Week 5 – Idea Extrapolation, Unreal Engine and AI

This week’s team focused goal was divergence, allowing our ideas to flow and spread out from our original concept. In the previous week’s critique, Ross, Jane and Mitchell challenged us to further explore not only the concept of a virus that you are trying to spread to infect an entire town, but also some of the ramifications of such a scenario. To help illustrate this they also suggested some stimulus in the form of the book Guns, Germs and Steel by Jared Diamond. Bob and I had a look at this and found that the 3rd chapter, particularly relating to the spread of disease through colonialism could provide an insight and potential backstory to this idea. This greatly with developing one of my extrapolated ideas for the team, which I will cover soon.

Other work I did this week mainly revolved around installing Unreal Engine and having a playaround, not only with the creator but also with the code view to see how C++ code makes the engine itself tick, what we need to extend and what the engine already does for us. As we have 3 different programmers in our team, it grants us the opportunity to specialise slightly and one aspect of the development I have taken a particular interest in is AI, with special regard to machine learning and how feasible it is to use it to power NPC behaviour (wrapped in AI structs to control anomalous behaviour, of course). The two main AI methods that were looked into were Artificial Neural Networks (feedforward with backpropagation) and genetic algorithms, which are actually not as difficult to implement as I had imagined when I followed a nifty little introductory tutorial that can be found here.

Finally, as part of the team we have agreed to present 3 rather different ideas after brainstorming and developing our original ideas individually and picking the best ones out. Of the three that were selected, I am proud to say that my idea #2 made it to the final cut; however I will also precede this with my first idea as it is a much closer offshoot to our original idea of spreading the virus, exploring the emotional and moral consequences rather than only killing off a village with no discrimination.

Idea 1: Extrapolation of I Am Zer0 – First person vs third person
One player controls Patient Zero as per the original idea, whereas the other player is placed into the map from a 1st person view; it it their job to try and identify Patient Zero and cure the village. The 1st person view and the possibility of allowing the 1st person player to interact with NPC villagers and possibly receive dialogue has the goal of giving the 1st person player a personal view and connection/motivation to saving the village, where as the 3rd person player controlling Patient Zero, with no interaction with NPCs, can assume the role of a non-discriminatory, emotionless virus and its Patient Zero.

Idea 2: “Watch Your Back”  – selected for this week’s presentation
Players form two teams in an arena style map. Random players have demonic entities possessing them; these entities will kill the player they are possessing if another player successfully views their back; however, the the entity will then transfer to the viewer, making them vulnerable. The last team with surviving players wins the game, so players in a team may strategically sacrifice themselves in order to win. Obstacles in the playing field will be creatively placed or designed to allow players to obscure their back by scaling walls; with risks such as movable walls or obstacles allowing a player’s back to be seen from certain vantage points adding more risk and reward to gameplay, hopefully enabling a sense of flow when a player skillfully traverses the map hiding their back successfully, whilst killing other players by viewing their backs. Also considering a free for all mode where entities don’t transfer, however every player has one.

Before idea #2 it was accepted for the presentation I worked with the team to make it more feasible, in ways such as scrapping the “entity transfer” idea when you kill another player and simply having all player characters start with an entity, in a free-for all style base mode. Other suggestions were having non-entity-bearing NPCs walking around, posing as an oblivious environmental hazard, and also requiring the player to always be moving in order to prevent camping scenarios.

Well, that concludes quite a long blog post for this week’s work! Our team had some trouble accepting that we needed to extrapolate on what we already thought was a good idea, so quite honestly this was quite a challenge (given the team’s wont to Scope It Down). In the end however, I’m glad we could agree on some good offshoots to our original idea, and hope that we get the opportunity to select one and finally develop it into something amazing!


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s