This entry will be a bit more technical mostly about our game development experience so far. I was thinking on what I would advise me if I could send back a message in time to the point where I started game development. I've collected some of the major things I would do / do different now. Most of them are common mistakes.

1. Scope

We are working on our game Warriorb for about 2,5 years now. The funny thing is that it started as a small 6 months project. Back then we were working on a more ambitious epic fantasy story driven action game focusing on melee combat and traps. When we realized how long that project could take we decided to start with a more easy to make game. We ended up at the concept of a side-scroller game where the player controls a funny ball character who has two limbs.

Although the core idea was quite simple as it turned out we couldn't keep to it. We added more and more ideas. Like there should be a fight system, you know just a simple one. Or we should add dialogues, that really doesn't take much time. Maybe the most significant change was the idea of using splines defining the path of the character instead of straight lines. This way we ended up with a game where you need full 3D environment since many parts can be seen from different angles.

Back there we couldn't estimate well enough how much work those changes mean. This is a sort list of some of these elements and what they did:

- Dialogues:
- Concept: just some text, maybe the less dev time consuming content
- Cost:

• We ended up creating a dialogue plugin for UE4 which basically took about half year for one of our programmers. We decided that this plugin will also help others so we released it as open source and you can find it int the Unreal Engine marketplace or you can get it from the Git repository.
• We got negative feedback on dialogue UI so we remade it from scratch. It also took a ridiculous amount of time. (Maybe a month for our 2D artist and weeks from our programmer.) (Btw you can check the result here.)
• We still need to hire a writer to proofread our texts.

- Spline based path instead of side-scroller
- Concept: it's fun and unique
- Cost: level design time multiplied. There are some smart area where we can make advantage of reusing environment elements for other puzzles or using them as background but it also means a ton lot of additional work we could have skipped if we stayed at the side-scroller concept.

- Combat system:
- Concept: nothing fancy but still fun
- Cost: a hell lot. We need to create 3D enemy characters, design combat gameplay, write AI. Even if it looks really simple it is complicated.

- Skill tree:
- Concept: you can't make a game where the player can't level up
- Cost: way too much and we couldn't really make it work with both the combat and puzzle system. We found out that we are trying to develop 2 games at the same time. We dropped this idea and instead focused more on items, weapons and clothes.

As conclusion this is how I would start an easy project now:
- focus on only a few gameplay elements and make those as good as possible
- do the math! Search for games using similar elements as yours. Check how many people worked on it and how long did it take for them.
- don't start to work on fresh ideas. Each idea looks super cool when you make it. Let yourself some time to decide if it is good or not.
- ideas are hard to dismiss but you won't be able to complete all of them. Compare the new idea to the others. Does it work well with them? If you can choose only some of them which ones would you pick?

2. Working order

Prototyping the ideas is crucial in a work which takes years to complete. In the beginning I always wanted to create parts to their final form. Later we needed to completely rework or delete these areas because they didn't work the way we expected. After events like this I had to accept that no matter how cool my idea is, it just won't work for the first time. I need to iterate and test the ideas before working on details.

We had a year when I worked full time and our gameplay programmer could work next to his job and that made it even more complicated. I preferred to work on environment and 3D assets because there I could work on my own. Or I just used the existing features to build parts. This is something I want to avoid in the future. Everything is relevant to gameplay so that is what needs the most work in the beginning.

3. Player learning progression

It is really easy to forget about the fact that I've been playing my game for 2,5 years now, I know exactly how the gameplay mechanism works and others won't know what is trivial to me.

When we started testing I was quite desperate because players failed to complete what I thought was super easy. We realized that teaching the player how to play is not done when you give them hints on which button to press. We broke down the puzzles to smaller elements and made the environment so that it forces the player to do the expected solution. Using this method we started to create 'learning puzzles'. We also repeated these puzzles in a bit different form between more complex puzzles where the same move was needed just to refresh what the player learnt.

My note on this that testing should be started as soon as possible and fresh testers are always needed.

4. Concept, story, time and motivation

Warriorb started with almost no story content, based on a joke. Later it turned out that a joke can look funny for a week or even for months but it will definitely lose its charm with years. As time passed the story become more and more complex and serious but there were some elements I couldn't replace without breaking the overall concept. I'm happy with the final result of the story and mood, yet if I ever create a new idea I will definitely ask myself: 'Is this something you could work on for years?' I know it might sound odd but in some way you must be compensated for your work. There must be something what keeps you going on. If you are doing your own game it must be something you like even after years.

This is it for now, see you next time! In the meantime I would love to hear what you would do different in you first game.