At the forefront of Artificial Intelligence
  Home Articles Reviews Interviews JDK Glossary Features Discussion Search
Home » Interviews » Gaming » General

Mike Ward

Mike is a Certified Public Accountant who, prior to his involvement with RoboForge, was managing an active private investment company in New Zealand for four years. This involved the acquisition and sale of substantial property and business assets. Mike oversaw all aspects of the transactions including the sourcing of deals, their negotiation, due diligence, funding, legal, operational and financial reporting. Previously Mike was based in Hong Kong where he established and managed a financial services company.

G5: What inspired the development of RoboForge? What were the primary objectives in creating RoboForge?

The inspiration for RoboForge came from Darren Green, the founder of Liquid Edge Games. Darren was doing some contract work in Tokyo and was kicking back in his hotel room flipping through numerous channels on his TV, all in a language he didn't understand. One thing he could work out, though, a lot of people liked a game where cumbersome mechanical robots attempted to knock each other over. He thought the robots were kinda cool, but figured that they were extremely expensive and you probably needed a Dad with a PHD in robotics to build one. As a kid he always liked building stuff (model planes, Lego etc) and he thought that the idea of a computer game where you could actually build a robot and then let it fight it out with other robots would be amazing (and much cheaper than doing it for real). It took a few years of contemplation (read: procrastination) for the idea to gel enough to tell someone else about it. In 1997, Darren spoke to two of his programming friends (Hugh and Young) about the idea and received an enthusiastic response from them. The idea for RoboForge and Liquid Edge Games now became a reality. They spent a year designing the game in their spare time. Once the design specification was sorted, they all chipped in some money to fund the development for a year from Darren's basement. The money was used to develop a working prototype of the game (crude but functional). The prototype was then used to successfully attract funding.

The primary objective in creating RoboForge was to produce a new genre of game that allows the players to build their own gameplay and use the net to bring them all together in a forum that lets them flex their creative and technical muscle. Players are totally in charge of the construction and design of their bot and solely responsible for how it performs. It is a very much constructive and thinking game with most of the gameplay being focused on the building and testing of the robot. Players will get huge satisfaction and a sense of ownership from creating, nurturing and training a champion contender. Because the robot is autonomous we can organise huge tournaments and run them overnight. If everyone had to be there to control their robots, it could take months for us to run a big tournament. Also, to play for money, everything has to be fair. Connection speeds greatly affect how a player performs in online games.

G5: Why did you chose Java as the development language? What are your ideas on Java as an AI language?

Most people associate Java with small applets that run in browsers. Java is actually much more powerful than common perception. Large advancements in speed have been made in the last few years. We chose Java as it provided the best integration of 2D and 3D (RoboForge needs an unusually high level of 2D GUI integration due to the construction nature of the game). Java also has multi-platform support and is at home on the net, which is important given the Internet-based nature of RoboForge.

G5: What are your tips on building robots?

Tournament winners are typically those designers that have thought outside the square. Dare to be different and fate will smile upon you. Because of the diversity RoboForge allows there really aren’t any dos and don’ts we can share with you. Maybe over time we’ll start to see a pattern of designs that have advantages in certain areas, but the good thing about the game is before too long you are always going to find someone who will counter you next great design or attack move. Having the dominant bot is the ultimate, but usually it doesn’t last for long.

Do your research, watch old fights from previous tournaments. Jump on the challenge rooms and meet other bot creators. Absorb what everyone else has done and evolve it into something they are not expecting. Winning tournaments is not only about bot onstruction; it's about guessing what everyone else is going to build and countering it.

G5: What are your tips on programming the AI for the robots?

Start with a simple, functional AI, become familiar with how it works and then build your expertise from there. Don’t expect to master it straight away. Spend time in the RoboForge forums and learn what you can from the guys who have been before. There’s a lot of skilled designers who are happy to share their highs and lows in mastering the AI. Test and test again also helps.

G5: The RoboForge community has come up with some incredibly original designs. What are some of the more impressive robots/AI behaviours you have seen during RoboForge's lifetime?

We have been really impressed with the designs the players are coming up with. They are always changing and what’s hot one week usually wont stay hot for too long as players counter top designs. There are some really smart guys out there. Given the diversity and number of different designs I honestly can’t single any out. Watching the featured fights after each tournament is run is a great way to experience the diversity.

G5: What future expansions would you like to add to RoboForge?

We’ve working on some new robot races, weapons, arenas and parts. We’re also looking at user defined textures and some special weapons. We’d also like to produce a similar game but allowing the player to control their robot in battle.

Submitted: 13/11/2001

 Article Toolbar
BibTeX entry


Latest News
- The Latest (03/04/2012)
- Generation5 10-year Anniversary (03/09/2008)
- New Generation5 Design! (09/04/2007)
- Happy New Year 2007 (02/01/2007)
- Where has Generation5 Gone?! (04/11/2005)

What's New?
- Back-propagation using the Generation5 JDK (07/04/2008)
- Hough Transforms (02/01/2008)
- Kohonen-based Image Analysis using the Generation5 JDK (11/12/2007)
- Modelling Bacterium using the JDK (19/03/2007)
- Modelling Bacterium using the JDK (19/03/2007)

All content copyright © 1998-2007, Generation5 unless otherwise noted.
- Privacy Policy - Legal - Terms of Use -