Question about draw2D Image in post processing

[The_Glitch]

I've taken a look at some examples around Irrlicht and I've seen people taking a RTT texture and using draw2D on some of them. My question is, is this Irrlichts method of taking a rendered result of a screenquad and drawing it back into a render target?? If so this may be my issue as I have some shaders that need to use the new result on a image of the previous screenquad and I wasn't sure how to make a rendered screenquad effect back into a usable texture again.

[The_Glitch]

Many sad faces LOL

[mongoose7]

It is difficult to understand what you mean. With Irrlicht's implementation, you can render to a RT and then, when you detach it, you can use it as an ordinary texture. So you can use it in draw2DImage. If you render such a texture into a screen quad, and you've set a render target, then you have a texture which is the result of the render. This is Irrlicht's *implementation* of render targets, but it is not Irrlicht's *method*, since the engine itself doesn't use render targets.

Anyway, a render target becomes a usable texture as soon as it is detached.

[The_Glitch]

Okay I see.

Thanks

Let's say I have a render target

Code: Select all

video::ITexture* rt = 0; 

Code: Select all

 rt = driver->addRenderTargetTexture(core::dimension2d<u32>(512,512), "RTT1", video::ECF_A8R8G8B8);

now
I pass that to a full screenquad and it's blurred. My question is from there how can I put that result back into rt or even a new rt if that's not possible to reuse the old one. So I can pass it to a bloom shader which uses it as a texture register slot 1.

[mongoose7]

Create another RT rt2. Render a screen quad into rt2 using rt. You now have two textures which you can apply to different purposes. Or you can just create rt2 in the same way you created rt.

[Mel]

You need two rendertargets. While you render to one, you use the other, and viceversa, this allows to chain effects.

[The_Glitch]

LOL, Thanks but I figured out most or all the problems I was having.