What game engine for an FYP

Connavar

Hi,

I am going into my final year in Multimedia and Computer Game Development in UL.
I am planning on doing a Goldeneye esque game as my FYP.
Is there anyone here with experience of any game engines that could recommend one to me.
The engines that I am currently looking at are Source, Gamebryo Lightspeed and Unreal Engine. I will probably be using open source models and audio

Thanks in advance

NeverSayDie

Connavar wrote: »

Hi,

I am going into my final year in Multimedia and Computer Game Development in UL.
I am planning on doing a Goldeneye esque game as my FYP.
Is there anyone here with experience of any game engines that could recommend one to me.
The engines that I am currently looking at are Source, Gamebryo Lightspeed and Unreal Engine. I will probably be using open source models and audio

Thanks in advance

Well, Source and Gamebryo aren't available in free/hobbyist versions AFAIK, so they're not options (though you could do a mod for an existing game built with them I guess). Same with Unreal up to recently, but Epic now have the UDK, which makes it an option;
http://www.udk.com/

A popular FOSS engine for hobby/academic projects would be Ogre3D, it's been going a while;
http://www.ogre3d.org/

Unity3D is another one I've been hearing a bit about lately, appears to have a lot of cross-platform support;
http://unity3d.com/unity/

RedXIV

Gamebryo is available for education licenses I believe but from what i've heard of it, its a steep learning curve.

I've also been hearing alot about Unity and it's appearing to have a very strong indie scene at the moment which is good.

Also, the Torque engine from gamesgarage isn't too bad either.

Connavar

NeverSayDie wrote: »

Well, Source and Gamebryo aren't available in free/hobbyist versions AFAIK, so they're not options (though you could do a mod for an existing game built with them I guess). Same with Unreal up to recently, but Epic now have the UDK, which makes it an option;
http://www.udk.com/

A popular FOSS engine for hobby/academic projects would be Ogre3D, it's been going a while;
http://www.ogre3d.org/

Unity3D is another one I've been hearing a bit about lately, appears to have a lot of cross-platform support;
http://unity3d.com/unity/

I have access to both source and gamebryo engines so they are still in the running. I will have a look into those other four

Anima

I think Unity would be a good choice. Source can be buggy in places, plus sometimes their updates break the engine and you need to find a patch. I wouldn't risk it for something like a final year project.

Lord Panic

Found Gamebryo to be a pretty steep learning curve... Ogre3D's good, fairly easy to get started with, excellent support community too, I'd recommend it!

Edit: Though Ogre3D's really only a rendering engine with animation integrated, if you want a fully fledged "game engine" it might not be for you.

DesperateDan

Hi we had a second year project using the unreal ed, and I have to say its really nice to get to grips with, just make sure you have a relatively powerful machine and a legitimate copy, as the rip I had crashed, often. Unreal tournament 3 comes with the editor and has a really incredible set of free video tutorials provided by 3dbuzz.com that are the best I've ever seen. I think the unreal ed only allows fps development really.
But it depends on the focus of the game you're making, is level design a big factor? Or Ai? or Animation and models? etc.

Connavar

Lord Panic wrote: »

Edit: Though Ogre3D's really only a rendering engine with animation integrated, if you want a fully fledged "game engine" it might not be for you.

Was hoping to make the whole game alright

DesperateDan wrote: »

But it depends on the focus of the game you're making, is level design a big factor? Or Ai? or Animation and models? etc.

I havent decided weather to make it single or multiplayer yet. If its multiplayer, then level design will be my main focus.
If I go single player AI will be the focus.

I am beginning to play with the idea of a 2d game as I heard that there would be enough work in it. Really wish I had put more thought into this earlier

satchmo

I've heard lots of good things about Unity too, it's worth checking out.

If you're considering multiplayer, make sure whatever engine you choose has the required networking support - adding multiplayer from the ground up would probably be much more work than adding AI for single player. Even if the engine does have networking support, multiplayer would probably be just as much work as implementing simple AI would be, so it all depends what you like working on and what you want to show off on your CV after you're finished.

pwd

You should look at xna. I think it's oriented towards this sort of thing. (2d rather than 3d though, I think)

noxqs

Some of these already posted are very complex. Unless you understand scene graphing, rendering pipeline, input handling etc.. Making a game is not that easy.

Youre better off working ontop of a working implementation such as Source/Quake/Unreal or other games which provides modkits which lets you plug in new textures, objects and scripting.

Source is your best bet in my opinion but there are open source FPS games you can modify as well but they dont come with such a friendly editor, and unless you are focussing on the computer science part of the project - just getting to know any one of these projects will be overwhelming.

With that said. I have worked with Panda3D and found it quite nice to work with and there is a project which is similar to your goal you can peek at here: http://a3p.sourceforge.net/ built with Panda3D.

RedXIV

Just to note, if I were you, i'd concentrate on a single player experience with AI as opposed to multiplayer. Even with an engine built to support multiplayer, its tricky enough to get it even resembling what you know as multiplayer. At least with AI you can get the bad guys to chase and attach at you with about 4 lines of code.

Connavar

pwd wrote: »

You should look at xna. I think it's oriented towards this sort of thing. (2d rather than 3d though, I think)

I have used XNA before alright. Its not a game engine though. You have to do everything yourself with it

noxqs wrote: »

Some of these already posted are very complex. Unless you understand scene graphing, rendering pipeline, input handling etc.. Making a game is not that easy.

Youre better off working ontop of a working implementation such as Source/Quake/Unreal or other games which provides modkits which lets you plug in new textures, objects and scripting.

Source is your best bet in my opinion but there are open source FPS games you can modify as well but they dont come with such a friendly editor, and unless you are focussing on the computer science part of the project - just getting to know any one of these projects will be overwhelming.

With that said. I have worked with Panda3D and found it quite nice to work with and there is a project which is similar to your goal you can peek at here: http://a3p.sourceforge.net/ built with Panda3D.

I know it won't be an easy task. I am just looking into it as one of the lecturers put it down as something that we could possibly do

Connavar

RedXIV wrote: »

Just to note, if I were you, i'd concentrate on a single player experience with AI as opposed to multiplayer. Even with an engine built to support multiplayer, its tricky enough to get it even resembling what you know as multiplayer. At least with AI you can get the bad guys to chase and attach at you with about 4 lines of code.

Thats true alright. I will discuss this aspect with my supervisor once I have one picked

Gybe

This could so easily be done with the udk, most of the work would just involve getting the art assets done. The only actual code needed, that I can think of, would be for the weapon & round behaviour. Everything else can be done through the editor and kismet. The only real issue I could see would be presenting what actual work you have done to make this game as the udk takes care of alot. Either way you should definitely check out the eat3d tutorials on the udk, they are immense.


From my own experience of making a game for my FYP (this was for a normal comp sci degree!) support was everything. For that reason alone I would look for an engine with the biggest community around it as it's always the unforeseen problems with the tech that causes the biggest headaches! I've played with both ogre3d and irrlicht and they're pretty simple to get up and running. It wouldn't take much code to get the game up and running.

That said, as it is a game dev degree it may be more beneficial to just pick an API and start from the ground up. This is just looking at it from a college pov, I know the argument would be why re-invent the wheel etc, but this does demonstrate good fundamentals.