Story and art synergy: how both shape Storyngton Hall’s art pipeline
Plot is key in our project, so our screenwriter’s needs shape our entire art pipeline. But sometimes the screenwriter shapes the story based on already-drawn assets, too. Here’s a peak behind the curtain at our whole art process!
Hello everyone! Storyngton Hall is a classic match-3 game, but it also has a plot with well-developed characters, and actually, quite a lot of people participate in creating art for the game — the entire process involves about 25 team members from our studio, as well as a number of outsourced talent. My name is Vadim Udaltsov, and I’m a 3D artist at MY.GAMES. In this article, I’ll discuss how the Storyngton Hall art pipeline works.
This article covers just a slice of the big picture: a large number of specialists from different fields are involved in the process of creating games. In order to coordinate their work, we need a clear pipeline that clarifies who is responsible for different aspects of the work and at what point they need to step in. And at MY.GAMES, we provide the necessary environment to ensure that the work process is comfortable for everyone. Our close communication between different departments means that we are able to create games that make our users around the globe happy!
As mentioned, Storyngton Hall is a classic match-3 game with a plot that sees the Green family inheriting a mansion — however, the premises turn out to be quite dilapidated, and thus, in need of repair. Additionally, we later come to learn they’re not the only ones living there; more and more characters gradually appear in the story, and the machinations of the plot begin to further unfold.
In fact, the work involving the artistic components of the game begins with the script itself. A screenwriter comes up with ideas, writes them down, and creates a draft. At this stage, it’s important to clearly describe the main concepts, then, later, the writer will synthesize everything together into a story — almost as if it were a novel. Indeed, there really are some similarities with the novelistic format to note: the game consists of many chapters, so the script can actually be presented in this way.
When the main story is ready, it’s divided into chapters, quests, and cutscenes, as well as a list of dialogues and character behavior. That looks something like this:
Next, everything is entered in a table.
Tasks are then created for the concept artists, animators, layout designers, UI artists, VFX artists, and translators. And when the table is ready, the artists gather to read the script and distribute tasks. During this, it becomes clear what animations, characters, and objects will be needed.
After this, the main process begins, we call this the “shuttle pipeline” — reflecting the regular travel of assets as they are constantly transferred from person to person, iterated on and improved.
It is often actually the visual components that determine the script: we have laid a lot of groundwork and created numerous ready-made objects, so a writer can evaluate the location layout and decide which ready-made objects can be used and which need to be drawn. They then edit the script and fill in the list of plot objects.
When plot objects are drawn, they’ll end up in a cutscene and this helps the cutscene artist navigate their business. Next, the draft version of the cutscene is sent for review — if everything is good, then all the objects in the scene are finalized and sent for review. After this, they enter cutscenes where the final build takes place, and on top of all this we implement VFX effects.
Creating an environment
In terms of temporal setting, the game takes place during the Regency era, and this is an important detail to note because we strive to comply with all the key aspects of this time. Graphically speaking, the location itself looks like a period-accurate estate in an isometric view.
All of these 2D graphics are created in several stages: searching for references, concept modeling, modeling itself, rendering, post-processing, and exporting to Unity.
Below is an example showing how an artist creates a location draft based on references — and it’s important to pay attention to all details: the interior, clothing, furniture.
Then comes the object concept stage. By way of the script, the artists are able to understand the necessary objects to draw; accordingly, they begin to select shapes and colors for the concepts.
Once the concept is approved, it goes to a 3D artist who does the modeling in Maya. Incidentally, the actual number of polygons is not so important to us — the most important thing is that the object looks nice and accurately portrays the concept.
Then the model goes into rendering, where we use Maya and Arnold. We render everything in a certain lighting: for the buildings, we use a gray fill as well as material from Arnold, and we render in two light sources looking in opposite directions. After rendering, we add masks to make it easier for the artist to apply colors during post-processing.
That said, we don’t always use gray for materials. For example, with this chair, it’s clear that the legs are made of wood, so we can use the appropriate material on that rendering. Also, to simplify any subsequent post-processing, the 3D artist applies a certain texture to the model.
Once the model has been fully rendered, it’s handed over to the artist, and they begin the post-processing process where they complete all the necessary elements; all of this happens in Photoshop.
If you take a closer look at the image above, you may wonder: why isn’t the vegetation modeled? There’s a simple answer to this question: a 3D artist will spend much more time creating leaves than a 2D artist.
