We Keep Coding

C# OpenGL OpenTK - Camera moves away from drawings as I adjust eye.X value

I've been trying to set up a camera in OpenGL for a few days now, but I just can't get it working properly.
As the eye.X or eye.Y value increases/decreases, instead of "looking arround", the camera(I know there is no such thing as a camera...) moves away/gets closer to the drawings!
I hope following information can help you.(I also uploaded a Video, which is showing the error):
https://www.youtube.com/watch?v=OlD5X0EzkUw
These are my ModelView calculations on OnUpdateFrame.
_worldPosition = Matrix4.CreateTranslation(_worldX, _worldY, _worldZ);
_cameraMatrix = Matrix4.LookAt(eye.X, eye.Y, eye.Z, center.X, center.Y, center.Z, up.X, up.Y, up.Z);
_modelViewMatrix = _worldPosition * _cameraMatrix;
Here, I adjust the values with OnMouseMove function:
if (_mouseDown)
{
_mousePos.X = 2.0f * Mouse.X / Width - 1;
_mousePos.Y = 2.0f * Mouse.Y / Width - 1;
_mouseDeltaY = (_mousePosOld.Y - _mousePos.Y) / 10;
_mouseDeltaX = (_mousePosOld.X - _mousePos.X) / 10;
eye.X -= _mouseDeltaX;
eye.Y -= _mouseDeltaY;
}
And finally my Vertex shader:
#version 440 core
layout (location = 0) in vec4 position;
layout(location = 1) in vec4 color;
layout(location = 2) uniform mat4 projectionMatrix;
layout (location = 3) uniform mat4 modelViewMatrix;
out vec4 vs_color;
void main(void)
{
gl_Position = projectionMatrix * modelViewMatrix * position;
vs_color = color;
}

To look around you have to rotate center around eye.
e.g If you want to do a rotation around the x axis (up, down) the you have to setup a rotation matrix dependent on _mouseDeltaY:
float angle_degree = _mouseDeltaY;
Matrix4 rotateY = Matrix4.CreateRotationY(angle_degree * (float)Math.PI / 180.0f);
Compute the vector from eye to center and rotate it:
Vector3 toTarget = center - eye;
toTarget = Vector4.Transform(new Vector4(toTarget, 0.0f), rotateY).Xyz;
Finally compute the new center
center = eye + toTarget;
eye and center are assumed to be of type Vector3. Possibly you have to change the scale of _mouseDeltaY for your needs.

Can a shader rotate shapes to face camera?

I made a scene where balls appear in 3D space. Triangle balls spend a lot of resources. So I did this using 2d surfaces (quad) with a ball texture. But now I need to adjust the direction of the shape every time the camera moves. I do this using position transformation and the LookAt method. The question is can I optimize this? If it is possible to rotate the shapes with shader, this will greatly help.
using UnityEngine;
public class WorldSurf : MonoBehaviour
{
GameObject[] matrix;
int xSize = 20;
int ySize = 20;
int zSize = 20;
// Start is called before the first frame update
void Start()
{
matrix = new GameObject[xSize * ySize * zSize];
//var shader = Shader.Find("Legacy Shaders/Diffuse");
var shader = Shader.Find("Sprites/Default");
//var texture = Resources.Load<Texture>("Textures/Ball_01");
var i = 0;
for (var x = 0f; x < xSize; ++x)
{
for (var y = 0f; y < ySize; ++y)
{
for (var z = 0f; z < zSize; ++z)
{
var texture = Resources.Load<Texture>("Textures/Ball_" + ((int)Random.Range(0, 15)).ToString("00"));
matrix[i++] = CreateQuad(x * 3, y * 3, z * 3, shader, texture);
}
}
}
}
static GameObject CreateQuad(float x, float y, float z, Shader shader, Texture texture)
{
var quad = GameObject.CreatePrimitive(PrimitiveType.Quad);
quad.transform.position = new Vector3(x, y, z);
quad.transform.forward = Camera.main.transform.forward;
var rend = quad.GetComponent<Renderer>();
rend.material.shader = shader;
rend.material.mainTexture = texture;
//rend.material.color = Color.red;
return quad;
}
// Update is called once per frame
void Update()
{
var pos = Camera.main.transform.position;
foreach (var itm in matrix)
{
itm.transform.LookAt(pos);
}
}
}

