How does one make a 3D game prototype?

I planning to make a game Similar to Castlevania: Lament of Innocence with getting to different areas with Portals.
Does a Prototype require the main area, and each Level?

What is the goal of the game?

How does making a prototype serve your goal with the game?

It is up to you to figure out what the questions you need to answer are. Generally, a prototype is just whatever the fastest way to answer the important questions is.

For instance, if you want to confirm the production pipeline for how you’ll create each level and then the code archietecture to move from one to the other, then you know that you’ll need to build enough code to move from level to leave and create at least two levels to switch between. And you’d want to create at least one piece of each part of the levels so you can understand how that process will go.

But no, there wouldnt be an yreason to create everything becasue that is not the fastest way to answer the question.

1 Like

So there’s a LOT sitting in this question. I’m fairly amateur when it comes to game development but a “prototype” means different things to different people.

Below are list of components that I tend to check off when I’m thinking of something to build out.


  • None
  • Bare minimum
  • Minor polish (varies from person to person)
  • Release Ready


  • Startup Level / Debug room
  • Demo Level
  • Graybox
  • Polished
  • Release Ready

Visuals / Audio

  • None/Defaults
  • Basic Representation / testing samples (unfinished animations, uncustomized models, etc)
  • Minor polish
  • Release Ready

The basic premise is that for each element of the final product that I want to see when released, I want to know how far towards “complete” I need to go.

The biggest question you face when choosing this is what is the purpose and who is the audience?

If you’re aiming to show off some simple mechanics on your twitter for some early production discussions, just having the basics of the mechanic is good enough.

If you’re talking to investors or producers, you may want to have some polish to your visuals and audio, even if your mechanics aren’t 100%.

1 Like

For prototype you need basic movement and single level that fits movement. Then you add some fighting and enemy. And you have prototype.

Next step would be to make vertical slice of your game. This is just like prototype, but you need to have prototype of all systems in game, like crafting options, multiplayer, lobby. Everything your final game would have.

1 Like

As others have said, what a prototype requires really depends on you. Or more specifically, what you’re looking to answer.

Are you trying to figure out if you can make a basic combat system that feels good? Then your prototype needs a character (even if it’s just the basic Epic Mannequin), the bare bones of your combat system, and something for your character to fight.

Are you trying to figure out if your concept of traversal across levels works well? Then you need a basic movement system, a representation of what a level might be laid out like (even if the level itself is just untextured blocks), etc.

Are you trying to make a flashy demo to convey your vision to people, in order to convince them this would be a cool game and that it’s worth putting in money/time/effort/whatever to help make it?

And so on.

As an example, a friend and I are working in our spare time on an exploration-heavy multiplayer cooperative rogue-like, which incorporates some parkour-y movement and some unusual combat mechanics. That’s a lot of pieces that don’t usually go together in quite that combination, which leaves us with some big unknowns. We took those unknowns and boiled it down to three questions we want to answer.

  1. Will the parkour elements feel smooth and enjoyable? If not, that’ll detract from the game rather than adding to it. So we’ll need a prototyping step to address that question.
  2. Will our procedural generation make levels that feel interesting to explore using that movement? This being a rogue-like, the worlds will be generated at runtime, on-the-fly. But if the worlds we can generate feel bland and uninteresting – or alternatively, if it feels like the same large set pieces smacked together in a limited number of ways – that suggests we might need a different approach to this. Moreover, if we can’t generate levels that provide good opportunities to use that parkour-y movement system in a satisfying manner, that suggests either we need a different approach to generating them, or that the parkour system will just be needlessly complicating.
  3. Will our combat system feel good? We’re looking at some some weird stuff with combat, and it’s an open question if that’ll feel good/fun/engrossing, or whether it might get in the way.

We’re focusing on prototyping in phases as a result. First phase? See if we can put together a movement/parkour system that feels good and fluid. Second? See if we can generate levels that it’s fun to explore using that movement system. And third? See if our idea for a combat system works well and is enjoyable when mixed in with our movement system, while traversing those environments.

Maybe we learn as part of this prototyping phase that what we have in mind just doesn’t quite work the way we want it to, and we should rethink our idea. Maybe we see that it mostly works and needs a bit of fine-tuning. Maybe it just plain clicks and we can’t wait to start fleshing the systems out for real! We don’t know; the prototyping phase is there to help us figure that out. But we don’t need flashy graphics or background music or detailed models or anything in order to answer those questions, so those are things to tackle later, after we know the idea is viable. (Or not.)

My friend and I are skipping any visual glitz right now and just using basic shapes and the Epic Mannequins, because visual glitz isn’t needed to answer our questions. But in some cases, “flashy graphics” might be what you’re trying to answer, if you have an idea for an unusual visual style and need to know if you can make that work; I had an idea for a game done entirely in watercolor at one point, so my big question was “can I make the game look like a living watercolor painting, and do so in a compelling fashion that doesn’t distract from the gameplay”. (The answer was, unfortunately, “no”. But someday, I’ll get a watercolor shader I’m happy with…)

In the end, what bits of your game you should focus on putting into a prototype is something that, at least in large part, you’re going to need to decide for yourself; you know better than we do what the questions you’re trying to answer with a prototype are.