As a newcomer to the Unreal Engine, I’m loving the possibilities it offers for creating immersive games and stunning visual effects. However, as I delve deeper into the engine’s capabilities, I find myself facing questions that are not always easy to answer.
One such question is whether it’s possible to get a lower resolution of the SceneTexture’s PostProcessInput0? From my understanding, this texture contains the final output of the scene before any post-processing effects are applied? Can I somehow scale down this texture and continue my Blueprint with this texture/image?
Downscaling any texture can be as simple as just sampling it at every other pixel (or less) instead of every pixel. You can do this using the UV input of the sampler.
I’ve been trying to understand how to sample from the SceneTexture’s PostProcessInput0.
I was thinking the only option would be to translate SceneTextures to Texture Objects? Might have it all wrong, though.
Thank you so much for pointing me in a direction last week. I’ve been doing some tests and I think I’m one step closer but there is one thing I don’t quite understand.
It’s not enough to just divide, because UVs range from 0-1. If you divide, then you will only fill the 0-0.5 space.
You actually need to have multiple repeated coordinates to increase the apparent size of a pixel. Imagine you have a grid of pixels:
(0,0)(1,0)(2,0)
(0,1)(1,1)(1,2)
And so on…
To downscale, you would want to sample the same coordinate for multiple pixels. So the end result is:
(0,0)(0,0)(2,0)(2,0)
(0,0)(0,0)(2,0)(2,0)
(0,2)(0,2)(2,2)(2,2)
(0,2)(0,2)(2,2)(2,2)
Each pixel looks twice as big because it’s actually twice as many pixels sampling the same coordinate.
Unfortunately, I think the information is more aligned to what this question is about. To able to manipulate each pass / scale individually and then at the end bake them.
Thank you so much for your answer @BananableOffense. Yes, I think you are right @wiquid_78… I’m able to “pixelate” the post process material (see images below) but what I want to do is reduce the number of pixels in the texture. If the original screen texture is 1000x1000 px, then the scaled down version should be 500x500 px… then 250x250 px and so on.
Oh, interesting. I’ve been researching it and cannot find information on how to connect it to a post-process material (PostProcessInput0). Do you know how to achieve this?
I don’t think it matters what kind of material it is. Only that the desired output is connected to the emissive. I don’t see why a post process material would work any different than anything else here, but I haven’t had a reason to try it myself.
I made a video about an engine modification that does just that, scale the SceneColor so shaders run at a downscaled resolution : https://www.youtube.com/watch?v=K398K2VWSxQ
