← Return to topic

I have a few questions as a "starting designer"

jsfijol · 10 Sep 2026 at 00:58 · permalink

I was thinking i wanna try to do this in my free time, potentionally make a living from this since i've always been into the games and i feel like i wanna try this. So i have few questions to start with.

1. What engine should i use and what engine you are using these days?

2. If you were starting from scratch today, how would you learn level design?

3. What are the most common mistakes beginners make in level design?

4. What advice do you wish someone had given you when you first started learning level design?

benceeh · 10 Sep 2026 at 00:58 · permalink

Use whatever engine you want. The most popular is going to be Unreal or Unity, but Hammer and Hamer 2 have a very strong following still and strong documentation / free tutorials and resources. What type of game are you trying to make a level for?

Youtube. Believe it or not when I first started learning level design youtube -didn't exist yet-, so I spent a lot of time reading forums and official documentation. Now there's a step by step guide for anything and everything you want to do.

Going too big too fast. "I want to make an open world with" literally years to make that as a team. "I want it to be super custom and unlike anything" you're going to be doing a LOT of 3D modeling and texturing for that. default textures and community assets are absolutely perfect for starting out and learning.

Learn the restrictions and limitations of your tool, and research how professionals and established independent creators set their design flow. My origin and true level design love is still Hammer for HL2 engine games. It took me a while to come to terms with area portals and leaks and how objects loaded in. Once I had a lot of experience I started designing my levels around area portals FIRST and thinking about how to move the player around with the best culling possible so the levels played buttery smooth on the worst hardware from the first alpha to the completed project. That's not a limitation with modern engines, but they too have their own quirks. Looking at forums and discord communities and seeing what's holding people back first and learning from those mistakes will rocket your education forward. That said mistakes often lead to better ideas, so experiment as much as you can at first.

Secondly PRE-PRODUCTION IS KING! My levels start on paper every time. I've got tons of sketchbooks full of buildings and hallways, and my cell phone is full of pictures of interesting architecture and details out in the world. Knowing what you're trying to build and how they look irl (when applicable) is how you go past making square rooms and square buildings. Learning a bit of architecture helps too.

Let's say you're building a normal office building. Ok, so square shape usually, square front door square windows. Does this building have air conditioning? Where are the units? Are they on the roof? Are they out back behind the building? How does the AC get into the office? Is the ceiling inside drop ceilings or exposed (vents in the tiles or exposed ventilation shafts)? What kind of office? Do those usually have a reception? Is it a production building? Maybe they have a natural gas tank in the back or a dumpster. Is it an old building? Then it's probably made of bricks. What style bricks and ornamentation did buildings of that tim get? Is it an office building from the 60s or 70s? Then it's probably got some brutalistic features like concrete around sunk back windows. From the 90s? Probably a plate glass front. Did it used to be something else? Maybe you can make it one of those "used to be a Pizza hut" style buildings that's been retrofitted. How many people work there? Is it a cubicle farm or is it a legal office with individual offices. Maybe it's a storefront that's temporarily a tax office so all the offices are just larger cubicles in a big empty room.

Knowing what you're making is your starting point for research both irl and images online. From there the details flow wildly to fill the room.

chroniclesofdoom · 10 Sep 2026 at 00:58 · permalink

There are a lot of really great books out there.
“101 things I learned in Architecture school” and “Architectural approach to level design” are great books.

Generally with level design, you are thinking about how people interact with spaces and trying to create space to evoke emotions.
So learning about architecture can be very helpful.

Im a big theme park nerd and I’ve found that studying Theme Park design has also been very helpful.
In both practices you are creating interactive, immersive spaces and experiences.

Basically any books on design can be helpful.

I think in terms of learning an engine or beginner tips, the best thing you can do is find a game you like with a level editor and start building levels for that game.
Most editors like Unreal and Unity are good but lack any sort of gameplay mechanics and level design is all about making an environment FOR game play.
Most in-game level editors function relatively similarly to other engines so when you do eventually move over it won’t be too much of a shock.
Plus then you can actually play your levels (sometimes with other people) and get feedback and improve thst way.

Some common pitfalls I see, most people want to make something that looks nice, but level design is less about making something pretty and more about making something fun and functional (Funtional?)
When starting out, I recommend just keeping your designs and block outs simple. You would be surprised how much you can do with just a simple cube. Plus if you focus on layout and design first before figuring out all the details, it’s much easier to change things later since you won’t have dozens of little small details you also need to adjust.

