dixit.

Six, and therefore the hexagon.

Before building the first map, I decide that I want the first moves to be easy, then for gameplay to crescendo. By removing a few hexes around the central starting space, I can give players first the possibility of moving to one of three spaces (which helps them start in the desired direction), then one out of four, then one of five or six, which is where the adventure begins!

I think I missed my graphic designer destiny

Players now have missions. They must find two spots of the same color (one of them is probably on a purple hexagon) without stopping on the forbidden tiles marked "Ø". The game is team-based and real time. In each team, a leader provides information by giving one or two cards to make their teammates move, then refills their hand to six cards. When receiving one card, teammates move their pawn from one space to an adjacent space. Receiving two cards enables teammates to move their pawn two spaces, which helps them move more quickly during the game, but which increases the number of spaces that they must consider, thereby increasing the risk that they go where the leader doesn't want them to. Receiving two cards enables them to move through forbidden tiles without penalty.

It's still May 15th. The night was fruitful, but nothing exists outside Google.

Episode II: Spies

I don't like cutting and gluing to make prototypes. This is an exhausting task that will be repeated 2, 3, 8, 42 times, possibly for nothing. Lots of ideas are here, crouching in a link in the toolbar of my browser, thirty pages of ideas, with 95% of them never having been used.

How to produce in an easy way around sixty cards? That's not difficult, but you have to commit an irredeemable act by selling your soul to the devil. Faint-hearted, please avoid reading the remainder of this paragraph and shield yourself from the image below while going directly to the next one. I gathered cards from Dixit that are related to mystery, doors, keys, races, vehicles, city, streets, documents, men in suits, money...and I used a pattern to cut them up. Four snips of scissors to shape the mythic Dixit cards into hexagons. Two evenings later, the result is quite satisfying.

Those with faint hearts should look away...

Technically, I have all that I need. On Tuesday, May 23rd, I informed Valentin. His answer: "Cool. You'll bring it tomorrow, won't you? - Oh… well... OK." The goal was to playtest it with colleagues before they would go to Libellud's games night.

Battle stations! I needed to come up with an ending for the game. Once brooding in bed, I came up with one, with poker chips standing for the markers that need to be collected because I love the Splendor effect.

I then tested it with my former partner in order to confirm that the movement mechanism does work. I set up the game. It looks like Codenames, a bit. I haven't figured that out before. At this moment, it freaks me out. (Today, I think that the game screen in Mysterium is also a kind of giant Codenames key-card. The matrix is just transcripted in a different way.) We start playing the game, me proud and anxious.

And it's a flop. Sure, there is only one team, but the game itself does not work. There is no tension, and we keep moving around the board several times without winning or losing. I am peeved. It's 10 p.m. I have a spatial Mysterium/Codenames that does not work. We are the day before the Incident.

I can't go to bed after this. I know I will not be able to sleep. I reopen InDesign, needing to correct two things: tension and experience time. Yes, the game will have more tension when two teams are competing in real time, but the game still needs more. However, there must be an accomplishment, too, and for the moment, that is too weak. I increase the number of markers in the game to nine, with them becoming briefcases containing intelligence. Some spaces now contain several briefcases in order to bring interaction and decrease frustration. By adding markers to the game, I create room for players to make mistakes; even if you start moving the wrong way, you have more of a chance of arriving next to another briefcase.

I create nine stickers ready to be glued on poker chips the next day. I also add two "Exit" spaces where you have to go after having collected three briefcases. The theme is slowly being shaped, largely thanks to the images: In a cold-war atmosphere, two CIA teams have infiltrated Moscow. Each team must pick up three briefcases, then be the first to exfiltrate.

To increase the stress, I add a new element: Each team's pawn is a 30-second sand timer. If a team arrives on a KGB agent space (the spaces that previously showed a Ø), the team has 30 seconds to move two spaces away from the agent without running into another agent. At 3 a.m., I go to bed. It seems to be clearer, but I didn't playtest it.

The day after, at lunchtime, all is printed and glued, ready to play.

Episode III: The Incident

I set up the game at 6:15 p.m. seven players are here to play. I start explaining in a desultory way. In addition to my fatigue, all the information seems to come out randomly. The aim of the game is hard to comprehend, even by my colleagues which are used to Dixit and Mysterium. I decide to observe rather than play.

The first game starts, some additional explanations are required, then it ends after a two-minute hesitation following a KGB agent's encounter.

The second game is longer and, for the first time, intense. The race is real, stress is present as well as some insults and bullying. I can't remember how it ends, but it is quite positive — except for comments about the 30-second sand timer mechanism.

A third game starts with Alexandre and Léa as chiefs of their respective teams. You can feel the stress everywhere.

Then the incident occurs: Everybody likes or enjoys the game. I hear some sentences such as "heart-stopper", "ready to be published", and "Régis must play to it!" — Régis being Régis Bonnessée, the founder of Libellud and the designer of Himalaya, Seasons, and Dice Forge.

To my mind, it was an incident, meaning something abnormal was happening. I was expecting a bit of employer benevolence with a sentence such as "Quite cool game, but you should maybe rework some mechanisms...". Instead, I was destabilized by this positive feedback. That was completely confusing!

Back home, I created a file to follow up the versions: V0 for the version of the previous evening, and V1 for the first playtest version:

• V0 - Moderately smooth. goals are too linear. We go round in circles. Multiple exit spaces are needed.
• V1 - Hard rules explanations. The aim is not easy to get. It works. Poor interaction.

The next day, I bring the game to the Game Design department to present it to Régis. They have already got some development ideas. One week after, they do the presentation to Régis, and I do not join them since I have work to do. A few days later, Régis calls me into his office to tell me that Libellud will take my game into development in order to publish it. That's a surprise, a very amazing surprise.

Mathieu Aubert