After looking around here it seems that alot of people have migrated from RCT3 Forums and have little or no knowledge of normal maps because that engine didnt support them. I thought i would make a thread explaining them and how they can be made easily.
First of all. Normal maps are a little like bump maps except they are 'pre processed'. Bump maps are greyscale images that are used to create the illusion of detail where there isnt any. These used to be used in games but the problem was that they required processing to calculate the faked geometry. Normal maps use Red green and blue to show the direction of the geometry so no processing needs to be done meaning its alot faster.
In lamens terms Normal mapping is just a new faster way of faking surface details. Its been around in games consoles since the Sega Dreamcast!
To make a normal map you must make a high resolution version of your model. It can be as many polygons as you want (which might frighten some of you RCT3ers!) You then take this high resolution mesh and project the details onto a texture (a normal map).
This normal map is then applied to a low resolution version of your mesh giving the illusion of surface details.
There is a free program called xnormal which makes this process very simple. Ill post a link at the end of this post. Theres plenty of youtube tutorials on how it works but its pretty easy anyway. This program takes your high res, UV Unwrapped mesh and creates a normal map. The it applies that texture onto the low res mesh and voila you have a great yet low poly model.
Below is an example;
On the left, you see the high resolution mesh. In the middle is the very low poly version of that mesh and on the right is the low poly mesh with normal mapping. As you can see its a great way to save polygons.XNORMAL DOWNLOAD: http://www.xnormal.net/1.aspx
NVIDIA also make a really good plugin that turns diffuse maps into normal maps. Theres also a program called ndo that does the same thing.
Another amazing feature of xnormals are its Ambient occlusion mapping capabilities. For those who are unaware, Ambient occlusion are the small shadows that form where surfaces meet
The left temple has ambient occlusion on the textures whilst the right has none. It looks infinitely better because of this small tweak.
Here is what an occlusion pass looks like.
Basically once you have uv unwrapped your model and takes it into xnormal you can generate an ambient occlusion map. You can then open your diffuse texture in photoshop or gimp. Place the ambient occlusion layer ontop of the diffuse channel, and set it to multiply. This will add the shadows to your texture and make it look 100 times better.
Once i have the time i will possibly make a video tutorial detailing a simple object from start to finish i.e. from sketchup to an in game object but since i neither have the game nor any indication of the acceptable formats it could be a fair amount of time.
There are alot of tutorials on youtube on xnormals and there is a good overview of the program on the xnormal site itself.