Firstly, I’m sorry if you’ve been anxiously awaiting an update, we’ve been busy working on the game, but it doesn’t always bring shiny new things. I’ve also been busy travelling across America with Jenny, having spent some time in NYC and Chicago, we’re now in San Francisco for GDC.

My first GDC has been amazing, San Francisco is a beautiful place and I can’t wait to come back. The panels, talks, parties and lunches in the park have been inspiring, galvanising and immensely positive, especially after the past year and the ugliness which has clouded the industry. I’ll leave GDC today loving games and wanting to make them more than ever.

Meanwhile, Pal has been toiling away on data persistence, or how we save, load and otherwise handle game data. Ensuring that we have a robust solution here is important and we want to make sure that saved games will be compatible as we keep working on the game, too.

I spent most of January re-factoring some overly complex code that handled the game start up, world construction and spawning. We’re waiting on some changes in Unity 5 before we can achieve our principal aims here, so for now the game world we construct consists of a single landmass.

We’ve been travelling since early February, but I’ve been working on the road; adding more complex spawning for animals. Each species now have ecological preferences, including the range of altitudes they prefer to live at, proximity to forests, predators and other environmental factors.

Re-working our spawning systems led me to investigate so called influence maps, which are representations of space, most often used to influence AI, but with many other uses too. For example we might have one map that shows the amount of tribe activity at any given position in the world, or one that shows density of trees. I’ve got a basic implementation working pretty well and it should be useful in a number of areas once I’ve worked out a few issues.

Since this post is lacking in pretty things, here’s a picture of a miniature village at the Chicago Field Museum.

In April, we’ll be bringing in another programmer to join me and Pal (yes, it’s just the two of us at the moment) and Petur is hopefully going to take a break from Rust to help out with art. I’ll be away for another two weeks, so the next update will be in April, when I’m back in London working full time.

Bye for now!

In-case you missed it, we released our first trailer earlier this month! Me and Jenny flew out to Las Vegas to introduce Before at The Game Awards.

Making a trailer wasn’t something that we had planned, so when Geoff Keighley asked if we’d like to show something at the awards we had to switch gear and work long hours through November to get it together. I’m happy with how it turned out, and really thankful to everyone who helped make it (you know who you are!).

It’s a cinematic trailer, all captured in the Unity editor. The idea was to paint a picture of the world, the atmosphere and tone of the game; giving people a glimpse of the kind of things that will be happening when you play.

We won’t be doing another cinematic trailer any time soon, next time it’ll be gameplay!

Music

One of the most exciting things about releasing the trailer was finally letting people hear the main theme music that Jenny has composed. For me, the theme conveys perfectly the tone of the game. It’s beautiful, but also raw, primal and unnerving in parts. It also serves as a starting point for one of features I’m most excited about.

Music is going to work in two ways: diegetic and non-diegetic, or rather “inside” the game world, and “outside” of it.

We’ll be dynamically arranging music as you play, based on different factors and events. So you’ll hear different lines and motifs play at key moments during gameplay, for example a change in the weather or at the violent conclusion of a hunt.

You’ll also see and hear people playing drums, chanting and singing. Eventually we want the music created by your tribe to reflect their characteristics, cultures and even history.

The voices you can hear in the main theme are that of Jenny and her father Phil Minton, who has also done some voice recording for the cavemen. We’ll be recording with more singers and vocalists in the future, and will put together a blog about that process and share some previews of how things are sounding in a few months time.

Progress

Making the trailer took up most of our time since the last blog, but it pushed us to implement some new features and improve others along the way. The basics of our Biome system are working nicely, which means Biome specific seasonal changes and weather are going in too.

We also identified a few slowdowns in our scene production pipeline, and I’m writing tools to make it quicker for us to put together new areas for gameplay. The game world is made up of islands of various sizes that will be randomly spawned and positioned when you start the game.

We’re waiting on some improvements to the navigation systems in Unity that will make this possible (you can’t move a navigation mesh at runtime right now), so for now I’ll keep working on the tools we need.

