Interactive Fiction vs Visual Novel, Who Draws It

Interactive fiction vs visual novel is usually framed as text against art, which puts the difference in the wrong place. The real split is who supplies the image. A visual novel draws the scene and asks you to look at it; interactive fiction writes the instructions and lets you assemble the picture yourself. Almost everything else the two forms do differently — how far they can branch, what a new scene costs to make — follows from that one division of labor.

• A visual novel hands you the picture; interactive fiction hands you the instructions for making one.
• In Ren’Py, every state a character can be seen in is a named image file sitting in the game’s images directory.
• Art is combinatorial: expressions times outfits times backgrounds, so each new variable multiplies the drawing rather than adding to it.
• A text-only form can afford states an illustrated one cannot, because a state costs a sentence instead of a file.
• Neither form is the more literary one; they buy different things with the same hour of your attention.

What does interactive fiction vs visual novel come down to?

One question: who does the imagining. Interactive fiction is a story told in prose that responds to what you do, whether you type a command or choose a line. A visual novel tells much the same kind of story with character art, backgrounds, music and a text box laid over them. The words on screen may be equally good in both. What differs is whether the room you are standing in arrives as a drawing or as a description.

The tooling makes the split visible. Inform, the language most parser interactive fiction is written in, introduces itself as “a programming language for creating interactive fiction, using natural language syntax” and notes that it is “widely used as a medium for literary writing.” Twine, on the choice-based side, calls itself a tool for telling interactive, nonlinear stories. Neither mentions art, because in those forms there is none to manage.

Why does a branch cost more in a visual novel?

Because a state you can see has to exist as a file before you can reach it. Ren’Py’s documentation is precise about this: “An image name consists of one or more names, separated by spaces”, where the first name is the tag and the rest are attributes — mary beach night happy. Images go in a directory named images under the game directory, and “the extension is stripped, the rest of the filename is forced to lowercase, and the resulting filename is used as the image name.”

Read that as a production fact rather than a syntax rule. If you want your character angry in her coat at the harbor, someone has to have drawn mary harbor coat angry, or the layers that compose it. The states are not written; they are inventoried.

The arithmetic compounds the way branch counts do in a printed gamebook, which we worked through in Choose Your Own Adventure vs AI narrators. Four expressions and three outfits is twelve drawings. A second location doubles it. A scar in chapter nine that ought to persist doubles it again. This is why illustrated stories reuse backgrounds, keep casts small, and bring routes back together: not timidity, but the cost of every visible state being an object somebody made.

Text pays in a different currency. A new state in a Twine or Inform story costs a passage, sometimes a sentence. Prose can put a character in a coat she has never worn in a room nobody has seen for the price of naming them, because the reader’s mind renders both at no cost to the author.

What does a visual novel give you that text cannot?

A face you did not invent. That sounds small and is not. When an artist decides how a character looks when she is lying, that decision has authority the way a photograph has authority, and it stays fixed across every reader. Text cannot do this. Prose gets its power from the opposite property: it stays slightly out of focus, and the focus you supply is your own.

Three other things travel with the picture:

• Timing. A visual novel controls the pace of revelation. The tap that advances the line is a beat, and a well-built scene uses it the way a comic uses panel breaks.
• Shared reference. Two readers see the same image, so they can talk about it exactly. In a text story they compare impressions rather than views.
• A picture without the work of picturing. Not every reader builds images readily from description. For those readers a form that supplies the picture is not the lesser form. It is the one that works.

There is a fair objection to text-only reading here, and it deserves saying plainly: the argument that prose is more imaginative is often just a preference dressed as a principle.

Where does an AI narrator sit between the two?

On the text side, but with the sentence cost removed as well. In pre-written interactive fiction a new state still costs an author a written passage, so the reachable states are finite and drawn up in advance. A narrator that composes each scene at the moment you arrive has no inventory at all: the passage is written for the state you reached, or it is never written.

Fabledrift, an interactive fiction app for Android, is built that way, and it inherits both sides of the bargain. Nothing has to be drawn or written before you go somewhere, so the story can follow a move nobody planned for. Nothing arrives with an artist’s judgment in it either. Our image budget is spent once rather than per scene: every world has a shared cover painting, and each story gets a single illustration of its own, drawn at the story’s midpoint from the scene the reader has reached. The Emberwood and Ironreach Station each carry that one cover; the scenes are text.

Interactive fiction vs visual novel, side by side

The same six questions, asked of both forms.

Question	Visual novel	Interactive fiction
Who supplies the image?	The artist, in advance	You, while reading
What does one new visible state cost?	A drawing, or a layer set	A sentence
What is your input?	A choice, at a chosen moment	A choice, a typed command, or a written move
What is on screen while you read?	Sprite, background, text box	Text
Can two readers compare exactly?	Yes, they saw the same frame	Only impressions, not views
Where does it fail?	Recycled art, routes that converge	Flat prose, states with no detail

The middle row is the trade in miniature. Everything a visual novel can show you is finite because somebody had to make it; everything interactive fiction can show you is finite only in the way sentences are.

Which form fits the hour you have

Pick by what you want your attention doing, not by which form sounds more serious.

• You want to be shown something. Read a visual novel. Direction, music and a composed frame are pleasures prose cannot counterfeit, and they carry a tired evening better.
• You want the room to be yours. Read interactive fiction. The description is a set of instructions and you are the one building from them, which is slower and more absorbing.
• You want to go somewhere nobody drew. Only the text side can take you, and a live narrator takes you furthest, since not even the sentence exists yet.
• You want a place to start in one genre. Our guide to fantasy interactive fiction and where to begin sorts the entry points by what each asks of a reader.

The two forms are not competing for the same hour. One is a thing to watch closely; the other is a thing to build while you read. Most people who love one turn out to have room for the other, and the honest reason to choose between them tonight is which kind of work you feel like doing.