Why isn’t Rivals of Aether Free to Play?

Dan Fornace
11 min readOct 18, 2017

--

We get this question a lot whether it is on Discord, Twitter, Reddit or anywhere else where Rivals fans hang out. It is also a common discussion when comparing Rivals of Aether to other platform fighters especially since Brawlhalla has found success in the space and that most genres are looking at the Free to Play model to see how it could fit their game.

Truthfully, before Rivals of Aether had begun any actual production, it was planned to be a Free to Play experience. As a gamer, I started playing League of Legends in 2011, after a friend failed to convince me to get into Heroes of Newerth. I was blown away by the low barrier to entry. I was also impressed by the game’s holding power as my friends and I all stayed up late playing League of Legends even though we had active World of Warcraft subscriptions running at the time. The game did a good job making you want to spend but not have to spend to get addicted. I bought a few skins in my first week because I wanted to stand out.

LOLOLOL OUYA…
Hey that Roguelike thing sounds sweet. Sorta like Abyss Mode + platforming? We should do that…

If you go back and look at the original design documentation for Rivals, then you would find verbiage about Free to Play and how it would work. It was basically just straight up League of Legends model though. I wanted to have skins that you could purchase and characters that you could grind to unlock. You could earn points by playing either single player or multiplayer and we would do a character rotation just like League of Legends.

This type of model for my first game after leaving Microsoft Studios would not have been that surprising. I had just worked on Season 1 of Killer Instinct before starting Rivals of Aether in 2014. Killer Instinct had a similar model where you could play for free and either buy characters one by one or buy them in a set for each season of around 8 characters at once.

So what gives? If Rivals was originally slated to be Free to Play, a model that had already proved successful on PC, then why did it release on Steam Early Access in September 2015 for $14.99? To answer that question, you have to go back to the very first month of Rivals development.

I began Rivals of Aether in April 2014. I had just left Microsoft Studios and had saved up enough money to have about 2 and a half years of living expenses as a runway for launch. I started the game in Unity 4.3 before the 2D tools had been relaunched with new features. I saw what Moon Studios was doing in Unity with Ori and the Blind Forest and I saw that Unity was picking up steam in terms of online support and active forums. I knew it would be a good engine to build a platform fighter in especially if I needed to hook into a server later to control a marketplace and other free to play features.

But I hit some snags early in. From grade school through college, I had made my fun game projects in Game Maker from version 4 up to version 8. I knew how to use the engine and how to build 2D games the way I wanted. Rivals of Aether was designed as a 2D pixel art experience because I would be handling all the art and the creation of pretty much everything initially with the exception of audio. I knew how to manipulate sprites the way I wanted in Game Maker but was struggling in Unity. I remember one issue that halted me in Unity was controlling animation speed based on attack timing. I was having issues having it draw the frame of animation that I wanted. I knew exactly what I would have done in Game Maker. As a quick test to see how I would solve it in GM, I started up a project to get to where I was at in Unity. In 2 hours I was able to get even with over a week of work in Unity mostly thanks to familiarity with Game Maker. I realized that if the art was going to be 2D then I might as well work where I was most comfortable so I left the Unity project behind and continued in Game Maker Studio.

That decision about 2 weeks into development did shape the project. Rivals of Aether was never meant to be a huge experience. I knew that I wanted to build a small fighting game. I wanted to establish the world of Aether and move onto bigger projects ideally in the same world. At the time, Flashygoodness and I were a two man team and we had experience in building local 2D games. Features like online multiplayer, DLC and other features core to a free to play design were really pipe dreams and not even on the table initially.

Rather than trying to expand into a larger team while using Game Maker as our platform, we decided to scope the game down. We were looking at 8 characters, 8 stages, a Story Mode, and some practice features as all we would like to accomplish and to do it in a year and launch in summer 2015. We decided this would be a one-off game priced around other indie gems at the time. ($10–$20 range) Also since we were doing hand animated pixel art sprites, we weren’t really planning on having skins or other Free to Play cosmetics that 3D or skeletal animation games excel at. Don’t worry though, we eventually figured out a way to make some cool skins while still having a custom color option for players.

Rivals of Aether on the big stage at Gamescom 2015.

After we showed off the first trailer in July 2014, we did get a little buzz. The ID@Xbox team was interested in us so we joined the platform in addition to planning on a Steam release. It was also at that time that we realized we couldn’t get away with local only so we went out and found a team who could help us with online multiplayer. We ended up working with Code Mystics who have done a lot of work on both console games and fighting games. They provided us with a Networking framework that we use in Game Maker to have synced input 1v1 online matches.

Thanks Code Mystics for the Multiplayer Library!

Like pretty much every game deadline, the 2015 summer goal was unobtainable. The game was increasing in scope and adding online multiplayer was a big challenge as that meant our game logic had to be deterministic. We only send inputs back and forth so both games have to run the match exactly the same as the other would. That meant syncing up all our random variables and making sure all game logic took place inside of a game loop controlled by one object. We did however manage to launch on Steam Early Access in September 2015 so I guess we weren’t too far behind.

