Building Worlds Adrift in UK

We were extremely lucky to talk to the developers of Worlds Adrift – one of the most unusual multiplayer online games of the last couple of years. The developers from Bossa Studios were kind enough to discuss some of the technologies used during the creation of the project.

Introduction

Bossa Studios is an independent outfit based in London, UK. We started the studio in 2010 to create games with flair by inventing new genres. The idea was to make things no one else was doing, and standing out for that. These days we’re about 35 people who have worked on a wide array of games: Monstermind, Merlin, Surgeon Simulator, Deep Dungeons of Doom, Twelve a Dozen, I Am Bread and, of course, Worlds Adrift.

Choosing Game Engine for Worlds Adrift

Every year the tech team does an evaluation of what engines are available in the market, and a discussion is held about which one we should invest on as a studio. In the past, for our first games (Monstermind and Merlin) we created our own engine, down to the multiplayer layer. But we felt that it was a distraction of our core purpose of creating games, so we adopted Unity as the default platform for our projects.

There are many reasons for this decision, ranging from ‘speed to prototype’ (a very important measure for us as we jam every month)  to online references and support. While no engine is ‘perfect’, and ‘better’ is heavily dependent on context, Unity excels on all fronts we deem important for creating games.

I personally don’t know where the ‘it’s only for mobile’ comes from. Perhaps it’s the fact its shaders pipeline is not as developed or understood as Unreal or Crytek’s, but one only have to look at demos such as Blacksmith or Adam to realise its potential. Maybe it has to do with the fact a lot of beginners and indies prefer it. We find it to have its quirks, yes, but it seldom lets us down.

Game World

The main pillars of Worlds Adrift are: Sandbox, Freedom, Persistence and Simulation. Worlds Adrift is a game trying to make quite a fair amount of things that have never been done before, so we’ve got to be diligent to remain on target or it creeps. Anything we implement in the game has to touch on all these pillars at once, or doesn’t make the cut.

For instance, if you explode a rock to gather some metal from within, its pieces must stay behind as physical objects. Players must be able to interact with these pieces in many ways – perhaps hoisting them onto a ship to rain them down on another ship below; maybe hoard them on a cliff and set them rolling down onto an enemy party. Here you see these pillars at play enabling emergent gameplay, gameplay we neither script nor control, gameplay that opens up the door to endless creativity and possibilities.

That’s what beats within the core of Worlds Adrift and empowers its freedom to players.

Artistically there were several considerations, ranging from a style that stands out at first glance, to the practicality of low-polygon and simple textures done by a small team producing such a large game.

Lastly, the most important aspect of the project is the open development approach. Not just checking with our community to gather ideas, but go far beyond as with the case with the Island Creator tool, to have them directly collaborate with the development of the game – watching your biggest fans create more than 1,000 islands in a couple of weeks was quite humbling. We have experienced a lot of good moments thanks to our community, and spared ourselves from many mistakes thanks to their input on our work.

Colors

It has purely to do with what emotions we want to evoke as game designers. The stage of Worlds Adrift is called ‘Foundation’, a planet ripped apart by careless over-exploitation, a social criticism to our handling of our own planet. We needed a style and palette that invoked something lost, reminiscent for a glorious past now forgotten. That’s also heavily reflected in its soundtrack, which intertwines with the visuals.

Going for a more realistic style would slow down production and play against the focus of the narrative on large scale events (instead of details).

From a production angle, we use a wide array of tools ranging from Photoshop (2D), Maya (animation and modelling), ZBrush (organic modelling) and some pipeline plugins we created ourselves to allow the movement of assets from artist to deployment within the game.

Using Unity

There’s very little ‘out of the box’ functionalities or plugins at use in Worlds Adrift. Due to its very nature (a massive online multiplayer sandbox game) we had to use Unity in very unorthodox ways, including messing with its physics engine. For instance, the game dynamically remaps coordinates to overcome Unity’s 4Km square size limit for scenes — a limit we couldn’t put up with on a game of the scale of Worlds Adrift (as far as we know, the largest online multiplayer map ever created).

