First of all, moderators: if this is in the wrong place, please, move it. While I don't consider myself "new" to this site anymore, I still don't know exactly where this should go.

Now to the good stuff. My team and I are making a 3D zombie apocalypse simulator MMO soon to be available for Mac and PC. I won't go into detail about this, as this isn't what I'm asking about. Anyway, my programmer is building a system for Unity 3D, kind of like an engine inside of an engine. This is what our game mechanics are based on. We've been working on 3D models, so we would have them ready when our system is complete. However, in my whole game development life (about a year), I've always used one rule: Focus on gameplay, then 3D models. So, in my case, would 3D modeling and then doing gameplay work? Does anybody else work like this? Thanks for any answers!
Dakota.

1. Our game is made inside of Unity 3D engine, but a system is being made to add to that.
2. Thanks.
3. If you knew how our game works, you would know why it needs to be an MMO. You can read my dev journal if you like.

I'm with redthegreen. First decide what the gameplay is. Then make the models needed for the gameplay. Making the models first might result in having to go back and change some and/or add some and/or delete some.

Thanks, guys, we pretty much have exactly how we want the gameplay, so is it safe to start the models?

so is it safe to start the models?

Yes, you can start with the models. It is certain that there are some zombies running around, that you will have a some street sign etc.

Just be aware, that if you start with art which is gameplay relevant (e.g. the layout of a level, some special zombies etc.) , then you will most likely need to adjust it later on. The danger of reworking art or of disposing it completely is higher if the gameplay is not fix yet.

Anyway, my programmer is building a system for Unity 3D, kind of like an engine inside of an engine.

This rings a few alarm bells.

There is a problem that a lot of programmers have, where they feel the need to work on such 'systems' and 'wrappers' and 'useful engine tools' and generally making the code 'perfect' rather than getting on with the task of actually making the game.

While there may be some value to some of this work, most of the time you end up in an endless cycle of refining the 'engine' stuff, refactoring it to make it nicer and nicer, and adding more and more awesome features, and you never actually write the game. Trust me, I've fallen into this trap myself before.

I would try to encourage your programmer to drop the 'engine in an engine' angle - it's pointless, you already have an engine, it's called Unity - and get on with writing some gameplay. It doesn't matter if the code is ugly, what matters is that you can get the game into a playable state as soon as possible. You can always refine the code later, once you know you've got a product worth finishing - and once you know at last what features you actually need.