Hi,

All the articles I've read about putting together a portfolio all say that you should use it to demonstrate your particular
area of interest and for the job you're applying for. I've always been very interested in A.I. but when I think about putting a portfolio together I feel really overwhelmed by the sheer volume of sub topics the field contains.

I know from making my own games that starting out simple (I've worked my way through tic-tac-toe, pong, snake and I'm going to start on my own idea now) but if I was to put together some demos I think I'd be working on them for years to cover even a small percentage of topics (e.g. A* has multiple variations) that I read about in this forum, aigamedev, game/ai and so forth. I understand that as games increase in complexity so too does the need for A.I. to handle it but also that in certain cases the A.I. techniques used are specific to the game / genre.

I was wondering if someone here with vastly more experience than a novice like me could suggest a game that I could reproduce that would be a good test bed for implementing a few different techniques without getting too complex?

Thanks,

T

I have mixed feelings about the usefulness of demos as job-seeking tools for exactly this reason. If you present a demo that's not highly polished, it's not going to impress game developers (game developers who don't polish well don't last long in this industry). On the other hand, creating a highly polished demo is enormously time consuming. Thinking back, I don't know that I've ever personally seen somebody hired on the basis of their demo (at least, not as a programmer - artists are a very different story). Typically, for programmers, a lot more depends on the interview and the programming test than on any demo you might provide.

My recommendation would be to spend your time nailing down your knowledge of C++ (and to a lesser extent Python, Lua, and maybe C#), browse through the industry literature to broaden your knowledge of how games are built (good places to start are the Game Progamming Gems books, the AI Game Programming Wisdom books, and the aigamedev.com site), and spend enough time playing around in an actual engine (e.g. OGRE, Unreal) to have a feel for how things work.

A demo isn't essential, but a decent one can make or break your offer with an employer. For me it was my AI bot http://www.omni-bot.com/ that got me my job at Pandemic and helped a lot in getting my current job at id Software. The more experience you have the less difference a demo will make. For just getting into the industry though it's always better to have something. Bots for FPS shooters are not all that time consuming to do something interesting with, and it demonstrates something significant in an existing title.

Thank you both for your replies, I really appreciate them.

I guess working on a demo will help keep me focussed and help to guide my learning. Even if I don't get it polished enough for an interview I'll have learned a lot just by working on it. It is bound to be better to get stuck in than worrying about how and where to start.