Designer Diary – Fragments

Some ideas are enough to create a game.
Others you develop when you have the time, tuck away in a drawer, forget about for a few months before coming back to them.
And then there are those that grab you from the very first second, shake you up, and eventually swallow you whole.
Five years ago, Fragments literally swept me off my feet.
And it still hasn’t let me go.

FROM INSPIRATION TO CONCEPT

WHY? — Telling a story

What really drives me in board games is, without a doubt, interactive storytelling. Telling stories that players won’t just read or listen to, but actually experience.
I was lucky enough to have two of my Unlock! scenarios published, but then came the frustration of having the next two rejected. Carrying on putting so much effort in only to face rejections made no sense, but I was enjoying it too much to stop. From there, it became clear: I had to create my own system.

Guilty as a Precedent

I’d already taken this approach a few years earlier. Faced with the masterpiece that is Sherlock Holmes: Consulting Detective, I’d tried to theorise what my own detective game might look like. That question had eventually given rise to *Guilty*.
So I want to do the same with the adventure game. Not to create ‘my own *Unlock!*’. But to find my own angle.

Initial intentions

I already have a few firm convictions. I want the narrative to take precedence over the puzzles. And, having learnt from the experience of a very wordy *Guilty*, I want to tell the story through images rather than long texts.

Video games have been exploring these issues for decades. So I started playing adventure games and point-and-click games as if they were game design manuals: how do they drive a story forward? How do they make a setting interactive? How do they turn gameplay into emotions? Little by little, I’m discovering a whole host of narrative design concepts and am beginning to refine the experience I want to create.
But I’m still missing the game itself.

WHAT — The lightbulb moment: the jigsaw puzzle as an interface

Following the COVID lockdowns, like many people, I rediscovered jigsaw puzzles. And that instinctive pleasure: fitting pieces of cardboard together.
In autumn 2021, an idea gradually took hold: what if the jigsaw piece became the interface for my adventure game?

Why a jigsaw piece?

The jigsaw piece has some rather extraordinary properties for a narrative game.
Like cards: it can feature an illustration that enhances immersion and storytelling. Like cards: its reverse side can hide something. You can move it, rotate it, turn it over, connect it, or separate it from a set. But it has one huge advantage: its edges are both ‘keys’ and ‘locks’. A piece can therefore validate an interaction or reject it.

This little piece of cardboard will, without needing any explanation, carry both the narrative AND the mechanics.

And above all, it evokes emotions. Curiosity before turning a piece over, satisfaction when a connection works, surprise at what it reveals…
Films, video games and books have powerful tools for enhancing immersion and creating emotions. But board games have something unique: physical objects that players actually handle.

It’s confirmed: for Fragments, that object will be the jigsaw piece.

Find a reference

Connecting pieces. Turning some over. Detaching others. The general principle is simple.
But making the whole thing feel intuitive is much less so.

When can a piece be flipped over? How do you indicate an action? How do you ensure that a connection made here triggers something elsewhere? How do you prevent certain actions without cluttering the illustrations with symbols?

My initial solutions work, but the instructions take precedence over the images. Players are likely to look at the instructions before the scenery. The game becomes mechanical. And that’s exactly the opposite of what I’m aiming for.

The technical breakthrough

Having spent a lot of time watching players get stuck on narrative-driven games, I’ve learnt that a problem often resolves itself when you ask the right question.

And the question I’ve been missing is simply: how do I make a connection interact with a distant room? The answer suddenly unlocks a large part of the system: you just need to use the number of connected rooms as an event trigger.

Some of the symbols disappear. And, for the first time, the pieces are actually starting to become an immersive interface rather than a medium covered in instructions.

The first test

One concern immediately springs to mind: the price-to-playtime ratio. A puzzle piece costs much more to produce than a card. In Unlock!, around sixty cards can provide about an hour’s worth of gameplay. If Fragments cannot achieve a ratio of 1 piece = 1 minute of play, the project may prove impossible to publish.

So I 3D-printed four pieces. I put together a tiny sequence using my first few orders, convinced that solving it would take me five seconds, especially as I knew the solution.

Incredibly, it took me several minutes to complete it.

For the first time, I began to realise that, perhaps, the concept might actually work.

A first scenario

In less than a week, I’m frantically prototyping a short adventure comprising around twenty scenes: *La Cabane*. I’m avoiding text altogether. The storyline is simple, but there’s a beginning, a middle and an end. A sense of time passing, mistakes to avoid.
It’s time to get some testers to try it out.

