@RPJK (if you still read this after so long): Sadly it's impossible to create "useful" source code out of a binary. It is possible to create a "source code file" from a binary that will compile, but that's not useful. All the variables will names like "awf2f" instead of "number_of_bullets_left". Despite common belief source code is for humans, not for computers.Oh dear, oh dear. How can people always lose source code in some way? Even the bigger projects only are 1MB of source code in total.If I ever get my hands on a lot of free time I am willing to make "RA3". Before you start yelling "NOT GOING TO HAPPEN" and "YOU HAVE NO IDEA WHAT YOU ARE TALKING ABOUT", yes I do. I can't handle it alone, but if we can get a team together (maybe with http://gamedev.net/) we can do it.Just my 2 (euro)cents.
EFFEs still around?
If you are willing to join me (a C, C++, Python and Lua programmer) in creating a new game engine (and physics engine, unless we do some tests and Bullet fills our needs) for 'ra3' then please contact me. I already have another developer that is willing to work on this (eFFeeMMe, a Python programmer), but us two aren't enough.
Surely modifying an existing game engine (such as Valve's Source engine or the Unreal Engine) would be easier and better than starting from scratch? Bear in mind that most game engines are derived from the Quake engine.
Couldn't you at least try creating something with nig-rig soft body physics using rigid body physics? I mean, it'd be best just to get a robot combat game out there so you can build up enough steam to get one with a proper soft-body physics engine done.
There are a few problems with that, in my opinion:most of the engines have a very unflexible API, with concepts that are not needed in a RA-like game, yet have to be used in order to use the engine to its full potential This is true when it comes to the Unreal Engine, but bear in mind that the source engine was designed to have new features added and for it to modified - HL2 Cinematic mod springs to mind....which implies that a lot of custom boilerplate code would have to be written anyway, and, in my opinion, starting from scratch is as good or even better (as you develop the whole architecture from the ground up, and can make everything tailored to your specific needs) A very good point. However a counter argument could say that it would take a lot of time building a completely new engine from scratch. What I suggest is build a new engine but take some elements from existing engines.I know the obvious candidates (Source, Unreal, ...) don't have physics engines that I'd like to have in the game. They're boring rigid-body engines, and using only these would mean creating something awfully similar to RA2. There is one exception into which I will look, and that is incorporating the Bullet physics engine into any sort of current game engines, as it offers some soft-body physics, and is open source. I'm not entirely certain what you mean by soft-body physics. I understand you thinking that the Havok engine could be classed as boring, but it is realistic, and a certain degree of realism needs to be implemented into the game.personal taste, eg: Source being a bit old-ish and too tied with their internal physics engine, Unreal focusing on 'brown' FPS games... Completely true when it comes to Unreal. However, despite Source's age, it still goes on, and is capable of many different things, if you know what you're doing.
The first I'm not so sure of, there's a lot of choice here.The second I have a candidate for, I am very interested in Irrlicht: http://irrlicht.sourceforge.net/The third will probably be Python or even Lua. I'm working on a scripting language myself, but that's more of a personal project that I'm never going to finish.
Author
Topic: New Robot combat game- in progress. (Read 41131 times)