Rogue: Genesia

If you like me are avid retro gamers you couldn’t be attracted by this title! The Ghosts ‘n Goblins-sh look and feel and the pixelization take me back in time…Ok stop this emo thing and jump straight into the interview with Ouadi Huard, the developer of Rogue: Genesia.

Steam:	https://store.steampowered.com/app/2067920/Rogue_Genesia/
Twitter:	https://twitter.com/RogueGenesia

How did you jump into the game development world and why your started this project?

My adventure started when I was around 9 to 10 years old. My main time sink was making maps in a small free game named “Cube 2: Sauerbraten”. The map editing tools were incredibly easy and fun to use.
I think this is what really gave me the taste of creating stuff that would be played, trying to have nice maps, maps that are fun to play. Of course with the current me, I think they were seriously bad, but I believe they were important for me to understand how to make stuff fun.

A few years later, in my 12/13s  I started to practice 3D for fun. It was again my main time sink, putting all my week-end after school doing “youtube intro”, but I quickly preferred to work on making 3D models and modeling in general. 

This translated into me wanting to work in the video game industry as a 3D artist after I was done with High-School (I would have likely moved into Astro-physics studies otherwise).
However, at the same time, Modding tools were just created and made available to anyone on Dota 2. This is when I started to work on “Epic Boss Fight” . Just a small game-mode to explore a bit the tools available and for fun that I quickly left… 

However, a few months later I noticed a lot of players started to play the game-mode. I haven’t updated for all this time, noticing the huge gain in popularity, I started to work back on it, but this time, not just for me, but for the community.

It quickly rose to be the most popular mods, having today reached more than 4 million downloads.

Even so, this was my first time doing anything related to programming, and this quickly became impossible to maintain. I also made everything I could think of, and it was completed to me.
Having gained all this experience, I decided to make a similar game-mode, but you know BETTER and BIGGER. I started to think of custom inventory, persistent saves that are located in a server, weapon upgrades …

I achieved a good start of the game-mode after 6 months of work, but I quickly noticed, it wouldn’t work, for technical reasons, but for game design reasons, what worked for the first mod and made it popular, what how close it was to the normal game, it was easy for everyone to understand and jump into, this new game-mode was simply too different, it brought a bit learning curves to player that wasn’t looking for it.
This lesson learned in addition to the multiple updates of Dota 2 breaking mods, again and again, I decided to move elsewhere and learn Unity
There was then nothing really noticeable, I made a lot of prototypes, learned tons of new things, did my studies as a Game-Artist, I made a small Mod for Terraria named “Another Rpg Mod” that got mild success, and then worked in the game industry for a few studios.
Then I worked on a Planet generation plugin for Unreal-engine, World-Scape in 2021.

2022, I played a few months before Nova Drift, and then in early 2022 Vampire survivors, I really liked this whole upgrade idea, having a ton of enemies, and made a very small prototype that I quickly abandoned because it was just too similar to Vampire survivors.
A few months later, I was watching a YouTuber playing Round-Guard, sort of a mix between Peegle and Slay The Spire, and then I got a sudden idea!
What if I mix Vampire Survivors and Slay the Spire ?! 

This is how Rogue:Genesia was born, from the ash of my abandoned VS prototype.

After the success of Vampire Survivors many reverse bullet-hell / rogue-lite games have been released. Only a few of these games have succeeded. What in your opinion drove Rogue: Genesia to do better than other games?

Visual, It’s a bit of a sad trust, but the visuals of a game are super important to gain visibility. I am not necessarily speaking about Nice 3D model or good HD texture. The art direction, having a whole game with a coherent art style is very important. Of course, having a hook is also important.

 Is there any particular “twist” or any aspect that you think makes your game unique?

The Slay the spire type of stage progression, with artifacts, is I think one of the main hooks of Rogue:Genesia. Another one that is not obvious at first but definitely noticed by the player, is the depth and variety of builds you can have.

Games-stats.com* says that “Rogue: Genesia” generated something like $300.000 in 2 months! Did you expect a similar result? What did you feel when you noticed that the game was gaining that much traction? 

I have no issues being transparent about what I gained.
As of today (15 December 2022) Rogue: Genesia generated $448.426 gross, removing chargeback, VAT, and Steam cut, which is about  $280K, having about half going to me, and the other half going to my publisher.
(To the game-developer, A 50/50% share with a published is a very bad deal, I advise you to aim for a 70/30% – 70% for the developer – share, I went with the 50/50% with my publisher because I have a very special relationship with them)

When I was developing Rogue:Genesia, I expected to make about $30K, So I was very happy when I saw how well the game was selling. However I’m someone that is usually calm, so I did not have a crazy reaction ^^’.
However, I also know success such as Rogue: Genesia is quite rare for a game, especially for a first-shipped game.I would also thank the “How To Market A Game” community for their very helpful advice, feedback, and good practices, which definitely helped the game a lot to do well today.

* source: games-stats.com

Tell us more about your pipeline and the tools you’ve chosen to develop “Rogue: Genesia”?

There is nothing really crazy about tools, I’m using a lot of the native pipeline (or unity made package).
For the development, I mostly use the tools I use for any game I worked on during my other job beforehand. Photoshop for the sprites, Substance Designer for some texture work, Maya for the 3D modeling, Visual Studio Community 2022 for the programming.

You are mixing 3d and 2d in such a great way!
Would you mind sharing some hints about how you have achieved this amazing look & feel?

I worked for a few years as a Technical-artist, specializing in rendering, so doing Lighting and Shader was my job for quite a bit of time now so it’s not as easy as “having just a few tricks to follow”.
The art style is greatly inspired by Octopath travelers. I spent a lot of time analyzing the color they used, how the sprites were made, how many details were skipped, how they fit together 2D/3D, and what sort of post-process were used.
Then I spent a lot of time tweaking my game to fit the style I aimed for.
There was also a lot of time spent on making the sprite.

The player’s running animation actually took me 2 days of work, and I am a professional artist. This can give an idea of how long this can take for a developer whose main job is not art.

However, every skill can be learned, and it’s mainly a question of learning and training more than if you can do it in the end. Practice a lot targeting something and you’ll end up having good enough results

There are moments in the game where you see thousands of AI on the screen! How difficult has been to keep good performance and responsive controls?

AI are as simple as they can be, and Since there is a lot of instantiation done by unity, there weren’t huge performance issues for the Monsters, however, having one single Script handling all the monsters at once helped with it.

The main challenge in performance was the projectiles.
To be more specific, The huge speed of creation and deletion of projectiles was a big issue for the engine. I ended up using pooling on the visual, while also using a single centralized script to handle the projectile’s AI.

However, I think a lot of optimization can still be done,e.g. not using collision to detect if the projectile is touching enemies could be an optimization. I’m thinking of using “Spatial hash maps” for this part, also the Unity Burst compiler, and maybe the C# Job system if they can bring performance gain.

Another Important optimization in the game is the usage of the Particle system for the damage number. Instead of creating normal text mesh, I’m using a GPU particle system, this greatly helped reduce the floating damage numbers performance cost.