Generally yes, and in this specific case where you want a quad to align with the camera makes doing so very easy.
What you want is called a "billboard shader". Here is an example from Wikibooks:
Shader "Cg shader for billboards" {
Properties {
_MainTex ("Texture Image", 2D) = "white" {}
_ScaleX ("Scale X", Float) = 1.0
_ScaleY ("Scale Y", Float) = 1.0
}
SubShader {
Tags {"Queue"="Transparent" "IgnoreProjector"="True" "RenderType"="Transparent"}
ZWrite Off
Blend SrcAlpha OneMinusSrcAlpha
Pass {
CGPROGRAM
#pragma vertex vert
#pragma fragment frag
// User-specified uniforms
uniform sampler2D _MainTex;
uniform float _ScaleX;
uniform float _ScaleY;
struct vertexInput {
float4 vertex : POSITION;
float4 tex : TEXCOORD0;
};
struct vertexOutput {
float4 pos : SV_POSITION;
float4 tex : TEXCOORD0;
};
vertexOutput vert(vertexInput input)
{
vertexOutput output;
output.pos = mul(UNITY_MATRIX_P,
mul(UNITY_MATRIX_MV, float4(0.0, 0.0, 0.0, 1.0))
+ float4(input.vertex.x, input.vertex.y, 0.0, 0.0)
* float4(_ScaleX, _ScaleY, 1.0, 1.0));
output.tex = input.tex;
return output;
}
float4 frag(vertexOutput input) : COLOR
{
return tex2D(_MainTex, float2(input.tex.xy));
}
ENDCG
}
}
}
And an explanation of how this works:
The basic idea is to transform only the origin ( 0 , 0 , 0 , 1 )
of the object
space to view space with the standard model-view transformation
UNITY_MATRIX_MV. (In homogeneous coordinates all points have a 1 as
fourth coordinate; see the discussion in Section “Vertex
Transformations”.) View space is just a rotated version of world space
with the xy plane parallel to
the view plane as discussed in Section “Vertex Transformations”. Thus,
this is the correct space to construct an appropriately rotated
billboard. We subtract the x y object coordinates (vertex.x and vertex.y) from the transformed
origin in view coordinates and then transform the result with the
projection matrix UNITY_MATRIX_P.
This will produce an output like this:

Black hole distortion shader in Unity

I found shader code that has the effect of warping a space around a certain point. It's a cool effect, but it's missing some animation, so I've added something to it:
Shader "Marek/BlackHoleDistortion"
{
Properties {
_DistortionStrength ("Distortion Strength", Range(0, 10)) = 0
_Timer("Timer", Range(0, 10)) = 0
_HoleSize ("Hole Size", Range(0, 1)) = 0.1736101
_HoleEdgeSmoothness ("Hole Edge Smoothness", Range(1, 4)) = 4
_ObjectEdgeArtifactFix ("Object Edge Artifact Fix", Range(1, 10)) = 1
}
SubShader {
Tags {
"IgnoreProjector"="True"
"Queue"="Transparent"
"RenderType"="Transparent"
}
GrabPass{ }
Pass {
Name "FORWARD"
Tags {
"LightMode"="ForwardBase"
}
ZWrite Off
CGPROGRAM
#include "UnityCG.cginc"
#pragma vertex vert
#pragma fragment frag
#pragma multi_compile_fwdbase
#pragma only_renderers d3d9 d3d11 glcore gles
#pragma target 3.0
uniform sampler2D _GrabTexture;
uniform float _DistortionStrength;
uniform float _HoleSize;
uniform float _HoleEdgeSmoothness;
uniform float _ObjectEdgeArtifactFix;
uniform float _Timer;
struct VertexInput {
float4 vertex : POSITION;
float3 normal : NORMAL;
};
struct VertexOutput {
float4 pos : SV_POSITION;
float4 posWorld : TEXCOORD0;
float3 normalDir : TEXCOORD1;
float4 projPos : TEXCOORD2;
};
VertexOutput vert (VertexInput v) {
VertexOutput o = (VertexOutput)0;
o.normalDir = UnityObjectToWorldNormal(v.normal);
o.posWorld = mul(unity_ObjectToWorld, v.vertex);
o.pos = UnityObjectToClipPos(v.vertex);
o.projPos = ComputeScreenPos(o.pos);
COMPUTE_EYEDEPTH(o.projPos.z);
return o;
}
float4 frag(VertexOutput i) : COLOR {
i.normalDir = normalize(i.normalDir);
float3 viewDirection = normalize(_WorldSpaceCameraPos.xyz - i.posWorld.xyz);
float3 normalDirection = i.normalDir;
float2 sceneUVs = (i.projPos.xy / i.projPos.w);
float node_9892 = (_HoleSize * -1.0 + 1.0);
float node_3969 = (1.0 - pow(1.0 - max(0, dot(normalDirection, viewDirection)), clamp(_DistortionStrength - _Timer, 0, _DistortionStrength)));
float node_9136 = (length(float2(ddx(node_3969), ddy(node_3969))) * _HoleEdgeSmoothness);
float node_4918 = pow(node_3969, 6.0);
float node_1920 = (1.0 - smoothstep((node_9892 - node_9136), (node_9892 + node_9136), node_4918));
float3 finalColor = (
lerp(
float4(node_1920, node_1920, node_1920, node_1920),
float4(1, 1, 1, 1),
pow(
pow(1.0 - max(0, dot(normalDirection, viewDirection)), 1.0),
_ObjectEdgeArtifactFix
)
) * tex2D(_GrabTexture, ((node_4918 * (sceneUVs.rg * _Time * -2.0 + 1.0)) + sceneUVs.rg)).rgb).rgb;
return fixed4(finalColor, 1);
}
ENDCG
}
}
FallBack "Diffuse"
}
Now, the problem is that in order to make the distortion disappear after certain time, I need to include some variable into the equation - here I'm calling it _Timer. I'm not using the _Time built in because of obvious reasons - it's an ever growing value and I need something that starts from 0 each time the object using this shader is made active. C# code handling passing that parameter looks as follows:
public void Update() {
_timeElapsed += Time.deltaTime;
_renderer.material.SetFloat("_Timer", _timeElapsed);
}
The question is - can I do it better? I would like this shader's code to be more of a self-contained thing - without the need to pass parameters from cs script to it.