At the next stage, the asset is transferred to a technical artist, who imports it into Unity and configures it to create locations. Using the script, the image is converted into the corresponding prefab. In tech art, we divide objects into two types: ground objects and billboards.
Ground objects are located strictly on a plane, and characters move along them. Heroes are always above the level of these objects — these include carpets, grass, and earth tiles.
Billboards always “look” directly at the camera, creating the illusion of 3D. It turns out that they always rise above the ground, and the character can pass through them.
Since the characters in the story are renovating a mansion, each new room initially looks a little shabby.
Gradually, the heroes put the mansion in order.
We’ll note that there are active objects in the rooms, and there are also static ones. Static objects include everything that doesn’t change over time, and with active objects the opposite is true — that is, they can transform. In other words, since the player can fill the space as they wish, each active object has a state that can change during the game.
Accordingly, active objects must have modifiers to change their state. Let’s look at the example of a chair. Let’s say we have an empty game object on which we’ve applied a Decor Placeholder component — this component contains all the visual information about the multiple states of an object, and the player can choose any of them.
The next stage is adding VFX and SoundFX effects. The player can choose any chair and place it on the location, and for every action there are corresponding effects.
That wraps up location production; the next important part of the workflow is creating the characters around whom the whole story revolves!
Character Creation
All the heroes in our game are realized as 3D models, and the process of creating them is divided into several stages: concepting, modeling, texturing, rigging, animation, and exporting to the game.
First comes the concept stage; we write down all the characteristics of the character, their behavior, what animations they will have, and so on. Once this description is ready, the concept artist selects references to help facilitate the rest of the work. At this stage, it’s important to find the desired character and get the “feeling” of them.
When a reference is found, all of this is transferred to the concept. A concept is a 2D image in which a 2D artist depicts a character in an A-pose or T-pose. A concept will additionally see the artist depict the character in their characteristic stances, and it also shows the character’s size relative to others.
As soon as the general concept is ready, it’s sent for 3D modeling in Maya. (I don’t use sculpting in my work because it takes a lot of time — I’m a fan of modeling based on primitives.)
We have several internal requirements for character models. They should deform well and look smooth since players see them up close. This is why we make the mesh on faces denser; this allows you to achieve soft (and correct) facial deformation to convey emotions.
Since the game also has character customization options, we use several models to organize our prefab. When the main model draft is ready, it’s divided into segments: body, hair, clothing. This makes it easy to customize your character.
After the model has gone through a thousand reviews (and finally everyone is happy with it) the texturing stage begins. We perform this in 3D Coat and Photoshop. Since we don’t have lighting as such, all the nuances are depicted on the texture. We only use one color texture, which is simply filled with color. For ease of texturing, I divide the model into segments so that I can paint the tongue, oral cavity, teeth, and eye sockets.
The texture map resolution is 2048x2048. This is then compressed in Unity to the required size.
At the same time, the rig is being prepared. We pay special attention to the animation of faces and clothes — we strive to achieve believability and stylistic consistency.
We also use special bones in our hands so we can place any object in the characters’ hands.
And it’s not only characters and organic objects that have animation systems, vehicles do, too. For example, we have a carriage that can drift.
The next stage is animation. Below is the animator’s working window; this is the Graph Editor — here you can see keys, controllers and curves.
The models are then exported to the engine. In this case, the model is divided into several segments. Materials and textures are also exported. We don’t use light sources, so we have material without light sources — we use unlit.
When all the models are ready, we start creating cutscenes. We work in our own cutscene manager where we can add any animation for any character, set a specific path, and write dialogues.
This is one of the final stages of conveying the script’s story to the viewer.
Working on gameplay
Work on a new mechanic begins with the game designer: they look for a suitable idea and describe the concept. Then, the artist gets down to business — their task is to find a form and image for the tiles that will look harmonious within the game.
The final result is exported to the engine.
We have our own tool for developing match-3 levels which allows the game designer to arrange the tiles as desired. We can select any segments, features, and background as desired, and more than 1 thousand levels have already been created using this editor.
The pipeline for developing art together with a script is a complex process that combines the competencies of many specialists from different professions. In our project, the plot is of particular importance, so the entire process meets the needs of the screenwriters. We make sure that the art elements are appropriate to the chosen time period, and also support the writers’ desire to tell the story as much as possible.
That said, sometimes the opposite happens — the screenwriter shapes the story, taking into account the assets that have already been drawn. The result is an inextricable connection between different aspects of the game — they shape each other during the development process. Thanks for reading!
