The game's engine

[MR.Mic]

I already knew they were "just" rendering engines, that's an obvious answer. The point is to pick a library that does something, and does it well. And then combine it with other libraries that do other things well. (like physics, sound, and networking)

EDIT: That's why I also suggested Quake 3, it's an all-in-one package that contains everything to make a game. On top of that, there are other Open Source projects that have modified the game to support shaders and stencil shadows from which we can base our own code on. On the other hand, we could very well make a mod for an existing engine.

[goodkebab]

whichever is going to be the fastest and easiest for the programmers ...at least that is my opinion because it is going to be a lot of work.

[scope2005]

I agree kebab, if you use the Unreal 3 engine you already have proven to work netcode (in previous Unreal titles anyhow and considering UT2007 is primarilly multiplayer...), and a graphics engine that supports modern shader effects without having to program your own support.

Its a lot less work compared to the task you would have to make this game from an open source engine, meaning a working beta would be possible after about a year of work.. and perhaps an internal alpha after a few month's if your lucky.

[frvge]

How would we acquire the UE3 engine... ? I thought licensing was in the millions of dollars. Or at least hundred thousands

[Overstatement]

We can use any engine as long as it is written in C++ (sorry Mic) and you don't need to have owned the game to play it (but if we are going to do a mod, might as well be Doom3. it has cool lighting and 4 player multiplayer too). Preferability, it should be open source because CT is a unique game not the typcial FPS that most engines CAN run.

And OGRE and Irrlicht are both "proven" engines with lots of games made with them. Probably a lot more in quantity than professional engines because of their availability. And lets admit it, professionals screw up, that is why most of us are here instead of playing DA, right?

[Aj]

I would still vote for Unreal 2/3 seeing as it's the engine that SC uses.  I don't know if they will do it immediately, but for Unreal Engine 2 there is a non-com runtime for people to use to create things.  They might do that with Unreal 3 so we *could* use that if it is available and build on that.  Or just do a mod.

[goodkebab]

here is something new ....for free from Microsoft.

please, someone qualified to judge engines....look into it:

http://msdn.microsoft.com/directx/XNA/default.aspx

[frvge]

"Right now XNA Game Studio Express is only designed and tested for Windows XP SP2. Windows Vista and Windows Server 2003 are unsupported platforms, although we are working on bringing Vista support in a future update."

"The initial release of the XNA Framework on the Xbox 360 does not have any support for networking. We realize this is a big area of interest for game developers and are actively working on a solution for the next release."

These 2 bits seem to be to be the largest drawbacks for our project.

Aj said he could program a bit of C# and XNA. I think with a bit of practice, I should be able to do the same.

[goodkebab]

^^ I am assuming that all of us would be using XP sp2 to develop the game.  The gamer should not be limited to this....or am I wrong?  If it is a limiting factor to the gamer....how significant is this in comparison to using a UE 3.0 mod?

We wont concern ourselves with the 360 right now anyways.

[frvge]

I dont know if the games themselves are playable on Vista.

[Overstatement]

Ok, been looking at this for a while which is very hard because the online copy of it on MSDN currently have non-working scripts.

Short answer: XNA is a stupid name (stop it with the recursive acronym already. They aren't intersting, they aren't cool and they aren't here to stay), too low-level, no one knows C#, C# has less game libraries we could use and historically, Microsoft has already been poor at design and implementation (bugs) on the early versions of programming stuff (ok, I don't have any proof of this. It's just my impression after reading the horror stories of early DirectX).

Long answer:
For those who don't know. Direct3D is seperated into two parts. The Direct3D core is that part that has daily lunches and good working relationship with your graphics card (this is low level stuff). The Direct3D extention is just code  of stuff commonly used (like math and models) so programmers don't have to keep "reinventing the wheel". By extention, game engine pretty much do the same thing at a much higher level. XNA doesn't seem to be a "game engine". I would probably call it a D3DX library for C#. So far, only new thing I've seen in XNA is a lighting model based off of a callback effect system for customization (a plugin system with basic stuff as default). Which is not terriblly hard thing to code yourself. I suspect it only basic light because I've seen nothing to indicate any kind of mapping, it doesn't pretend to be an engine, it expects you to replace the default with your own (with all those cool lighting features all the young girls are screaming about). Most importantly, it doesn't include any shadow casting, HDR, memory management or space partitioning structures. There is also no physics engine, anyone know of any popular free C# physics engines out there that isn't just written in C++ and binded to C#? I could be compeletly wrong about this. That section of MSDN is very slow and buggy so I couldn't get a good look of it. But if anyone can find those things I see are missing, post them with a link here.

Also, in my experince, pretty much all games have slow-motions problems with pre-SP2 and using a Microsoft product completely destroys the hope of rewirting this to cross-platform.

Edit: Really short answer: It's a framework, not an engine.

[goodkebab]

Thanx Overstatement,  we need more of these kinds of critical analysis on engines from programmers.  At the moment it seems our biggest concern with UE 3  is that the players will have to buy UT 2007 first.

We do have the advantage of a very solid engine that the SC series uses anyways.  So look and feel should be very similar....leaving us with the question of how much do the programmers have access to the SDK in mod form.

Just because it comes out in 7 months, should not stop graphics production -modelling and animation need to be done before the coders can start anyways.

CT and DA players are in a decline now....but i dont doubt we can get a lot of them back if we can remain loyal to CT gameplay as much as possible.

[Overstatement]

Just because it comes out in 7 months

That is one huge assumption. I bet it will get pushed back to September. Apparently, they don't teach you how to estimate development time in school. Such a humungo risk for something we have no information about, awesome. But hey, what do I know about engines?

We do have the advantage of a very solid engine that the SC series uses anyways.  So look and feel should be very similar

See, you can't ever say that! They had the complete source code, that may not mean much from the outside looking in but it's a HUGE difference. And I don't know what you were playing but CT feels nothing like UTanything. I only recall vague memories of playing a UT demo but it was pretty clear how unCT-like game it was (god, I hope I'm not digging myself a hole with that).