Also, when starting out, try and get as much feedback as possible from others and don’t worry about failing.
Don’t also get caught up on making something perfect.
Try different things, get someone to play test it, get feedback and adjust it or move on and learn from your mistakes.

Constructive feedback is the best feedback! It’s nice to hear “oh this was great” or “this level is amazing”
But the best feedback is always hearing what you can improve because that is how you get better.

There is so much to learn in level design, the best way to start is just by doing.
It’s an incredibly tough industry to get into right now though, so again I just recommend finding a game that you enjoy with a level editor and building levels for that.
Find a project that excites you and focus on that.
Good luck!

howveryGracie · 10 Sep 2026 at 00:59 · permalink

1: Yo comencé con Unreal 4.27 allá por 2022 y sigo en esa versión (aprendí solo)
2: viendo gameplays, jugando de manera "objetiva" juegos sencillos y leyendo documentación y viendo tutoriales.
3: Delirarse con el tamaño del juego y lo que esperan como primer proyecto. (todos esos abandonan)
4: que me ponga un kiosco y deje todo.

ecstazy_ezy · 10 Sep 2026 at 00:59 · permalink

1/ What engine should i use and what engine you are using these days ?

* Unreal Engine
* Unity

3/ What are the most common mistakes beginners make in level design ?

* Jumping straight in the engine to start your project (working without LDD or any sort of preprod work)
* Confusing Level Design with Level Art/Environment Art
* Trying to learn the theory too hard instead of doing actual Level Design
* Designing the level you want rather than the level the game needs

2/ If you were starting from scratch today, how would you learn level design ?

If you want to get serious about Level Design just make levels, try out multiple engines and workflows. Unity and Unreal Engine are the industry standard right now so I would focus on that.

You have two types of projects I would recommend you:

* Find a game you love, with a decent Level Editor and make something for it
* Use ready to use project templates and design a level on it ( eg. AGLS)

Here is a list of a few stuff you could work on:

* TPS (SP): ALSV4 / ALS Refactored / AGLS
* TPS (MP): UEFN / Gears of War 1 & 3
* FPS (SP): Metro Exodus / Prodeus / Left 4 Dead / Quake / Doom / Dying Light 2
* FPS (MP): Counter Strike 2 / Halo: MCC / Halo: Infinite / Unreal Tournament 2014 / Black Ops 3
* Platformer: Trials Rising / Banjo-Kazooie / Mario Maker / Mario 64 / N++ / Battle Block Theater / Marble Blast Ultra / Marble It Up! / Mario Odyssey
* RPG: Skyrim / Fallout 4 / Divinity: Original Sin 2 / RPG Maker
* RTS: Impossible Creature / StarCraft 2 / WarCraft 3
* Puzzle: Portal 2 / HUMANITY
* Racing: Trackmania

Try to work on your workflow, find what works for you and try to be a little 'professional' about it. Try to follow a certain schedule, have deadlines that fits your life, work on documentation (LDD) instead of jumping straight into editor. Check resources and portfolio and see how other level designers work, read/watch some LD resources, etc...

Maybe after a few projects you can try to do collaboration with Level Artist/Environment Artist to start understanding their constraints more and getting used to making compromises.

Join "The Design Den" and "Funsmith Club" Discord server and you will have a bunch of design resources.

ttblsf · 10 Sep 2026 at 00:59 · permalink

1. Unity, it is easy to use (great integration with Photoshop and Maya) and have great example projects to get you going.
2. Study, like properly dissect, levels of related games to the one you're developing or want to make. Why and how do they work? Yours have to meet that standard, if not, try and try again.
3. Nobody outside Reddit, YouTube and Instagram cares about a pretty dioramas. So dont focus on the visual aspect and dont assume the player or audience to invest in your lore.
4. Find your target audience, test on them well and test on them often.

smigolek86 · 10 Sep 2026 at 01:00 · permalink

I find it super easy to use unreal's modeler to greybox levels quickly. Might need to enable a plugin for it to work in current versions however.

d4rkkauan · 10 Sep 2026 at 01:00 · permalink

The editor-first advice is strong. I’d graybox one route, note where testers hesitate, then change one affordance at a time. Fixed goal or free exploration?