uniform vec4 lightDir;varying vec2 vTexCoord;
varying vec3 vHatchWeights0;
varying vec3 vHatchWeights1;
void main(void)
{
gl_Position = gl_ModelViewProjectionMatrix * gl_Vertex;
vTexCoord = vec2(gl_MultiTexCoord0);
vec3 posW = gl_NormalMatrix * gl_Vertex.xyz;
vec3 normalW = normalize(gl_NormalMatrix * gl_Normal);
float diffuse = min(1.0,max(0.0,dot(lightDir.xyz,normalW)));
diffuse = diffuse * diffuse;
diffuse = diffuse * diffuse;
float hatchFactor = diffuse * 6.0;
vec3 weight0 = vec3(0.0);
vec3 weight1 = vec3(0.0);
if (hatchFactor>5.0)
{
weight0.x = 1.0;
} // End if
else if (hatchFactor>4.0)
{
weight0.x = 1.0 - (5.0 - hatchFactor);
weight0.y = 1.0 - weight0.x;
} // End else if
else if (hatchFactor>3.0)
{
weight0.y = 1.0 - (4.0 - hatchFactor);
weight0.z = 1.0 - weight0.y;
} // End else if
else if (hatchFactor>2.0)
{
weight0.z = 1.0 - (3.0 - hatchFactor);
weight1.x = 1.0 - weight0.z;
} // End else if
else if (hatchFactor>1.0)
{
weight1.x = 1.0 - (2.0 - hatchFactor);
weight1.y = 1.0 - weight1.x;
} // End else if
else if (hatchFactor>0.0)
{
weight1.y = 1.0 - (1.0 - hatchFactor);
weight1.z = 1.0 - weight1.y;
} // End else if
vHatchWeights0 = weight0;
vHatchWeights1 = weight1;
}
[/code]
pixel shader:
[code]
uniform sampler2D Hatch0;
uniform sampler2D Hatch1;
uniform sampler2D Hatch2;
uniform sampler2D Hatch3;
uniform sampler2D Hatch4;
uniform sampler2D Hatch5;
varying vec2 vTexCoord;
varying vec3 vHatchWeights0;
varying vec3 vHatchWeights1;
void main(void)
{
vec4 hatchTex0 = texture2D(Hatch0,vTexCoord) * vHatchWeights0.x;
vec4 hatchTex1 = texture2D(Hatch1,vTexCoord) * vHatchWeights0.y;
vec4 hatchTex2 = texture2D(Hatch2,vTexCoord) * vHatchWeights0.z;
vec4 hatchTex3 = texture2D(Hatch3,vTexCoord) * vHatchWeights1.x;
vec4 hatchTex4 = texture2D(Hatch4,vTexCoord) * vHatchWeights1.y;
vec4 hatchTex5 = texture2D(Hatch5,vTexCoord) * vHatchWeights1.z;
vec4 hatchColor = hatchTex0 +
hatchTex1 +
hatchTex2 +
hatchTex3 +
hatchTex4 +
hatchTex5;
gl_FragColor = hatchColor;
}
[/code]
我们实现hatching需要多张hatching纹理，
这些hatching是一系列又淡到浓的素描纹理，如下：
在VS中。我们通过 N dot L 算出了此顶点的受光亮度，如下
[code]
float diffuse = min(1.0,max(0.0,dot(lightDir.xyz,normalW)));
diffuse = diffuse * diffuse;
diffuse = diffuse * diffuse;
[/code]
后面把diffuse平方两次是为了，把亮度降低点，把整体亮度向暗方向偏移，
这里只是为了更好的视觉效果。
现在最关键的就是要通过这个亮度值决定使用哪些hatch纹理的权重(hatch0, hatch1, hatch2, hatch3, hatch4, hatch5)
在这里为了减少vs向ps插值变量个数，巧妙使用了两个vec3来记录每个hatch图所绘制的权重，
见以下代码：
[code]
float hatchFactor = diffuse * 6.0;
vec3 weight0 = vec3(0.0);
vec3 weight1 = vec3(0.0);
if (hatchFactor>5.0)
{
weight0.x = 1.0;
} // End if
else if (hatchFactor>4.0)
{
weight0.x = 1.0 - (5.0 - hatchFactor);
weight0.y = 1.0 - weight0.x;
} // End else if
else if (hatchFactor>3.0)
{
weight0.y = 1.0 - (4.0 - hatchFactor);
weight0.z = 1.0 - weight0.y;
} // End else if
else if (hatchFactor>2.0)
{
weight0.z = 1.0 - (3.0 - hatchFactor);
weight1.x = 1.0 - weight0.z;
} // End else if
else if (hatchFactor>1.0)
{
weight1.x = 1.0 - (2.0 - hatchFactor);
weight1.y = 1.0 - weight1.x;
} // End else if
else if (hatchFactor>0.0)
{
weight1.y = 1.0 - (1.0 - hatchFactor);
weight1.z = 1.0 - weight1.y;
} // End else if
vHatchWeights0 = weight0;
vHatchWeights1 = weight1;
[/code]
首先diffuse值被乘以6，是为了好方便区分区段。
下面一系列判断就是判断diffuse值到底在哪个区段里面，
以决定相应的权重。
vHatchWeights0.x 对应hatch0 的权重
vHatchWeights0.y 对应hatch1 的权重
vHatchWeights0.z 对应hatch2 的权重
vHatchWeights1.x 对应hatch3 的权重
vHatchWeights1.y 对应hatch4 的权重
vHatchWeights1.z 对应hatch5 的权重
接下来vHatchWeights0, vHatchWeight1 被插值到了ps
后面就很简单了，根据相应的权重，取得相应纹理的颜色，再合成起来。
代码如下：
[code] vec4 hatchTex0 = texture2D(Hatch0,vTexCoord) * vHatchWeights0.x;
vec4 hatchTex1 = texture2D(Hatch1,vTexCoord) * vHatchWeights0.y;
vec4 hatchTex2 = texture2D(Hatch2,vTexCoord) * vHatchWeights0.z;
vec4 hatchTex3 = texture2D(Hatch3,vTexCoord) * vHatchWeights1.x;
vec4 hatchTex4 = texture2D(Hatch4,vTexCoord) * vHatchWeights1.y;
vec4 hatchTex5 = texture2D(Hatch5,vTexCoord) * vHatchWeights1.z;
vec4 hatchColor = hatchTex0 +
hatchTex1 +
hatchTex2 +
hatchTex3 +
hatchTex4 +
hatchTex5;
