Saturday 15 January 2011

At times, the goal is to represent artificial stupidity rather than intelligence.

Wow, so I have a huge apology to make, It really has been longer than a month since I last wrote! I'm appalled at my lack of commitment to my own blog, but I can honestly say it's been a roller-coaster of a ride this last month. So many things happening, so little time and I'm sure I'm not alone when I tell you how busy it does get over the Christmas period... But enough with my excuses, I'm back and I promise to work harder to keep my blog updated!

Some good news, the pitch/presentation went really well, we had Rob Yescombe, a developer from the local company 'Crytek' attend and even though I'd never met the guy and had no real previous knowledge of him or his colleague he brought with him, I honestly felt a little star struck. I admitedly started off a little shacky when beginning, but I managed to pull it back and the rest of it went very well. We hadn't been told previously but there was an award for the best sales pitch, and we won! So it was a much longed for pat-on-the-back if I'm honest and re-lit the fire again.

So I started back at college last Monday, and there was no time to talk about what we'd been up to because I'd gradually realised over the break how much work was actually left to do. It's obviously hard, other than taking an educated guess and applying some common-sense, to judge even approximately how long steps of the  development/production are going to take, so this is where ultimately, as the Project Lead, I start to worry a little, but I've managed to mitigate it slightly by being a little tougher on the team and setting some realistic timelines for getting certain things done. as much as I love the game design side of the project, it's time to put it to bed for a while and start creating some assets, creating the level and playing around with the Artificial Intelligence (AI). Luckily I have a very dedicated Artist/3D Model creator and I really do mean it when I say I'm lucky to have him. He will give us the necessary boost we need in order to stay on track as he's competent and knows what he's doing with using the software we're creating most of the assets in. Also, my confidence has been restored a little in the programming side of things given that the programmer has shown some real enthusiasm for this side of things, taking time out to research the necessary techniques and skills required to build the demo level.

Anyway, more to follow soon, will keep you updated on how the development of the project goes and any other bits that I think you guys might find useful :-)

Bye for know...

P.S the title for this note in the blog is taken directly from the UDK tutorials on implementing AI into the game  you (me, anyone) are trying to create. To paraphrase the example given - It's very easy to programme the AI to target the player and shoot and be very accurate. But this would not make for a very interesting game experience as the AI will always have the upper hand. So the key is to make the AI less like AI and more like a human player (e.g. will not always aim successfully and will miss every now and again.) Ergo, the goal is to create stupidity rather than intelligence. It's perhaps a rather blunt way of looking at it, but I quite liked it for its irony. This is effectively what makes AI challenging to create in video games.