That means that if we rotate the camera then normals’ facing will also change. Normals retrieved in this way are camera-space normals. In order to unpack it, we need to call DecodeDepthNormal as above seen above. Unity documentation says that depth and normals are packed in 16 bits each. Normal = mul( (float3x3)_CamToWorld, normal) Now let’s start with getting the normal: half4 frag (v2f i) : SV_TargetĭecodeDepthNormal(tex2D(_CameraDepthNormalsTexture, i.uv), depth, normal) _CameraDepthTexture always refers to the camera’s primary depth texture. By declaring a sampler called _CameraDepthTexture you will be able to sample the main depth texture for the camera. Why is it called that way? You can learn about it in Unity documentation:ĭepth textures are available for sampling in shaders as global shader properties. For this case there is sampler2D _CameraDepthNormalsTexture Since we’re set up on the camera that is set to generate depth-normals texture, now we are able to access it. Finding out where to snowĪs I explained before, we’d like to put the snow on surfaces that are facing upwards. First, we need to capture all the data passed by ScreenSpaceSnow script: sampler2D _MainTex ĭon’t worry if you don’t know why we need all this data yet. Let’s now focus only on the important part – the fragment shader. Note that if you create a new unlit unity shader (Create->Shader->Unlit Shader) you get mostly the same code. O.vertex = mul(UNITY_MATRIX_MVP, v.vertex) Here’s the basic template: Shader "TKoU/ScreenSpaceSnow" Our snow shader should be an unlit shader – we don’t want to apply any light information to it since on screen-space there’s no light. It’s only the basic setup, it will not generate a snow for you. execute the shader on input texture (src) and write to output (dest) _material.SetFloat("_SnowTexScale", SnowTextureScale)
_material.SetTexture("_SnowTex", SnowTexture) _material.SetFloat("_TopThreshold", TopThreshold) _material.SetFloat("_BottomThreshold", BottomThreshold) _material.SetColor("_SnowColor", SnowColor) _material.SetMatrix("_CamToWorld", GetComponent().cameraToWorldMatrix) Void OnRenderImage(RenderTexture src, RenderTexture dest) GetComponent().depthTextureMode |= DepthTextureMode.DepthNormals tell the camera to render depth and normals _material = new Material(Shader.Find("TKoU/ScreenSpaceSnow")) dynamically create a material that will use our shader Public class ScreenSpaceSnow : MonoBehaviour In our case the input data is an image rendered by the camera and some properties set up by the user. Usually this shader instead of rendering 3D object, renders full-screen image out of given input data. Now if you’ve never created an Image Effect before, you should know that these are build from at least one script and at least one shader. Setting pthTextureMode to DepthNormals will allow us to read screen depth (how far pixels are located from the camera) and normals (facing direction). Since the second option can be easily set by the image effect script itself, the first option can cause a problem if your game is already using a forward rendering path. If you have any idea why that could be, please leave a message in the comments section.) The depth shader was just rendered incorrectly.
No changes to any of the textures has been made. The only difference is that in the second one I enabled snow effect on the camera. In the images above you can see two screenshots presenting the same scene.
#UNITY WORLD TO SCREEN ON SCREEN HOW TO#
We’d like to show you how to create an Image Effect (screen-space shader) that will immediately change the season of your scene in Unity. Have you ever wondered how much time does it take to apply snow to all of the textures in your game? Probably a lot of times.
0 kommentar(er)