We even added some exclusive colors for our Early Access players!

But our solutions to online and the game in general left us open to some issues. Cheating in Game Maker games can be a real problem. Players can download cheat engines to change pretty much any variable in the game. Since we were only sending inputs online, players could cheat their game but it would simply desync the match. We did not have to worry about cheating players ruining another player’s experience with double damage or increased movement. But it also meant that we have very little control on what one player’s client is doing. They could be sending messages to us / Steam that they won every match or that the match gave them 1 million coins instead of 50. We do place limits on any interaction with the tracking server but as a small team, we’re not really equipped for handling cheating and support for a large online game. Couple that with being in an engine where most users do not have to worry about live services or cheating and we would be alone in climbing a mountain of combating cheaters.

We went the route of ensuring that players could only affect their own client and experience, but we have some fixes to reduce cheating too. We have limits on what players can post to leaderboards and how many times a player can disconnect in ranked matches before their leaderboard rank is removed entirely. While these solutions aren’t perfect, they are just enough to keep people from ruining the game for others.

Is 12 seconds even possible? Who can say?

But opening the game up as Free to Play would make our solutions far from perfect. Players could cheat to access any character even if they do not own any. They could cheat to report wins, affect their currency, etc. We would need to rewrite the most of the client logic in order to safely sync with a server. This would be a huge undertaking and require us to redo a lot of the progression in the game. It would also mean that offline Rivals would be different too as we would likely not be able to reward currency when players are offline to avoid abuses.

Also being Free to Play opens us up to way more players who are looking to circumvent the rules. Purchasing a game at an upfront cost is an investment and the type of player who buys it is more willing to play by the rules than those who would not.

The biggest reason, much more than the engine and cheating which are both solvable problems, is simply our team size. Rivals of Aether started as a two man production and is still a small development team with seven people working on it regularly. Our strength is in what we do well as a small team — character designs, original movesets, cool pixel art, amazing music, and a balanced competitive roster. Our weaknesses include most of the aspects of a good Free to Play game including fast and regularly released content. Characters take us three months each and skins take a week and a half while taking time away from our main animators and programmer. We work hard on getting new stuff out but are not able to get things on a tight content schedule without our main development suffering.

These are the non 2D art people. We work with some amazing artists who are cooler than us.

Secondly, Free to Play games have potentially large userbases and therefore require extensive online support - Ranked Modes, Fun Modes, Ladders, Seasons, Rewards, Daily Challenges, Loot Boxes, etc. Our online features are functional but are not as equipped as some of our competitors. Part of that is because of our initial scope did not include online, but the main factor is simply people power. If we had a programmer specifically on online features then we could improve them, but finding someone with enough familiarity with Game Maker Studio and netcode has proved challenging.

If Rivals could not be Free to Play, then why are we releasing DLC characters now? Well, DLC characters are a bit different for us. We can rely on Steam to handle the ownership (which works both online and offline) and it allows us to expand the roster without handling the security. Players could likely find a way to cheat and access these characters locally, but they have already purchased the game so would be less likely than someone who has the game for free.

Could there be a future where Rivals of Aether is Free to Play on Steam? Honestly, it is possible. It would not be the type of Free to Play that I originally imagined where earning currency could unlock characters. But I could see us doing a version where you get access to a couple characters and the rest are behind a paywall. Free to Play would certainly help our playerbase and is a great way to inject new players into a multiplayer focused experience.

But this would be shoehorning the design into Rivals similar to other games going Free to Play later into their life cycle. And like those games, the impact would not be the same. It would read as a last ditch attempt rather than the initial design. The systems we have now would not fit well as we would still not be able to rely on earned currency due to our lack of server control.

But wait, does that mean if a Rivals 2 was created then it would surely be Free to Play? Well if we ever did build a Rivals 2, then it would depend on where the market is headed. We would want the game to have as many players as possible but also be a competitive and fair experience. We are not blind to the negative reaction to loot boxes and other practices that are currently inhabiting AAA multiplayer games. There is a struggle right now between accessibility, popularity, revenue, fairness, and engagement. Certain practices that generate tons of revenue don’t feel great as a player and certain practices that players love don’t work for everyone, which can be seen when the League of Legends model of grinding for characters fails in smaller free to play games. We would want to build something that is designed from the ground up to fit its business model so the competitive design and free to play design are not at odds with each other.

So in summary, here is your TLDR:

Rivals is not Free to Play because of our engine choice, inability to combat cheaters and small team. It will likely never be Free to Play, but that option is still on the table as a way to increase the player base later. If a Rivals 2 is ever made, then maybe it will be Free to Play and maybe it won’t.

I hope you enjoyed that insight into the painstaking process of deciding on our business model for Rivals of Aether.

--

--

Dan Fornace

Game Director and Designer. Creator of Rivals of Aether. Worked on Killer Instinct (2013) and other games at Microsoft Studios.