The Goal State Kickstarter is Funded! Stretch goal season begins!

We raise £25K in 3 days, and now it's time to build on it some more!

Goal State went live on Kickstarter on November 15th, with the goal to raise over £25K to fund a new avenue of content and material from

AI and Games

. I’ve talked about it a lot in the past few months, but here is everything you need to know, and how to find out even more about it. I’m pleased to say first and foremost, that it is now funded, and we have plenty of stretch goals to hit in the coming weeks!

Key Facts and Links

• Where’s the Kickstarter at?

  • Linked to here: https://www.kickstarter.com/projects/aiandgames/goalstate

• When does it go live?

  • 3pm GMT on Friday 15th November.

• How long does it run?

  • 30 days. We close funding on 15th December 2024.

• What’s the funding target?

  • To cover the first phase of the work - lectures for the Game AI 101 course - it’s £25,000.

  • This funding target is now met, and we’re in stretch goal season! Scroll down for more info!

• What is Game AI 101?

  • It’s the first of several planned courses under the Goal State banner. It covers foundational knowledge in AI for games. You can find out more about it in the dedicated article here on Substack.
    

    Goal State's 1st Course: Game AI 101

• How can I support the Kickstarter?

  • You can visit the Kickstarter page for more information. But, I did provide a full breakdown of how the funding will work in a recent blog here on the AI and Games website.

    How to Support Goal State

---

What is Goal State?

After 10 years of making videos on the AI and Games YouTube channel, and almost one full year of writing the AI and Games Newsletter on Substack, I want to start building high-quality, deep dive educational materials on artificial intelligence for game development. Goal State is the brand under which all of my work will operate.

---

---

Why is it Called ‘Goal State’?

In classic AI literature, we refer to artificially intelligent systems as ‘agents’. And typically when we use AI to find solutions to problems, we feed it what is known as the ‘initial state’ (the current configuration of the problem), and the task is for it to reach one of the possible goal states that exist.

So if you think about pathfinding in a game, getting a character to reach its destination by manoeuvring around the map, would mean it has reached its goal state.

So Goal State is a pun of sorts, in that it’s all about satisfying our ambitions. I felt that would be a fitting name for a series of online courses designed to help people enter the world of AI for games.

---

Stretch Goals

At the time of writing, the Kickstarter has now passed its original funding target, and that means we now have stretch goals to reach! If you’re not familiar with the idea: the more funding the Kickstarter attains, the more content we will add to it provided it hits key milestones.

Part 1: Game AI 101 Stretch Goals

The funding goal allowed for us to build all of the theory and non-technical homework assignments for the Game AI 101 course. But of course, the thing I want to do next is ensure that people can get their hands dirty in the code and start building stuff. But also people will want to explore it in game engines of their choice. As such, the technical component is broken up into stretch goals for each game engine that we wish to support.

But also, one little flourish that I thought would be great fun to add, is that we should take the time to dig even further into some classic games, and explore how they work. So in conjunction with the technical stretch goals, we also have source code deep dives, where we look at some of the most popular games of all time, and how the AI was built within them.

#1: Game AI 101 - Unity Tutorials

£31,000 (GBP) / €37,000 (EUR) / $40,000 (USD)

First things first, let’s get you putting some of that knowledge to practice. The Unity tutorial series will be spread across the Game AI 101 course. Showing you how to develop core algorithms in C#, then deploying them within the Unity engine. We will do a deep dive into how to work with tools such as Unity’s navigation mesh system, and how to build your own Game AI systems in their ecosystem.

#2: Source Code Deep Dive #1 - DOOM

£34,000 (GBP) / €40,700 (EUR) / $43,000 (USD)

A fun project that I think would be great for newcomers to Game AI, let’s take what I do on the AI and Games channel and turn it into a full-fledged analysis. The Source Code Deep Dive series will be a complete deconstruction of the AI in popular games, and how it relates to everything you’ve seen so far.

I’ve covered it on AI and Games but on a very high level, and so let’s begin our source code dives with one of the most influential games of all title: id Software’s 1993 classic DOOM.

