Quick and easy question reguarding textures

[le Tigre 0]

Hey all, here a quick question. I've been reading a few tutorials in reguards to texture modling, across all types of models, When textureing things like metal and wood, I have noticed that many of the tutorials have the viewer adding things like high lights to the textures. my question is: do people accually do this?

Kind of an odd question, but I always thought that things like tight reflections off a metal surface happened in real time based on in game light sources. No?

Can someone explain this too me?

Thanks, loki!

[deleyt 0]

Your question is short but the answer is quite long

I'll try to explain in an easy way.

A model is made of geometry. Geometry is a collection of triangles, connected to each other.

A render program like 3D Studio Max, Maya, etc, adds a lightsource and shoots rays of light at the model.

The program then calculates where highlights and shadows should be put for that model in that position from that lightsource and does this in shades. It converts a 3D model into a 2D image, which still looks 3D to us, because of the shades.

The shades put depth into the 2D image.

Now, one step further:

To make the model look as real as possible, it needs to have as much detail as possible.

But to make a really detailed model, it also takes more triangles. And more triangles means more calculations. More calculations take more time.

Since present computers can not handle that huge amount of calculations, we take a shortcut.

This is where textures come in.

Textures are images of parts of the model which we put on the model, so we can cut down on the triangles (called polygons).

Since the model is toned down, it looses also parts of it's detail.

To give the illusion that the model still has a lot of detail, we put highlights and shades back into the texture.

That's why people paint highlights and shades onto their textures.

Make sense?

[le Tigre 0]

Great thanks!

[SnowFella 8]

To take D's quite exhausting answer alittle further in the game engine sence.

In a game engine that uses realtime lightsources you don't really have to paint highlights and shadows onto the models texture, it will all be taken care of by calculations done by the engine. What you might have to do though is add additional texturemaps to your model....like bump, specular and specular colour maps.

With a game like Ghost Recon, that doesn't have realtime light sources, you are however forced to paint in your own highlights and shadows into the texture map. The engine won't calculate where highlights and shadows will appear on the model and unless you paint them into the texturesheet your models is bound to end up looking very plain and without depth.

[le Tigre 0]

Great thanks guys! Really helpful. Im thinking of mabey doing a little moding myself when the new game comes out, we'll see what we have in terms of resources available to us. Ive accually studied 3D to a limit, but never in the sense of video games.

[Sup 0]

Great thanks guys! Really helpful. Im thinking of mabey doing a little moding myself when the new game comes out, we'll see what we have in terms of resources available to us. Ive accually studied 3D to a limit, but never in the sense of video games.

←

Well, with something like GRAW we should be seeing specular and normal maps. This will mean there's still quite a bit of painting to be done for highlights, but not into the diffuse maps.