WHO – January 2022

Bruno Cathala

I’m lucky enough to go cycling with Bruno Cathala on a regular basis. After one such ride, I laid out the scenario for ‘La cabane’ in front of him.
For thirty minutes, I watched him play, his heart rate far exceeding that of the cycle ride. Finally, he turned over the last piece and looked at me:

– “Sit down and listen carefully.”
I sat down and listened. (It is Bruno Cathala, after all.)
“This thing of yours is a rare Pokémon.”

-… (I was listening but didn’t understand)

– You’re in a rather unusual situation. One I’ve never even come across before. The sort where EVERY publisher will want this game. So you’re in a position where there’s no point in showing it to everyone. You have to choose. And your problem now is who. If you get it wrong, the game will never work. Because your thing is fun, but it’s a huge publishing challenge.”

Strangely enough, it’s not joy I’m feeling. But immense pressure.

I’ve been lucky: I might just have drawn two aces. But it’s just a stroke of luck. And as poker players know, with a pair of aces, there are a thousand ways to mess it up.

And above all, everything would hinge on that. On the WHO. Depending on the WHO, the game would be different: my role in the project, the format, the deadline… Everything.

This idea of not squandering this opportunity is going to be a huge source of motivation for all the years to come. But it’s also going to send my stress levels through the roof. It might even, later on, upset the usual author–publisher balance.

Choose

Bruno explains to me the pros and cons of big companies and small ones… A few names are mentioned. But I’ve already made up my mind. It’s that obvious to me.

Iello placed their trust in me throughout the years it took to develop Guilty, my first game.
And the years spent on Guilty taught me just how much the relationship with a project manager can transform a game. So I don’t want to ‘choose’ a publisher. I want to ‘choose’ a project manager.
I need to be trusted, given how much energy I intend to put into this. A project manager who will help me make the game of my dreams, rather than one who will sideline me to make their own.
For every project, there’s the right person. The right co-author, the right illustrator, the right graphic designer…

For Fragments, I’m convinced: the right person is Xavier Taverne.

February 2022, Cannes FIJ Festival

Absorbed by Guilty, I’ve only got one business meeting in Cannes: with Xavier Taverne to show him the progress we’ve made on Monaco 1955. A few metres away from us, the Ravensburger jigsaw puzzles distributed by Iello are staring at me. His next meeting is running late. I’ve got The Cabin prototype in my bag.

“Got a moment? Fancy giving something a go?”

FROM CONCEPT TO GAME – you can go wrong at any stage

It’s all sorted. Amélie Scordilis, who manages the new puzzle range at Iello, will be my project manager. Under the leadership of Xavier Taverne, who’s still busy with Guilty. Perfect.

But testing quickly revealed the paradox of Fragments: the game has very few rules, and yet opportunities to ‘play it wrong’ are everywhere. Ergonomics, readability, connections, storytelling… With some groups, a single detail is enough to completely ruin the experience.

The prototype seemed almost ready for release. In reality, this is where the development of Fragments really begins.

THE URGENCY: We need to act quickly now.

At this stage, it all seems to have happened very quickly. Yet I’m convinced we need to move even faster.

The idea of a narrative puzzle strikes me as too obvious to remain original for long. And I have absolutely no desire to relive the development of Guilty, to witness the arrival of an avalanche of detective games, gripped by the fear that mine will no longer be relevant.

I’m a general practitioner and, up until then, the game had taken up whatever time I could spare for it. Fragments seems to demand much more. So I drastically cut back on my medical practice until I could devote most of my days to the project.

I thought I’d be taking a break for a few months. It’s been going on for five years now. And just as with Guilty, I’ve watched, with bated breath, as numerous narrative puzzle games have emerged. Two of them have even won the As d’Or…

TRAINING: to minimise iterations

Whilst Iello spends the next six months assessing the industrial feasibility, I can’t afford to wait. If the go-ahead is given, we’ll have to go for it.

So I’m turning this period into a crash course: I’m playing and replaying dozens of point-and-click games, studying videos and podcasts on game design and narrative design. The absurd combinations in Day of the Tentacle? The risk of frustration is too great… The hours spent stuck in Myst? Players might lose interest… Above all, I’m trying to identify what Fragments should take from video games, and what it should leave behind. To get a feel for where playful thinking ends and frustration begins.