Pal has been crunching away in the background, moving towards another big milestone with our AI. We’ll do a more detailed blog about these systems when they’re implemented.

Questions

Lots of questions have popped up in the various comments and tweets since we released the trailer, so I thought I’d answer the most common ones.

When will the game be released?
When it’s done.

Will there be an alpha/beta I can play?
Probably, but we’re not committed to early access yet.

Will the game be released on consoles?
Maybe. We’ll try it out when the time is right. Desktop first (PC, Mac and Linux).

How does the game play?
Mostly from a top-down camera, but with other modes available. You can click on people and tell them to do things. The core of the experience is the relationship between player agency and AI autonomy.

It’s not Rust without textures. It’s offline and single player (for now).

Finish Rust first!
That’s not a question, silly imaginary questioner! Different people are working on different games.

Holidays

Needless to say, it is the holiday season. So I will be lighting the fire, dragging the carcass of a dead animal back to my cave and feasting on the raw fleshy goodness.

Whatever you’re doing, have fun and we’ll see you in the new year!

Seasons & Weather

The passage of time, the ebb and flow of the world; these are some of the most important parts of our plan for Before. The relationship between the player, their tribe and the world is integral to how we want the game to play and feel. You will need to be mindful of the elements, reacting to changing weather conditions and challenges they will present for your tribe.

I’ve begun exploring how we might implement these systems and while it’s quite simple at this stage, it’s still nice to see the landscape start to change. We want each season to feel distinct, both in how the environment looks and the way in which people and animals behave.

This video shows the transition from Autumn to Winter and then into Spring. This process will normally take several months, but I’ve condensed it into roughly 48 hours of sped-up game time.

Diogo has written a new version of the terrain shader that I had been prototyping, and some great extension scripts to help manage how everything fits together. It’s still a little rough around the edges, but this all means that can I achieve the style I want and still make use of Unity’s native terrain features. We still have a few issues to solve in this area, but we’re nearly there.

AI & Simulation

Most importantly, Pål has continued his work on the core systems that govern agents in our world. We’ve reached a major milestone with the initial implementation of two of the three major systems here.

The final piece of the puzzle is the simulation of “needs” for each agent, which are the driving force behind most AI behavior. Once that system is implemented we can start to iterate more rapidly, designing behaviors and tweaking the values that will bring the world to life.

Land ahoy!

For the next blog entry we’ll have some video to show where things are headed. The next few weeks promise to be a real turning point in the development, and we’re really excited to see things come together.

There are comments now.

As Garry explained in his post, we have joined Facepunch to work on Before full time. When he first contacted me back in April, I was just starting to plan for a Kickstarter campaign, but Garry’s offer was too good to pass up. However, the team I wanted to bring with me weren’t all ready, so we agreed that myself and Petur would join Facepunch for a few months until everything was in place to get started on Before.

Working with me on Before now are Pål, a programmer whose skills I have trusted for nearly a decade and Petur, an exceptional artist who I’ve worked with on countless projects. Having built a prototype, documented ideas and explored the style of the game by myself, I’m really happy to be working with these guys, and having a wealth of talent alongside us at Facepunch is a huge boost as well.

We’ve had our hands full over the last few weeks porting the project to the Unity 5 beta, cleaning up assets, and tackling some of the technical and artistic challenges that come from the scaling up development. With the addition of new shaders, lighting improvements, atmospheric effects, Speedtree and more, things are looking good.

But of course, it’s what’s happening under the hood that really counts now, and turning those prototypes and ideas into something you’d actually want to play is no small task. Pål has been implementing the underlying structure of our game systems, the real foundation work that will allow us, to iterate quickly and focus on the fun stuff. This is the slow bit – but we’re starting to see results, and hopefully it won’t be long until we can share some of that with you.

Ah yes, you. I have to say “thank you” to everyone that has voiced their excitement, enthusiasm and support for the game, even in its early stages. If it wasn’t for your support we would not be in the position we are today. So yes, thank you!

Onwards, to prehistory!

