Long life to Embark studio and its fabulous procedural artists dream team !
truly excellent and inspiring to read. Would have loved to read some on the texturing since that is top-notch.
great environment with a lovely serene sense. Thanks for the write-up!
Supergaming‘s CEO Pavel Agoshkov talked about their new title for Android Evertile: Battle Arena, discussed the way their studio functions and what is required for developing and publishing a game alike.
80lv: Pavel, could you please tell a bit about your company? Where are you guys based, what do you do, what projects have you worked on? What were your main releases?
In the past, I was CEO and co-founder of Israeli studio Playtagon which was backed by ex-playtikans and Plarium. Playtagon was working on mobile MMO – Heroes Rage. However, the game hasn’t shown any significant KPIs during early access and investors decided not to keep the studio any longer. Meanwhile, my friend has started a small game venture and invited me to join him after I left Playtagon. I’ve joined the great team of professionals who had a huge experience in gaming.
Evertile: Start of the Project
80lv: Could you talk a little bit about Evertile and it was originally envisioned? What were the things that you wanted to have here and what inspired you? What was put at the core of this production?
I’ve been playing 2048 game for quite a long period of time and suddenly I realized that this game might be significantly improved with a slightly different approach. The one that we all used to play – RPG. Like, what if tiles were heroes? With attack and HP characteristics? What if tiles don’t stick one into another but just hit one another? I’ve shared this idea with a friend of mine and surprisingly he followed and started testing it. There was a goal to find fun around it. Current gameplay has shown up after dozens of attempts with a team. However, that was only beginning of a journey and it required to bring meta game in. As a matter of fact, Clash Royale have had the greatest metagame ever. The idea becomes simple – mix success stories to come up with a new one. In the current case, it was a mix of new fun core gameplay with the successful metagame.
How Does the Studio Function?
80lv: Let’s talk about the way you approach art production. How is it organized in your studio, how much of it is done in-house/outsource?
Well, we are a small indie studio where all processes are pretty simplified for the sake of speed of development. I remember we used to have a very complicated production pipeline with a bunch of specific software at my previous job. I mean JIRA for bug tracking, Confluence for documentation with flowcharts etc., Bamboo for continuous integration plus a pile of other useful and sophisticated stuff. However, it never added up speed. I have nothing against the proper pipeline unless it harms the speed of development. We cared about bugs too much. Here at Supergaming we use a simpler set of tools which allows us to progress fast and not distract to managing process too much. At least at the beginning.
Regarding production, we outsource only sound design. The rest is done within the studio efforts even though some team members work remotely.
80lv: Could you talk a bit about the way you’ve started working on the effect you’ve been using on the battlefield? What is the way this process is organized, how do you do this crazy animation of skulls, lightning, and physical attacks?
Sure. We take a few approaches here. Some assets we buy at the asset store, some are produced in-house. It depends on the needed requirements. As for skulls, it is pretty easy.
Skulls are controlled with a UI script that defines its “levitation effect” and light comes out as a particle effect. Sometimes we use Spine animation if we can’t provide a complex approach by engine means. But that approach has some pros and cons.
80lv: What’s the way you’ve crafted that amazing battle area?
Honestly, arenas are the most complicated art assets in the game. And here is why.
As you understand Evertile looks like a board game with 5 x 5 cells battlefield (arena). That requires us to keep the board properly faced a player. Which means no perspective is allowed. The first restriction. From the other hand, if you make all arena surroundings as the top view, you would probably see nothing exciting as only roofs and tops of all subjects. So, it would require to rotate all subjects in a way player see them either in isometric or orthogonal projections. Isometric doesn’t really help. See the first restriction.
So, orthogonal is the only option we were left with. Now, how to combine the top view board with orthogonal surroundings? And that was a trick. Artist simulated camera view in 3D software and placed the board. Then we were adding up the rest of the surrounding to get a whole picture. Not mentioning that they should stick to the concept of arena and grid of coordinates for tiles positioning at the same time. All arenas are made as 3D objects, then they were rendered to let us use them in the 2D game environment.
80lv: What were the main restrictions you’ve faced during the production of the content for this particular game? What are the restrictions to art given by the mobile platform? How do you manage to overcome them?
