Question: I'm a general Google Sketchup modeller (not a game developer). Why would I ever want to bother converting my 3D model to a file format a game engine can run, and then run it in this game engine ?
Answer: Simple, if you want to let your customers you sell your 3D models to walk around in the 3D model, and if that 3D model is fairly large, then you'll have a significant performance advantage when converting your model and running it in a game engine. If the 3D model is on the other hand small, then converting the model and using a game engine won't be worth the trouble and you best just use the "walk around" mode in Sketchup.
Question: This project revolves around making a first person/third person simulation game (see [Overview]). So why is there information to be found here regarding converting files in between game engines, and regarding making 3D towns and such ?
Answer: This project wants to be more than just a single project involved with just making a single game. We also want to speed up 3D development (pretty much any project that uses 3D to benefit real life urban development). We also focus not just on urban development which uses 3D for the sake of visualisation alone, but also specifically use it as a tool to help see which improvements can be done to the existing cities. For example, it can be used as a tool to make a quick proof-of-concept of integrating say green spaces (parcs, gardens, single trees) in a current street, making stormwater run to reservoirs, bioswales or going directly into the ground (by swapping macadam streets by gravel/dirt roads or pervious concrete roads). This would all help to drain or make better use of excess rainwater (and at the same time prevent the earth cities are standing on of drying out and thus come lower in relation to the sea), cool the city, ... It can also be used to show the possible conversion of existing fossil fuel power plants to run on a renewable fuel, making a farm to provide food directly to its nearest city, painting house rooftops white, improve the layout of electrical grids and water distribution piping in cities, quake-proof buildings, change the road infrastructure to make the roads safer and keep cars out of city centers and promote alternative modes of transport within the city centers, implement ring roads around cities and distributions centers in the outskirts of the city to transfer cargo from large trucks (coming from the inter-city roads) to smaller vans, cargo-bikes,... which can manouver better and are more environmental friendly in crowded cities. (1)
3D models of cities (of which the cities also exist in real life), can at the same time also be used for making serious games, such as simulation games (first person shooters, third person shooters, or even run games like Mirror's Edge), taking place in actual cities. Besides still being fun for gamers, it will then still perform a useful function as well (it will learn their way around in these cities, and such game projects will help accelerate the making of more existing cities).
1: Note that besides using using the game engine to show particular sections of the city and how it has been altered, other tools too may need to be used. For example, for any premilinary data that needs to be recorded to make the first analysis in cities on what particular roads need to be improved, we can use custom google maps using google my maps.
Also, to be able to know the right of way of streets for modelling them out in 3D, we can use waze map, available here. Knowing the right of way won't be necessary with most streets for the general planning won't be needed though, as city centers need to be made entirely car/large vehicle-free anyhow, so there will never be an issue with right of ways for the general planning; only for the 3D modelling of actual streets, waze maps can be handy, but then again you could also see this using Google Street View.
Also note that besides that this project can be of use for environmentally-conscious infrastructure changes, we also put up a few petitions which too might help in accelerating towns to become more ecological. These petitions can be found via following links:
350 campaign on sustainable road construction
350 campaign on promoting fragile forest land sales to protect them
350 campaign on providing power plant information to the public
350 campaign on civilian ITS's
Question: How did you come to select the game engines you did (Enemy Territory: Legacy, AssaultCube: Reloaded and Urho3D) ? The Enemy Territory engine is all ready quite old (2005), and the AssaultCube too is all ready fairly old (2008 respectively) ? Are such game engines still relevant today ?
Answer: They surely are. In fact, when comparing the system requirements vs the graphics offered, they can't be beat. Certainly not if you also want the game engines to be open-source, and specialized on displaying outdoor environments. In fact, even games made today (made with a tailor-made, new game engine) sometimes get a much, much worse game engine. Think of the immensely popular Second Life and its many recent off-shoots, and even the DarkPlaces engine -the latter being a Quake 2 engine and is still very popular for making new indie games-. Even the open-source Torque3D game engine, OGRE3D game engine, Quake 4 engine and the Tesseract engine don't offer the same favorable ratio between low system requirements and graphics. Sure, they're nicer, but the amount of computing power required to display them is far greater. Most people simply don't have such a computer with much graphic power, so having a game engine that has a very good ratio between system requirements and graphics is a much more preferred option. Also, even if you use a game engine that is more powerful/beautiful and you/your audience all have powerful computers, it isn't even certain you'll indeed be able to make advantage of this increased power/beauty of that game engine (as this depends of how well you are able to make your 3D maps, ...). If you are a bad mapper, you might end up making a map that is not well lit, has gaps, ... and is as such much more ugly than a well made map, shown in a slightly less beautiful game engine. To keep development of the game engine ongoing, it was also important that we selected open-source game engines. Open-source game engines can and are being continuously updated by the community and as can as such remain in use for a very long time (just look at the DarkPlaces engine). It is also important for the tools that can be used; often open-source game engines also rely on open-source/free tools. Engines like Source, Unity 3D, Unreal 3, are non-free and rely also on non open-source software, like Valve Hammer editor/Source SDK), Unity Editor, UnrealEd, ... In some cases, this non open-source software may be payware, there might be usage restrictions (of the program, perhaps on the media created, ...), ...
Question: You mentioned several times that Urho3d is a game engine, but I read that it's actually not a game engine at all, but a "graphics engine" (similar to OGRE3D ).
Answer: That's correct. We use the term "game engine" for simplicity, and assume that the user has either downloaded Urho3D with the rest of the components (sound, ...) that make up a game engine. For the newer game engines, this is also the way we have to go, simply because game engines tend to be made to use all features of the latest graphics cards, and use all CPU power. This is done by using the latest versions of these components. The downside however is that with these latest versions, the game engine won't run on your PC at all, if you have an older graphics card (for example a Shader Model 2.0 or Shader Model 1.0 graphics card), or not as much CPU power. So, we must use a graphics engine nowadays to be able to choose the components ourselves (in which case we can use older versions of these components). Since Enemy Territory: Legacy and AssaultCube were made a longer time ago (Urho3D is much, much more recent), this wasn't an issue with these before. Note btw that Urho3D allows to still use Shader Model 2.0 (and above) graphics card by swapping one of the components to an earlier version. Urho3D will btw also still be able to be run with a low-speed CPU (say 800 MHZ), all though it hence does need a slightly better graphics card (Shader Model 1.0 cards can't be used anymore, whereas with Enemy Territory: Legacy and AssaultCube: Reloaded, this was still possible). Shader Model 2.0 cards (and above) are cards produced after 2003 (NVIDIA), 2002 (AMD) and 2004 (Intel). Examples are ATI Radeon 9600, Nvidia Geforce FX 5000/Geforce 6000, Intel Graphics Media Accelerator HD3000, S3 DeltaChrome S8, ... The reason why Urho3D is hence at least a bit heavier, despite perhaps being better coded is because the shaders and per-pixel lighting, shadow mapping, and bullet rigid body physics are more extensive than before (even if you downgrade everything to lower versions or even remove some features completely).