Much like the blogs for Rust, Deuce, Space C***s, and Arcade, we will be posting regular updates on how things are going. You can also spy on our commit log via Twitter, where you can follow the new @BeforeGame account, too.

See you next time!

It’s been a busy few weeks since my last post… news of our plans to take the game to Kickstarter stirred up quite a lot of interest, which has has lead to some really exciting opportunities. It’s a little early for me to talk specifics, but I will say that as things stand we are no longer planning to run a Kickstarter campaign.

We’ll be working out the details and plotting a course for development over the next couple of months. I look forward to sharing more details about the future of the project later this summer. Between now and then, I’ll be posting updates as often as I can.

Lastly, I’d like to thank everyone who has followed, favorited, re-tweeted, liked, smoke-signalled and e-mailed for your support, kind words and excitable grammar.

Until next time!

– Bill

So… it’s time I explain what this project is all about and where things are headed. It has been a busy few months not working on the game, but that’ll be changing soon.

Until recently I had been working on Before by myself, in my spare time. However, I’m very happy to say that I will no longer be tackling this ambitious project alone, having enlisted the help of good friend (and vastly superior programmer) Päl Trefall.

Together we have been working towards a refined prototype of the game, based on my early experiments.

So, how will it play?

The game controls much like a real-time strategy or “god” game, though I think of the player as more collective conciousness than deity. You command members of your fledgling tribe to perform various actions and pursue goals; moving around the landscape, hunting animals and gathering resources.

The core of the experience is survival; guiding your tribe through their life in an often dangerous world. Ensuring that they find food, keep an eye out for dangerous predators and find a safe place to sleep each night.

As the tribe grows, so will their desire for somewhere to call home. Deciding where to settle down is the first step, before gathering wood for the fire and materials for building.

The game is open ended and somewhat of a sandbox. Should your tribe die out, the game is over. At present, we plan to have “collective permadeath” as our default setting.

Each creature in the world has some basic simulation which determines behaviour; eating when hungry and sleeping when tired. As time passes, creatures grow older and will eventually die of natural causes, though it is more likely they will be killed by a predator.

At this early stage the tribe’s AI is fairly basic; you are responsible for most of their behaviour, and that of the other creatures in the world is simple. A herd of elk may flee at the sight of approaching hunters, but a hungry pack of wolves may chance their luck in a fight.

What next?

Beyond the core theme of survival, the scope of the game design is quite broad. We want each player’s tribe to feel unique; the culture they develop and the mark they leave on their surroundings serving a visual history of the time spent playing.

You’ll be free to explore the world with your tribe; building boats to cross the seas, descending into underground caves, unearthing mysterious artefacts, even discovering other tribes. I’ll be talking more about these ideas future posts.

Our plan is to refine the prototype before launching a Kickstarter campaign around the end of April. To make this game a reality, I need the time to work and money to pay other people for theirs, Kickstarter seems like the best opportunity to move things forward, hopefully we’re not late to the party!

Thanks for reading!
– Bill

First off I will say that my assets are clearly simple, low poly models with little or no use of  smoothing groups. The majority of my meshes do not have diffuse maps, or even UVs, just plain colours. I am using the Unity Toon shader (the Lighted variant) for the majority of my materials, again with a few exceptions such as the skydome and clouds.

So onto the central topic, lighting and post processing, or Image Effects, as they are referred to in Unity.

My basic scene looks like this:

Tonemapping is the first script I add to my camera. There are a variety of options here, I’m trying to create a pastel-y, somewhat washed-out look with my image, so balance the settings until I’m happy with the result.

I prefer Global Fog  to the fog found in Render Settings, and it’s an important addition. Fog helps create a sense of scale and distance, adding hugely to the atmosphere.

Depth of Field comes next. I use a script to find my focal point when the game is running.

One of the most dramatic changes comes with Colour Correction. You could spend hours playing with curves and depth settings here, creating really drastic effects. I’m not 100% happy with the balance I have in this scene, which is often the case!

Last but not least, everyone’s favourite (and often mocked) post-processing technique: BLOOM! This helps the colours bleed around the image, creating that hazy feel.