There are a few plugins worth mentioning, though. We use a custom version of Puppetmaster for ragdolling our characters; the Steam integration plugin helps us talk to the platform; and our own collection of shaders to create the volumetric clouds you see in the game and which are shared amongst all players so they can serve their strategic purpose of cat and mouse playground for battling flying ships.

SpatialOS

SpatialOS does many things at the same time, but of particular interest is simulation and scalability. Traditionally, when the subject is online multiplayer games, there was always at play a trade-off between processing power and gameplay: physics are expensive to run on a world inhabited by thousands of players exchanging states through the network, so MMOs became a stats affair giving rise to MMORPGs, while action massive multiplayer games never took hold.

With the scalability and swarming capabilities of SpatialOS we now have the resources to run advanced AIs (as opposed to linear and conditional scripting), simulate ecosystems (no more creature spawning points), allow persistency of millions of entities through time ranging from pebbles to shipwrecks, and more importantly run physics for thousands of players, finally empowering emergent gameplay in large multiplayer scenarios.

It’s a paradigm shift in game development that above all opens up a myriad of design possibilities that, before, just couldn’t be explored at scale.

Blending the Game World and Gameplay Mechanics

It’s a good mix of lore, balance, Occam’s Razor applied to gameplay mechanics and a large dose of risk-taking from the team. We’re navigating uncharted waters with the game, a lot can (and does) go wrong, so in a way the biggest challenge is to work smartly on features that unfold onto many gameplay possibilities, paying off a larger amount of fun per mechanic; as opposed to something specific, that only affects a single aspect of gameplay or is restricted to a single use.

When working on a sandbox game it’s important to consider mechanics that interact with one another creating exponential results. Here less is better: give the player just what can be subverted and used in many different ways. Finding these mechanics is a huge challenge, by far the largest we ever faced as a team and even as individual professionals that have seen and done a lot before working on Worlds Adrift.

Making an Artistically Stunning Game with a Great Technological Backend

I believe it’s down to two elements: an amazing team capable of anything thrown at them; and a setup where this team can work in a flat structure, with lots of collaboration and zero egos or politics getting in the way of doing their best.

We all are aware we’ve undertaken a project that is anything but straightforward to realize. If a drop of our energy was to be wasted on something other than solving the problems we face in a daily basis in order to progress, we would fail.

To set out to do something unique, difficult, is to understand it won’t be easy, and be ready to fight all the way there. Only a team that is supported within a good environment can do that.

Now let’s see if we can pull this off…

Henrique Olifiers, Gamer-in-Chief of Bossa Studios

With a games career spanning more than 25 years starting on 8 bit computer development, Henrique has been in the forefront of every gaming revolution ever since, from the demoscene of the late ’80s to mobile, online, freemium, VR, social games and premium indie games.

His previous experiences include Head of Gaming at GloboTV in Brazil, in charge of exploiting IPs in the gaming format such as Heróis e Vampiros, a very popular online trading card game in the early ’00s; and My Big Brother, the first online mobile game ever made, where players followed in real time their favorite Big Brother show contestant in a Tamagotchi format.

Afterwards, Henrique held the position of Head of Game Development at Jagex, overseeing the overhaul of its main title Runescape HD; and the team behind the cancelled MMORPG Mechscape /Stellar Dawn. He left Jagex to join Playfish as Studio Director in the UK, where his team successfully produced the BAFTA-nominated FIFA Superstars.

As the co-founder of Bossa, Henrique is responsible for the studio strategy, management and game design.

INTERESTING LINKS

LIGHTING SCENES: REALISTIC LIGHTING FOR AUTODESK MAYA AND MODO

3D CREATION WITH MAYA, UNITY AND 3D COAT

STUDYING LIGHTING & MATERIALS IN UNREAL ENGINE 4