Along the way, I’m learning a whole new vocabulary: onboarding, affordance, friction, guidance, stakes management, UI… It doesn’t provide the solutions, but it helps put the problems into words.

Little by little, the direction is becoming clearer: to make players think without stalling them, to guide without spoon-feeding, and to let the story unfold.

ERGONOMICS — Preventing the player from playing wrongly

Ergonomics quickly became one of the project’s biggest challenges.

Fragments has few rules, but they must be followed precisely. And during testing, this sometimes leads to rather puzzling results.

I’ve seen experienced players, used to big games, making mistake after mistake where eight-year-olds were making progress almost without a hitch. Some groups grasp the logic of Fragments straight away, whilst others need several rounds to get the hang of it.

I’m not yet entirely sure what makes the difference. But one thing is certain: the learning curve is a bit steeper than it appears.

Unlock! reassures me a little on this point: today its mechanics seem obvious, but the first few games could be very complicated. The challenge is crucial: to shorten this learning curve as much as possible, until the rules become second nature.

Streamline the commands

I was quickly able to do away with certain commands, which were used too rarely, such as ‘no entry’. Others change form several times.

FORECASTING PRODUCTION COSTS

At this stage, every game design decision begins to have industrial implications. Decisions will be made both to enhance the player experience and to streamline production.

For example, the pieces remain on their punchboards until they are needed: less set-up, fewer mistakes, fewer spoilers.

To reduce production costs and accommodate more text, the use of a booklet with dividers subsequently became the norm.

CONNECTIONS: No room for error

Alongside this visible ergonomics lies another layer, one that is completely invisible to the player: connection management.

Each piece must fit perfectly with those it is designed to connect to, whilst rejecting all others. The pins must work in every direction, the connections must line up correctly, and the cutting files must be accurate.

A single error can be enough to break a sequence.
We therefore ended up compiling a veritable ‘connection bible’, complete with tables, dimensions and check files to track every interaction.

ILLUSTRATIONS — An interface to balance

In Fragments, illustration is not merely decorative. It is part of the interface.

A mechanic may be sound, a rule perfectly written, and yet a test will fail because a character is not visible enough, or the function of an object is not understood.

We add a shadow, tweak the contrast slightly, move an element a few millimetres, or change the direction of a character’s gaze. Then we test it again.

Sometimes, after three or four almost imperceptible changes, a sequence that was consistently failing suddenly starts to work.

That’s where all the magic of the illustrator comes into play. But to allow them this freedom without compromising the game’s clarity, you first need to build a very precise foundation.

Designing sets

My first step is to recreate the scene in 3D to find the right angle. Jocelyn Millet had explained to me that this was the approach he took for Guilty. It’s a very effective way of using the space coherently.

Next, I use the most practical approach to design the optimal layout of the rooms.

And so began days spent searching for images and copying and pasting to match the illustrations to the pieces.

A new series of tests then begins. An object is too inconspicuous, a path isn’t clear enough, a piece of information gets lost in the background: we reposition, we emphasise, we retest.

It is only once this structure is sufficiently robust that the illustrator receives the brief and can make it their own, transforming it into something far more beautiful and immersive – without losing sight of what makes it functional.

GAME EXPERIENCE – Guiding without spoiling it.

Developing a game over the long term means having the time to get to grips with it. The numerous rounds of testing allowed us to fine-tune the gameplay experience to satisfy as many people as possible. And to challenge my preconceptions.

Adding goals:
Once again, the idea came to me from video games. Clearly stating what’s at stake – even reiterating it – doesn’t detract from the experience. Quite the opposite, in fact.

The fantasy of the open world

I was also obsessed with the open world. I wanted players to explore freely, choose their own path and discover scenes in whatever order they wished. The second chapter of Fungataï still retains much of that ambition.

But freedom comes at a huge cost: the loss of control over the narrative.

You need to know what the player has already seen, what they don’t yet know, write text that works in different orders, and account for a multitude of special cases. I’d practised this with Guilty. But the player experience in Fragments is greatly compromised as a result. Because the more freedom there is, the more lost players become. And they try to piece things together without really understanding.

By simply making Chapter 2 more linear, we went from 4 hours of gameplay to less than 2 hours. But above all, we went from a laborious experience to something more fast-paced and immersive.

Scoring or not scoring

For a whole host of reasons, I was convinced that an adventure game had to measure performance and satisfy players’ egos. So Fragments originally had a penalty system: certain mistakes would tick boxes that affected the final score. This created tension and prevented players from trying absolutely everything at random.