Can I do it better?
In-short, yes and no. If you want the shader to behave differently per material you simply cannot avoid passing a property from C#. You can however avoid doing this in Update by passing a start time and computing the elapse time in the shader.
C#
void OnEnable ()
{
_renderer.material.SetFloat("_StartTime", Time.timeSinceLevelLoad);
}
Shader
uniform float _StartTime;
float4 frag(VertexOutput i) : COLOR
{
float elapse = _Time.y - _StartTime;
}
Now, although this will tie directly into the setup you are currently using, it should be noted that accessing the .material property will clone the material (which can break batching, among other things).
This can be avoided with the more recent introduction of MaterialPropertyBlocks.

Unity provides a handful of built-in values for your shaders: things like current object’s transformation matrices, time etc.
You just use them in ShaderLab like you’d use any other property, the only difference is that you don’t have to declare it somewhere - they are “built in”.
https://docs.unity3d.com/455/Documentation/Manual/SL-BuiltinValues.html
there is a clever way of giving you 4 variations of the value, potentially saving you a multiply operation by re-using the pre-multiplied value for every pixel being rendered. There are 4 values available.
_Time.x = time / 20
_Time.y = time
_Time.z = time * 2
_Time.w = time * 3
this is a simple example that show you how it works:
Shader "Example/Circle"
{
Properties
{
}
SubShader
{
Cull Off
Pass
{
CGPROGRAM
#pragma vertex vert
#pragma fragment frag
#include "UnityCG.cginc"
struct appdata
{
float4 vertex : POSITION;
float2 uv : TEXCOORD0;
};
struct v2f
{
float2 uv : TEXCOORD0;
float4 vertex : SV_POSITION;
};
v2f vert (appdata v)
{
v2f o;
o.vertex = UnityObjectToClipPos(v.vertex);
o.uv = v.uv;
return o;
}
float circle(in float2 _st, in float _radius){
float2 dist = distance(_st,float2(0.5,0.5));
float result = step(dist,_radius);
return result;
}
fixed4 frag (v2f i) : SV_Target
{
float WaveTime = sin(_Time.z);
float3 color = float3(1,1,1)*circle(i.uv,WaveTime);
return float4( color, 1.0 );
}
ENDCG
}
}
}
In comments you mentioned that you want reset time value when you enable it , so here you need to Initialize Time value with script.
so you should use your own Time In shader:
Properties
{
_Timer("Timer",Float) = 0
}
float WaveTime = sin(_Timer);
using System.Collections;
using UnityEngine;
public class Circle : MonoBehaviour {
public float _timeElapsed;
void OnEnable(){
_timeElapsed = 0;
}
public void Update() {
_timeElapsed += Time.deltaTime;
var _renderer = GetComponent<MeshRenderer>();
_renderer.material.SetFloat("_Timer", _timeElapsed);
}
}