#3: Game AI 101 - Unreal Tutorials

£40,000 (GBP) / €47,900 (EUR) / $50,500 (USD)

Expanding on our Unity engine tutorials, let’s migrate over to Unreal Engine. Epic Games’ popular suite of tools are richer in some capacities when compared to Unity’s. So we will be able to explore their navigation systems, take a deeper look into their Behaviour Tree tools, and how to integrate a lot of game AI algorithms using C++.

#4: Source Code Deep Dive #3 - Quake III Arena

£43,000 (GBP) / €51,500 (EUR) / $54,300 (USD)

The source code deep dives continue, and while there was many a popular first-person shooter of the 1990s that inspired the development of Game AI, I would argue the second most important after DOOM is none other than id Software’s other big 90s title: Quake III Arena. We’ll look not just at the underlying AI behaviours, but critically the navigation systems they built that to this day continue to influence game AI development.

#5: Game AI 101 - Godot Tutorials

£49,000 (GBP) / €58,600 (EUR) / $62,000 (USD)

Our final round of game engine tutorials for the Game AI 101 class is in the increasingly popular Godot. We will reproduce the same content covered in the Unity tutorials - or as closely as we can - using the Godot engine.

#6: Source Code Deep Dive #3 - Half-Life

£52,000 (GBP) / €62,250 (EUR) / $65,700 (USD)

Moving away from id software’s popular shooter franchises, there is of course the other bit first-person shooter of the 1990s: Valve’s Half-Life.

As explained in an episode of AI 101 on AI and Games Half-Life expands on the finite state machine architecture of DOOM to make it goal-driven. But what does that mean? How do you get that to work? We’ll explain how it all works.

#7: Source Code Deep Dive #4 - F.E.A.R.

£55,000 (GBP) / €65,800 (EUR) / $69,500 (USD)

We close out our source code deep dives with arguably the most critically acclaimed Game AI of all time: the enemy soldiers in F.E.A.R. by Monolith.

I’ve explained it in a high-level format again on AI and Games, but this time around we’re going to go into much more detail on how F.E.A.R. uses so many Game AI tools at once, all in an effort to create an intense combat scenario.

---

Part 2: Goal State Course #2 - ML for Games 101

#8: NEW COURSE - Machine Learning for Games 101

£70,000 (GBP) / €83,400 (EUR) / $88,500 (USD)

Should the funding continue to be successful, then our next stretch goal will be significantly farther off, but critically it will introduce my second course: Machine Learning for Games 101.

In our first course, we make little mention of machine learning, given it typically is not a practical approach for game AI. However, machine learning and in-turn deep learning have numerous practical applications in the games industry, both on gameplay side, as well as in production. So it is important that we then dig into this some more and cover this in more detail.

#9: ML for Games 101 - Python Tutorials

£75,000 (GBP) / €90,000 (EUR) / $98,000 (USD)

The first stretch goal for ML for Games 101 is showing how to build many of the ideas we explore in the course detached from a game engine. Largely because when we work in production we are building our ML codebase outside of game engines. So we’ll showcase how to use this in one of the popular ML development environments

#10: ML for Games 101 - Unity Tutorials

£80,000 (GBP) / €95,750 (EUR) / $101,000 (USD)

Unity's game engine enables for ML integration in a number of ways, but critically with the ML-Agents toolkit and the computer vision libraries. We will dig into how you can implement these in the game engine, build simple practical examples and understanding how it can be applied in different contexts.

#11: ML for Games 101 - Unreal Tutorials

£85,000 (GBP) / €101,700 (EUR) / $107,400 (USD)

Epic’s Unreal engine allows for machine learning integration courtesy of the learning agents library. In fact it’s how they build bots for Fortnite! So let’s dig into how that works and build your own ML-driven characters in the Unreal engine.

---

Part 3: Goal State Course #3 - Game AI 201

The final stretch goal, which frankly is here in the event that we reach a staggering level of success in crowdfunding is the sequel to the original class: Game AI 201. This would build on the first course, but also expand into new areas that we have not covered.

At this point I’m not going to get into too much detail on this, but in the event we start hitting that level of interest, then I will give a deeper dive into the structure and format of it all.

#12: NEW COURSE - Game AI 201

£110,000 (GBP) / €131,000 (EUR) / $98,000 (USD)

---

Thanks for reading, and a special thanks to all who support Goal State.