Iello pushed for us to remove it. I resisted the idea for a long time, before realising they were right. Nobody misses the score. It adds another layer to balance, so that a mistake doesn’t turn into frustration.

In a video game, we don’t measure the quality of our experience by the number of mistakes we make. So I gave in and applied the same logic here: if a score has no bearing on the story, there’s no need for it to exist.

From offering help to providing the solution:

I wanted to leave the help system until last. Fortunately, the gaming experience has evolved to the point where it is no longer necessary.

For some games, having a help system is important. But testing (and analysis of recent video games) has shown that when storytelling takes precedence over puzzle-solving, players mainly just want to progress.

Our attempts weren’t ideal anyway.
So simply providing the solution is more logical than adding another layer of frustration.

I had envisaged a streamlined and functional system. But the tests spoke for themselves: having visual aids showing the connections and available pieces laid out on the table is much more effective and reassuring for the players.

FROM GAMES TO SCENARIOS

NARRATION — When the gimmick is no longer enough

Very early on, another concern arises: the concept behind Fragments isn’t enough to make Fragments on its own.

The first time you turn a piece over or a connection reveals something elsewhere, it can be a surprise. The tenth time, far less so. After a few hours, the system becomes routine.

At that point, the gimmick is no longer there. What remains are the characters, the atmosphere, the revelations and the emotions.

So I set myself a fairly simple criterion: if I remove the puzzle pieces, is there still a story interesting enough to stand on its own?

For Solara, for example, I didn’t just want a science-fiction universe that served as a pretext for mechanical effects. The world and its characters had to have something original about them. And something to tell.

The same requirement applies to the overall universe of Fragments. I didn’t want a time machine, an interdimensional portal or an ‘it was all a dream’ plot device serving only to justify the possibility of telling any old story.

I needed to find something coherent and poetic enough to give the saga an identity, yet open-ended enough never to become a constraint on future storylines.

Mechanics driven by the story… or rather the other way round!

In fact, my approach to writing the scripts changed completely during development.

At first, I’d find an interesting mechanical effect, then I’d look for a narrative situation that would give it meaning. The early adventures were largely built around mechanical interactions and the core mechanics of point-and-click games: searching, associating, repairing, transforming, combining, revealing.

Nowadays, I try to work the other way round.
What do I want to tell? What do I want the player to feel? And only then: what kind of interaction could make them experience that?
At first, I wrote stories for the mechanical effects. Now I try to create mechanics for my stories.

WRITING — The text is part of the game

I was already trying my hand at writing novels whilst I was at university, and that part of the work is probably the one I enjoy the most.

I enjoy spending several months imagining the broad outlines of a world or a plot twist that turns a storyline on its head just as much as I enjoy spending a weekend on a single page, deleting a word, putting it back, moving a sentence around, and then finally starting all over again.

I haven’t had any specific literary training. Iello has given me free rein over the texts. It’s an enormous freedom, but also a huge responsibility. So, just as with game design, I’m teaching myself as best I can: through books, podcasts, analysis and rewriting. I’ve also learnt a great deal by watching Mathieu Rivero revise my texts for Guilty. Seeing someone reshape a paragraph you know by heart can be far more instructive than an entire chapter of theory.

Initially, however, my aim with Fragments was to do without text altogether.

The images and visual effects were meant to tell the story on their own. It was elegant. I took the idea so far that I built Chapter 2 of Fungataï in this way. But it was a failure: it extended the playtime to five hours!

Removing text does not, therefore, necessarily streamline the experience. Sometimes, it simply means abandoning the player… and losing them.
The opposite problem exists too. Too much text, and many players skim over it. Important information buried amongst four sentences can therefore disappear completely.

Writing becomes a balancing act: juggling functionality and immersion. Fine-tuning this accounts for almost half the time spent balancing a scenario.

And that’s also why I find it hard to completely separate writing from game design. A single sentence changes what the player understands, what they look at, what they anticipate or how they feel.

Rewriting a paragraph is still part of developing the game.

THINKING ABOUT THE GAME RANGE — A bigger story

I didn’t want to treat each scenario as a completely standalone adventure. Each box must work on its own, but also contribute a piece to a wider puzzle.
I think the ambition ultimately stemmed from a much more personal question: what did I really want to convey with this range?

