What exactly is meant by "render to texture? "

Is that one temp form during the whole rendering ?
It means your application is rendering an image into some memory that in a subsequent rendering pass can be used as a texture.
In a racing game you'd want to render the scene surrounding your cars into a cube map and then you can use this cube map as a texture to render reflections over cars
 
The advantage of this way is:
1. No need to do multi-rendering to save computation.
What else ?
For me, it is still like a trade off. The efforts on the texture sampling is increased.



It means your application is rendering an image into some memory that in a subsequent rendering pass can be used as a texture.
In a racing game you'd want to render the scene surrounding your cars into a cube map and then you can use this cube map as a texture to render reflections over cars
 
The advantage of this way is:
1. No need to do multi-rendering to save computation.
What else ?
For me, it is still like a trade off. The efforts on the texture sampling is increased.


You can do some interesting effects, games use render to texture for lets say a person talking on a monitor screen, you can get higher quality rather then using animated picture files, or movie file.

Other things that its used for, render to texture for shadow maps, procedural effects, interactive water (interactive bump maps), motion blur, depth of field, and there are quite a bit more uses ;)
 
Last edited by a moderator:
Because you're rendering a frame into a something that's to be used as a texture - one is 'rendering to a texture (surface)'.
 
The advantage of this way is:
1. No need to do multi-rendering to save computation.
What else ?
For me, it is still like a trade off. The efforts on the texture sampling is increased.
What do you mean by "no need to do multi-rendering"? Rendering to a texture obviously means another render pass on top of the one(s) to render the back buffer. However, RTT allows you to do things that are very hard to do without it. The alternative to dynamic RTT environment maps, for example, would be to raytrace the scene to get the reflection.
 
What do you mean by "no need to do multi-rendering"? Rendering to a texture obviously means another render pass on top of the one(s) to render the back buffer. However, RTT allows you to do things that are very hard to do without it. The alternative to dynamic RTT environment maps, for example, would be to raytrace the scene to get the reflection.
I mean

It means your application is rendering an image into some memory that in a subsequent rendering pass can be used as a texture.
In a racing game you'd want to render the scene surrounding your cars into a cube map and then you can use this cube map as a texture to render reflections over cars

In the racing game, if we do not render the scene surrounding my cars intoa a cube map and then
use it as a texture, I would end up with rendering the scene eacn time when my car is moving ...
 
I mean in the racing game, if we do not render the scene surrounding my cars intoa a cube map and then use it as a texture, I would end up with rendering the scene eacn time when my car is moving ...

No, you still have to render the cube map each frame.
 
true no way around rendering to texture each frame from a realistic interactive reflective surface (realistic up to a certain point, texture sizes tend to create major pixilation)
 
Back
Top