I've read time and time again that game developers, like any other artist, need to produce works for their portfolio so that they can be considered. In a lot of ways, this seems logical, and I understand it on the abstract levels of consideration. On a lower level though, I have had a dream for a game (who hasn't had one?) for a while now, and I'm finally in a position with enough free time and enough resources to possibly get it done. I hope for it to be a massive project in terms of features--but I was left thinking: what if that's all I end up with when I'm done? I guess what I'm asking is this: do I sink 2, 3, 4+ years into one (hopefully) stunning game as a show of talent, or should I build up to "the big one" so that I'm sure to have some truly finished products before I move onto the game I might not finish? Additionally, what about graphics/physics engines: for a portfolio piece should it be self-built all the way up, or could I base my work initially off of a graphics engine, say, Irrlicht, and eventually make the tweaks, etc. that I find necessary for my game? And the final point that complicates all of the above: the emotional factor. I've sort of presumed that I'm not going to get hired (as a game developer at least) right out of college--or at least, I'm assuming the worst. So I would *rather* work on "the big one" -- I've got plenty of littler projects I *could* do if I wanted to spice up a portfolio, but... I guess its just not where my heart is right now. So... anyone have any insight or advice? Thanks in advance.

Define stunning and big? Are you intending to make something the size of Call of Duty or Sims 2 or some other big Triple A title? The days of one person working on a project that can compete are long gone. You wont be able to make a game that is as big as a current triple A title. If you do make it as big you simply wont be able to produce the assets to fill it at the same level of quality as a triple A title. Given that this is the case how will it be stunning? What are you measuring it against when you say that, a triple A title or a small shareware title.

If you're straight out of college go for the small ones. Make them great and build up your portfolio and get a job. Then you will be able to learn all the tricks of the trade that will turn you from someone with a dream into someone with the skills and experience to fulfil them.

Well, see, here's the deal. I'm not expecting the graphics to be triple A level, I'm expecting the concepts and the programming feats involved to be. I have no dream of this game ever competing, per se... I figure that perhaps I would be working on something that demanded some level of skill, but not necessarily a Hollywood masterpiece.

I guess, in a nutshell, its not that my game is really all that grand a task. The main "feature" of it, if you will, is I wanted a game that was TRULY seamless. As in all objects represented uniformly in one coordinate system (as opposed to most space/rpg/fps etc. that break things up into zones of some kind).

It's a task, I know. I just thought it would be cool. However, I also see your logic on working the portfolio aspect (in fact, I saw that logic slightly more powerfully myself which is why I came here asking for an excuse to work on the big one).

Anyway, I'm rambling on to myself now--thanks for the advice.

In that case I don't understand the point. Why spend 2 years working on something that wont compete? Is this just a pet project for your own amusement/improvement or is there a purpose?

Well, what if you spend two years on something and it fails. If you spend your time on smaller projects, less risk of failure, less risk of not making money. Also, it would be better to have a history of projects completed then one big one not completed. Also, what are you going for. Someone who is going for a programmer's job is going to want to have something different in there porfolio then someone who is a modeler. I'm not saying don't think big, i'm saying do not put the cart before the horse.

I think you should write a game document for the game you want to make and let it sit for a couple of years. Then once you have finished some projects take it out look it over. You may be able to add things in.

I think we just had a winner.

The point was to come up with something workably impressive that would prove that I wouldn't be too "risky" to hire--that I knew what I was doing. However, the completion part bothered me as well--I'm not sure someone who has spent most of his recent time on a game that isn't finished would look too good either.

Writing the game document is probably the way to go with that. I could still develop all my ideas--figure out what I want to do and how I'd do it and throw that in my portfolio too.

Actually screwing around with one of the smaller games I've been working on led one of my friends and I to beleive that it was more addicting than some of the newer AAA titles we had hanging around as well. There's something to be said for simplicity.

Anyway--I feel I have a direction to go in which I didn't feel like I had before. Thank you guys!

A tag-on question though, if I may: I've seen various other descriptions, but you guys seem pretty knowledgable (erm, and in positions where you could judge =P). From your standpoint, what does a "good" portfolio look like?

My feeling is, if you are at the point in your life where you think you can try and make something really great, you owe it to yourself to try. You may fail, but if you can take a risk, you should do what you are passionate about.

If you are not sure you can complete your goal, think about some piece of it that you could complete that would make a good demo. The renderer, the physics, the AI, the character models. Make sure that gets done well and in a form that can make an impressive demonstration. Then at least you have something to show for your time.

I don't understand your question about developer's portfolios. If it is just you, what you want to have is a resume and some sample work (which artists usually call a demo reel, programmers programming samples). If a professional game developer wants to find work with a publisher, that company is usually judged by the titles it has completed, or the experience of the staff, or the quality of the game they have to demonstrate.

For an entry job at a game company, no project that you complete will be as helpful as the right kind of educational experience. If you don't feel your college degree will open doors, maybe you can continue your education. A masters degeree in computer science from a good school, or an MFA, could help you and it is unlikely to be a waste of your time in any event.

A good portfolio would demonstrate your programming skills and depth/breadth of knowledge. Try to think of something relatively original but still simple enough to complete and really go to work on it.

A demo of some kind that the guy in the office can put in their CD-ROM and just start playing straightaway is most likely to get attention. Make sure it works first time, every time, and try to blow their socks off.

As a junior programmer with a company they won't expect you to be able to immediately write a graphics engine, a physics engine etc from scratch. They are likely to use a 3rd party engine especially for physics, or will already have their own engines for everything. Even a new game they'll use existing code so just showing you can program and use libraries would be a big factor. Another major thing is to show you REALLY want to program games, not just program. They'll expect you to be dedicated enough to work long hours, weekends etc for no pay - you need to convince them of this.
You also won't be responsible for completing a huge project on your own. Showing you can keep things manageable without being distracted and spending extra time is useful - they'll want to give you tasks which you can finish ASAP.