As long as you create a 2D game you have way fewer restrictions from the performance requirements than a 3D one. The only restriction you should really care about is the size of the build unless you have tonnes of content which is another story. But speaking of Evertile, I must say that this game has tonnes of artwork which must be properly organized with atlases.
Using too many atlases at the same time will just slow your performance down and this is the area of constant optimization. For example, you might have a button made of one sprite or a few. Depends on UI style the artist has set. I would say that experienced developers always keep that in mind of which approach to use. Simple UI makes life easier but it would look poor. And that’s a question of balance. Art directors wish to enrich the visuals and tech directors require it vice versa. Anyway, there are tonnes of tricks of how to fit them both. Experienced art-directors understand the way art assets get into the game so they try to predict the tech guy’s feedback to keep move forward. It is always a dialogue. Eventually, everyone works on the same global goal – success. That’s a teamwork.
Testing the Mechanics
80lv: How do you actually test the mechanics? You’ve mentioned you based some of the mechanics on stuff you’ve learned from Clash of Clans. Could you elaborate on that and explain the process of testing?
Good question. We always create different hypotheses that seem interesting for us to test. E.g. what if we speed up tiles movement or add the 4th tile into the hand? We drill down the idea and create a short document with mockups to try to cover all aspects that this idea might affect on. It is very important since you might ruin some core mechanics and destroy the fun. E.g. previously we used to have a random battle deck order where tiles spawn onto the battlefield randomly. We decided to test the idea as if a player would set the spawn order manually. In order to deliver a proper solution, we should revise some UI elements to let the player easily control this aspect without harming his will to play as he used to do it before. So, prior to every hypothesis implementing we look at all potentially affected areas to address it properly.
Regarding Clash Royale. I would say that these guys have taught millions of players to get used to a specific UX. Lots of small developers took the same UX approach (and we call it “Clash Royale party”) as they accepted the fact that millions of players were already familiar with “game menu” and some basic engaging mechanics. Like, where is the store? What’s the quest? What is inside? What is the chest? What will it provide? What should I do with that? What is the clan? What is the top and where is that? I mean, from the perspective of the game UX, we didn’t want to invent a new one. It is like a d-pad stick, console Joystick or shooter menu. We wanted to show off the original and unique gameplay supported with well-known game menu set.
Game Production: Time & Cost
80lv: How long does the production of the game of this size usually take? What were the longest tasks? Also, it would be awesome if you could approximately say how much it cost because there’re many myths about the cost of production and the burn rate.
Prototypes were made quite fast, in 14 days. If you can’t make a demo of the core gameplay in 14 days – walk out. I can’t disclose the costs obviously but I could point that these type of games would probably cost you not less than $200K with the experienced team. An inexperienced team could double up those costs easily.
However, the most important thing is a time. Usually, a quality game takes 2 years at least and we are far from that yet. We still have lots of stuff to do to deliver world-class quality. And I mean not only production but all aspects of the game development. Eventually, this type of games are likely GAAS (game-as-a-service) and developers must deliver this service properly
Publishing & Promoting
80lv: Finally, how are you publishing this product? How do you plan to push it to the audience on the platform, where the competition is so intense?
This is the hardest question. We believe that great games need great marketing. Also, we are quite aware of huge competion on this market and that player’s attention costs a lot. However, as I said earlier, fun never gets older. Players are looking for new fun every time.
Honestly, we have followed quite complicated sub-genre – a real-time turn-based game, as it is PvP with turns. Thus we are trying to focus our marketing efforts on sending the message like “royale players, look at this original gameplay”. 4 minutes turn-based battles in real time with real opponents are quite excited.
In regards to marketing details, we focus mainly on the Facebook audience for now. We hope we will grow fast when we split our efforts among different marketing channels. However, marketing doesn’t end with UA and it starts before the production. We build a community every day and listen to our players every day. Players are wise, they always know what’s happening. Developers must be in contact with the community all the time.
In fact, we believe that Evertile becomes a huge GAAS which opens a variety of opportunities for gaming and gamers which we will announce pretty soon.
P.S. By the way, at the moment we have an ongoing contest in Evertile. Make sure to check it!