Helm Indie games · Aotearoa ← Back to Helm

Guide · Getting started

From an idea
to your first playable.

Five steps for anyone who wants to make a game and doesn't know where to begin. No experience needed. Take them in order, and don't rush the first two.

Step 01

Choose your engine

A game engine is the software you build your game in. It handles drawing, physics, sound and input, so you can focus on the game itself. We recommend starting with one of these three. Each can make a real, finished game, so pick the one that fits what you want to make.

Engine Best for What you'll learn Cost to start Difficulty First small game

Unity

Best for2D and 3D, mobile and VR. The biggest library of tutorials and ready-made assets.

What you'll learnC#, a widely used programming language

Cost to startFree (Personal) until your studio makes or raises US$200k in a year

Difficulty Moderate

First small game1–2 months

Unreal Engine

Best forHigh-end 3D: realistic visuals, shooters, big open worlds.

What you'll learnBlueprints (visual scripting) first, then C++

Cost to startFree. 5% royalty only after a game earns US$1M

Difficulty Demanding

First small game2–3 months, and it needs a strong PC

Roblox Studio

Best forMultiplayer games with a built-in audience, especially younger players.

What you'll learnLuau, a beginner-friendly scripting language

Cost to startFree. Earnings come through Roblox's own currency

Difficulty Beginner-friendly

First small gameA few weeks

Difficulty and time are rough guides for someone brand new, practising a few hours a week. Prices and licence terms change, so check each engine's own site before you sell anything.

Can't decide?

Brand new and want the gentlest start, or want players straight away? Try Roblox Studio. Want the most flexible all-rounder, for 2D, 3D or mobile, with the most tutorials? Start with Unity. Dreaming of big, realistic 3D worlds and have a strong PC? Go with Unreal Engine. You can always switch later, because the ideas you learn carry over.

Step 02

Decide what you actually want to build

Before you open the engine, get a clear picture of the game in your head. A small idea you can see clearly beats a huge idea you can only feel.

  1. Collect your inspiration

    List three games you love. For each one, write down the specific thing that hooks you: the jump that feels perfect, the eerie music, the satisfying upgrade. Those moments are your ingredients.

  2. Write a one-sentence pitch

    "A game where you ___, in a world where ___, and it feels ___." If you can't fit it in one sentence yet, keep cutting until you can.

  3. Find the core loop

    What does the player do every ten seconds? Jump and dodge? Stock shelves and serve customers? That repeated action is the heart of your game. Everything else is decoration.

  4. Build a vision board

    Gather screenshots, colours, sketches and music that capture the mood. One page is enough. It's what you'll come back to when you lose your way.

  5. Shrink it

    Your first game should be small enough to finish in a few weeks: one mechanic, one level, one goal. Save the big dream for later. You'll make it better once you've finished something small.

Step 03

Learn one engine and its true foundations

Pick one engine and stay with it. Every engine is built on the same handful of ideas. Learn these properly and everything else gets easier.

How to learn it

Follow one complete beginner course from start to finish, without skipping ahead. Then rebuild the same thing without the video. The second time is where the learning really sticks.

Recommended course · Unity How to Make a Game in Unity 6: Full Course for Beginners Jimmy Vegas on YouTube Watch on YouTube

Each engine also has free official learning material: Unity Learn, Unreal Engine learning and the Roblox Creator Hub.

Step 04

Make a rough start

Now take what you pictured in Step 2 and get a rough version of it into the engine. It's meant to be ugly. The goal is to feel your idea for the first time.

  1. Use boxes, not art

    Grey cubes for the player, flat planes for the ground. Developers call this "greyboxing". Art can wait until you know the game is fun.

  2. Build only the core loop

    Get the one thing the player does every ten seconds working. No menus, no story, no settings yet.

  3. Play it, then change one thing

    Play for two minutes, notice what feels off, fix just that, and play again. Small loops of change and test are how games get good.

  4. Save often, and use version control

    Keep backups, or learn the basics of Git, so one mistake never costs you a week of work.

You're done with this step when

Someone else can pick up your rough version and play the core idea for a minute without you explaining it.

Step 05

Keep building, and keep talking to people

This step never really ends. The developers who make it aren't the most talented. They're the ones who kept going and didn't do it alone.

Ready for the next stage?

When you're working on your first real project, the Building guides cover scoping, prototyping, working with others, and getting through the middle.