Volume Ray Marching rendered always on top of the other objects

I have a really big problem that has been bothering me for so long and I can't seem to find the solution. I have downloaded this project https://github.com/brianasu/unity-ray-marching/tree/volumetric-textures (Unitypackage with my project here https://dl.dropboxusercontent.com/u/27758186/ApplicationVolume.unitypackage ) , that is about volume rendering.
The problem is that, as you can see, the volume renders ALWAYS in front of everything else (try placing a cube in front of the cube with the volume). I have tried a lot of things but none seems to work.
I think it might be an issue of the shaders used. In the main Camera, a RayMarching script is attached, which contains an OnRenderImage method, which creates a new Camera (although disabled), and renders the volume.
I don't know if it is a shader issue, then, or more like a camera issue (is rendering with replaced shaders). I attach my current project for testing so you don't have to download from git, and I just want the volume (head), to appear behind the rectangles (not seen) when it's actually behind, and in front when it is, pretty much like an standard geometry, but it appears always on top...
Any help or suggestion would be GREATLY appreciated, I'm kind of desperate as anything works and I'm pretty sure it is a fairly easy issue.
The code of the shader of the Ray Marching is as follows. Should it do some kind of ZTesting to not show the fragments that are covered by any other object?
Shader "Hidden/Ray Marching/Ray Marching"
{
CGINCLUDE
#include "UnityCG.cginc"
#pragma target 3.0
#pragma profileoption MaxLocalParams=1024
#pragma profileoption NumInstructionSlots=4096
#pragma profileoption NumMathInstructionSlots=4096
struct v2f {
float4 pos : POSITION;
float2 uv[2] : TEXCOORD0;
};
sampler3D _VolumeTex;
float4 _VolumeTex_TexelSize;
sampler2D _FrontTex;
sampler2D _BackTex;
float4 _LightDir;
float4 _LightPos;
float _Dimensions;
float _Opacity;
float4 _ClipDims;
float4 _ClipPlane;
v2f vert( appdata_img v )
{
v2f o;
o.pos = mul(UNITY_MATRIX_MVP, v.vertex);
o.uv[0] = v.texcoord.xy;
o.uv[1] = v.texcoord.xy;
#if SHADER_API_D3D9
if (_MainTex_TexelSize.y < 0)
o.uv[0].y = 1-o.uv[0].y;
#endif
return o;
}
#define TOTAL_STEPS 128.0
#define STEP_CNT 128
#define STEP_SIZE 1 / 128.0
half4 raymarch(v2f i, float offset)
{
float3 frontPos = tex2D(_FrontTex, i.uv[1]).xyz;
float3 backPos = tex2D(_BackTex, i.uv[1]).xyz;
float3 dir = backPos - frontPos;
float3 pos = frontPos;
float4 dst = 0;
float3 stepDist = dir * STEP_SIZE;
for(int k = 0; k < STEP_CNT; k++)
{
float4 src = tex3D(_VolumeTex, pos);
// clipping
float border = step(1 - _ClipDims.x, pos.x);
border *= step(pos.y, _ClipDims.y);
border *= step(pos.z, _ClipDims.z);
border *= step(0, dot(_ClipPlane, float4(pos - 0.5, 1)) + _ClipPlane.w);
// Standard blending
src.a *= saturate(_Opacity * border);
src.rgb *= src.a;
dst = (1.0f - dst.a) * src + dst;
pos += stepDist;
}
return 3.0F*dst;// + dst;
}
ENDCG
Subshader {
Fog { Mode off }
Pass
{
CGPROGRAM
#pragma vertex vert
#pragma fragment frag
half4 frag(v2f i) : COLOR {
return raymarch(i, 0);
}
ENDCG
}
}
Fallback off
} // shader

Black Screen due to shaders ps_3

I'm currently using shaders in my game, it's working fine with a nVidia GeForceGT330m but with an ATI 4670 (which supports ps_4.1) I encounter a black screen.
Here is the source of the HLSL effect:
struct Explo
{
float3 position;
float4 color;
float power;
int time;
};
float2 DisplacementScroll;
texture colortexture;
int nb;
Explo explos[5];
float ambient;
float4 ambientColor;
float screenWidth;
float screenHeight;
sampler ColorMap = sampler_state
{
Texture = <colortexture>;
};
float4 CalculateLight(Explo ex, float4 base, float3 pixelPosition)
{
float3 direction = ex.position - pixelPosition;
float distance = 1 / length(ex.position - pixelPosition) * ex.power;
float amount = max(dot(base, normalize(distance)), 0);
return base * distance * amount * ex.color * ambient;
}
float4 Explosion(float2 texCoords : TEXCOORD0) : COLOR
{
//texCoords = tex2D(NormalMap, DisplacementScroll + texCoords / 3)*0.2 - 0.15;
float4 base = tex2D(ColorMap, texCoords);
float3 pixelPosition = float3(screenWidth * (texCoords.x),
screenHeight * (texCoords.y),0);
float4 finalColor = (base * ambientColor * ambient);
for (int i=0; i<nb; i++)
{
finalColor += CalculateLight(explos[i], base, pixelPosition);
}
return finalColor;
}
technique KaBoom
{
pass Pass1
{
PixelShader = compile ps_3_0 Explosion();
}
}

I remember once I had a similar problem. A shader just didn't work on ATI. The problem was that the vertex and pixel shaders were compiled to different shader models (vs_3_0 and ps_2_0). It worked for NVIDIA, but not for ATI. In your case you're only binding a pixel shader for the pass and who knows what the last vertex shader was.
Granted, this is relevant only if you're dead sure the problem is with the shader and not something else, e.g. your DIPs.
Good luck