There are a few solutions for you that fit within that budget range. If you want an industry-standard commercially available engine you're looking in the range of $50,000+ depending on the licensing agreement you have (and remember that this is just the license -- it doesn't include pay for your other developers).

Otherwise, you're best bet is to use something like the Torque engine which I beileve has a $100 indie license and then hire a programmer to do the rest. More importantly, you'll need to pay people to do artwork, sound and music, none of which is cheap. $5,000 just isn't going to be enough to develop such a game -- you'll need at the very leat $15,000 to be able to pay all of those people and even that is likely not going to be enough (might being for a couple of months for each but a few months isn't enough time to develop a commercially viable game). A programmer isn't the only developer you'll need.

I would plan on a one-year development cycle. This means that you'll need probably closer to $100,000 or more just to pay salaries. And this may be grossly underestimating it depending on what kind of quality you're looking for. Not to mention the costs of marketing and distribution. There's a very good reason so many indie games never make it to shelves because more often than not the 'brains' behind the game don't realize the astronomical costs involved in developing a modern game.

Quote:Original post by padlock
Torque 3D looks intesting but I'm not sure it could handle a large 3D map (of Italy) where you can zoom in on any point and still have high resolution terrain details.

I'm sure it can, it will just require some coding/scripting ;)

The problem is that you're looking for an all-in-one solution and such a thing really doesn't exist. Not one that's 'idiot' proof at least and as powerful as you seem to want it to be. If that's what you want than your best might be *cringe* game maker although I can't attest to how good as I've never used it.

The Torque engine requires that you're able to create your own content. It's possible to buy individual 3D meshes at lower prices versus hiring an artist to create everything completely new. It may even be possible to hire a freelance 3D artist to create models on a pay-per-model basis. This is a valid way of doing things but may not have the quality you're looking for.

As for a large map of a real terrain, if you're looking for real-world scale than you're out of luck once again. The costs in terms of resources are so high that a standard computer will more than likely not be able to handle such a thing. That, of course, depends on how everything is handled. If you're not looking for true-scale and instead something more realistic for current hardware that should be a very simple matter of using a height map which may be available from a stallite feed.

You've kind of hit upon the classic conundrum that is middle-ware, my friend: Either the solution is tailored to a specific genre -- encouraging you (if not locking you) into the vendors view of what that genre is -- or the solution is very general, and you must put a lot of effort into making it suit the genre at hand.

If we had a cheap, menu-driven application that could crank out games of all types with little effort, everyone would be making games -- but that tool simply doesn't exist. If you don't want to do any real programming (scripting aside) you're going to have to settle for a mod, or a genre-specific package like the previously mentioned Torque + RTS package.

It's not in the scope of this post to thoroughly cover hieghtfields but the basic idea is that you generate a mesh using an image composed of 255 levels of grey (e.g., black through white). Each color modifies the height (e.g., the y-axis) of a particular vertex in a height-field mesh.

I'm sure the Torque engine is quite capable of this as it's a straight-forward process and can be done in a few dozen lines of code with straight OpenGL (including initialization, etc.).

BASIC is an ancient language and, while great for beginners (hence 'B'asic 'A'll-purpose 'S'ymbolic 'I'nstruction 'C'ode) is really not viable for commercial games. I used to think it was back in the days of DOS and QBasic but that was a long time ago.

If you want to get into serious development you'll be looking at the industry standard, C or C++.

As Ravyne stated, to create a completely flexible SDK that's menu driven, scripted and not genre specific is like trying to create cold fusion. It's theoretically possible but to actually do it is an entirely different matter.

If your skill is not in programming you could develop other skills and either attempt to join another project or create a new one. Just know that if you do try to put together a team you're going to need to bring a lot more to the table than "game design". I myself am a programmer, team leader, teacher, graphic artist, web administrator and so on. I do a lot. But I also have other team members that are far better at concept art, for instance.

Anyway, the point is you're not going to find a tool for what you want to do for $5,000. I do wish you luck though as I know what it's like to want to do something but be unable to due to my own limitations.