Looking back, if there’s one thread linking the three Guilty scenarios, it’s that of relationships and, in two of them, the difficulty of having a child. I wasn’t necessarily aware of it whilst writing, but that was exactly what I was feeling deep down during the eight years I spent working on it.

Today, everything’s fine; I have two brilliant daughters. So what drives me now? What inspires me enough to want to write about it for years on end?

The answer lies in the question of passing on knowledge: how do we pass on our knowledge to future generations in a world that’s going downhill? And above all, what’s the point of passing it on when the future looks bleak?
I think we write better when we write about what truly moves us. And for me, that’s the question that takes precedence over all others today.

Fragments will therefore have a common thread – sometimes subtle, sometimes more obvious –: the difficulty of passing on knowledge, and the need to keep doing so despite everything. I obviously can’t say much more without spoiling certain plot points, but that’s the idea.
Naturally, things will evolve. Some ideas will fade away, whilst others will emerge. Some people will relate to them, others won’t. But having this overarching direction helps me maintain overall coherence and build the series.

CREATING A SCENARIO — The machine I thought would be simple

I came up with Fragments because I wanted a system that was easier to develop than Cartaventura or Guilty. In the end, I ended up creating something that’s harder to produce.

Designing a Fragments scenario involves imagining a poetic world, constructing a story and its twists, writing with very few words, devising physical interactions, visualising scenes in space, sometimes modelling them in 3D, creating prototype illustrations, testing, tracking down bugs, checking connections, monitoring the number of parts and the budget, then working with the project manager, the graphic designer and the illustrator on the final version.

And, of course, none of these layers operates in isolation. All this is done within the constraints of a tight budget. A new illustration might improve an interaction but could also make it less clear. Correcting it means changing the text. Then a new test reveals a fresh problem.
And the cycle starts all over again.

These days, I find it hard to imagine seriously designing a scenario in less than six months. Fortunately, every stage of the process is incredibly exciting.

AUTHOR — Dealing with Doubt

Being an author is a solitary endeavour. You swing between a few certainties and deep doubts. Sometimes you have to defend your ideas… only to be convinced of the opposite the very next day.

I’d like to wish those who aspire to embark on creating a game like Fragments – where the mental strain will pour out of every orifice – the best of luck in finding someone they trust, someone capable of answering, day (and night), the question: ‘Do you think this is any good?’ And who actually answers.

For me, that person is Jeremy Fraile. And quite naturally, he’ll probably be my co-writer on two upcoming storylines. I couldn’t finish without thanking him, as well as everyone who helped me keep going for 1,800 days, focused and motivated on a single project.

CONCLUSION – 5 years on

When I look back over the last five years, some of the figures seem slightly absurd to me.

I’ve probably spent over 5,000 hours developing the game. I’ve used 22 kilos of PLA to print over 7,000 puzzle pieces across seven scenarios – that’s about 7 kilometres of filament. 1,300 sheets of self-adhesive A4 paper. Thousands of prints, versions and tests.

And hours of almost daily discussions with my project manager (sometimes even with two!).

Collective effort

But these figures only tell part of the story. Over the years, the project managers didn’t simply stick to a schedule. The illustrators didn’t just deliver beautiful images. The graphic designers didn’t just put files together. Everyone was willing to adapt, make changes, redo work and sometimes scrap several days’ work because a test wasn’t conclusive.

It is probably this level of collective commitment that impresses me most today.

Why five years?

I would be unable to explain why this project took so long. It is probably the accumulation of thousands of small decisions, back-and-forths and moments of hesitation.

Spending a week on a bug that only occurs in one specific scenario. Two weeks making a mechanic robust, even though it seemed obvious. Redrawing an illustration because a detail always escapes the players’ notice. Removing a rule and reintroducing part of its effect in a different way six months later. Rewriting a paragraph ten times because it contains one word too many, or not enough.

For years, we’ve mainly been tweaking things: it’s too rough here, it’s too smooth there. A rule fits into a single sentence, and no one realises it used to take up eight. An image naturally draws the eye without anyone noticing the four previous versions. A paragraph takes twenty seconds to read, even though it took several weekends to write.

Letting Go of Fragments

For five years, Fragments has kept me busy, to the point where at times it has become a veritable obsession.
Yet, at some point, you have to stop tweaking things. You have to accept that a version exists, even if it’s imperfect.

In a few months’ time, Solara and Fungataï will no longer really belong to me.

It’s time to entrust them to you.