ps部分
ps_2_0
概览
| Instruction Set | ||||||||
| Name | Description | Instruction slots | Setup | Arithmetic | Texture | New | ||
| 绝对值 | 1 | x | x | |||||
| 加法 | 1 | x | ||||||
| Compare source to 0 | 1 | x | ||||||
| 叉积 | 2 | x | x | |||||
| 为采样器声明贴图维数 | 0 | x | x | |||||
| 声明顶点shader输出寄存器和像素shader输入寄存器之间的联合 | 0 | x | x | |||||
| 定义常量 | 0 | x | ||||||
| 2D点积之后加法运算 | 2 | x | x | |||||
| 3D点积 | 1 | x | ||||||
| 4D点积 | 1 | x | ||||||
| 全精度的2的x次方 | 1 | x | x | |||||
| 小数部分 | 1 | x | x | |||||
| 全精度的 log₂(x) | 1 | x | x | |||||
| 线性插值 | 2 | x | ||||||
| 3x2 乘法 | 2 | x | x | |||||
| 3x3 乘法 | 3 | x | x | |||||
| 3x4 乘法 | 4 | x | x | |||||
| 4x3 乘法 | 3 | x | x | |||||
| 4x4 乘法 | 4 | x | x | |||||
| 乘法之后加法运算 | 1 | x | ||||||
| 最大值 | 1 | x | x | |||||
| 最小值 | 1 | x | x | |||||
| 赋值 | 1 | x | ||||||
| 乘法 | 1 | x | ||||||
| 无运算 | 1 | x | ||||||
| 单位化 | 3 | x | x | |||||
| 幂运算 | 3 | x | x | |||||
| 版本 | 0 | x | ||||||
| 倒数 | 1 | x | x | |||||
| 平方根后倒数 | 1 | x | x | |||||
| sin和cos | 8 | x | x | |||||
| 减法 | 1 | x | ||||||
| 中断像素渲染 | 1 | x | ||||||
| 采样一张贴图 | 1 | x | x | |||||
| 根据w部分进行对贴图的level-of-detail bias采样 | 1 | x | x | |||||
| 根据w部分进行对贴图的投影划分采样 | 1 | x | x | |||||
部分函数细节
crs
语法
crs dst, src0, src1
叉积
算法
dest.x = src0.y * src1.z - src0.z * src1.y;
dest.y = src0.z * src1.x - src0.x * src1.z;
dest.z = src0.x * src1.y - src0.y * src1.x;
dcl_samplerType
语法
dcl_samplerType s#
声明一个像素shader采样器
· _2d
· _cube
· _volume
示例
dcl_cube t0.rgb; // Define a 3D texture map.
add r0, r0, t0; // Perturb texture coordinates.
texld r0, s0, r0; // Load r0 with a color sampled from stage0
// at perturbed texture coordinates r0.
// This is a dependent texture read.
dp2add
语法
dp2add dst, src0, src1, src2.{x|y|z|w}
算法
dest = src0.r * src1.r + src0.g * src1.g + src2.replicate_swizzle
// The scalar result is replicated to the write mask components
nrm
语法
nrm dst, src
单位化一个3D向量
squareRootOfTheSum = (src0.x*src0.x + src0.y*src0.y +src0.z*src0.z)1/2;
算法
dest.x = src0.x * (1 / squareRootOfTheSum);
dest.y = src0.y * (1 / squareRootOfTheSum);
dest.z = src0.z * (1 / squareRootOfTheSum);
dest.w = src0.w * (1 / squareRootOfTheSum);
sincos
语法
ps_2_0 和 ps_2_x
sincos dst.{x|y|xy}, src0.{x|y|z|w}, src1, src2
ps_3_0
sincos dst.{x|y|xy}, src0.{x|y|z|w}
算法
ps_2_0 and ps_2_x
写入通道为.x
dest.x = cos(V)
dest.y is undefined when the instruction completes
dest.z is undefined when the instruction completes
dest.w is not touched by the instruction
写入通道为.y
dest.x is undefined when the instruction completes
dest.y = sin(V)
dest.z is undefined when the instruction completes
dest.w is not touched by the instruction
写入通道为.xy
dest.x = cos(V)
dest.y = sin(V)
dest.z is undefined when the instruction completes
dest.w is not touched by the instruction
ps_3_0
写入通道为.x
dest.x = cos(V)
dest.y is undefined when the instruction completes
dest.z is undefined when the instruction completes
dest.w is not touched by the instruction
写入通道为.y
dest.x is undefined when the instruction completes
dest.y = sin(V)
dest.z is undefined when the instruction completes
dest.w is not touched by the instruction
写入通道为.xy
dest.x = cos(V)
dest.y = sin(V)
dest.z is undefined when the instruction completes
dest.w is not touched by the instruction
示例角度制转弧度制
def c0, pi, 0.5, 2*pi, 1/(2*pi)
mad r0.x, input_angle, c0.w, c0.y
frc r0.x, r0.x
mad r0.x, r0.x, c0.z, -c0.x
vs部分
vs_2_0
概览
| Instruction Set | ||||||||
| Name | Description | Instruction slots | Setup | Arithmetic | Flow control | New | ||
| 绝对值 | 1 | x | x | |||||
| 加法运算 | 1 | x | ||||||
| 调用一个子程序 | 2 | x | x | |||||
| 如果一个boolean寄存器不为0的话调用一个子程序 | 3 | x | x | |||||
| 叉积 | 2 | x | x | |||||
| 0 | x | |||||||
| 定义常量 | 0 | x | ||||||
| 定义一个 Boolean 常量 | 0 | x | x | |||||
| 定义一个integer 常量 | 0 | x | x | |||||
| 3D 点积 | 1 | x | ||||||
| 4D 点积 | 1 | x | ||||||
| 计算距离向量 | 1 | x | ||||||
| 1 | x | x | ||||||
| endif - vs...else - vs block | 1 | x | x | |||||
| 2 | x | x | ||||||
| 结束一个 repeat block | 2 | x | x | |||||
| 全精度计算2的幂 | 1 | x | ||||||
| 半精度计算2的幂 | 1 | x | ||||||
| 小数部分 | 1 | x | ||||||
| if块 | 3 | x | x | |||||
| Label | 0 | x | x | |||||
| 局部光照计算 | 3 | x | ||||||
| 全精度计算log₂(x) | 1 | x | ||||||
| 半精度计算log₂(x) | 1 | x | ||||||
| 循环 | 3 | x | x | |||||
| 线性插值 | 2 | x | x | |||||
| 3x2 乘法 | 2 | x | ||||||
| 3x3 乘法 | 3 | x | ||||||
| 3x4 乘法 | 4 | x | ||||||
| 4x3 乘法 | 3 | x | ||||||
| 4x4 乘法 | 4 | x | ||||||
| 乘法之后加法运算 | 1 | x | ||||||
| 最大值 | 1 | x | ||||||
| 最小值 | 1 | x | ||||||
| 赋值 | 1 | x | ||||||
| 从浮点寄存器到地址寄存器(a0)移动数据 | 1 | x | x | |||||
| 乘法 | 1 | x | ||||||
| 无运算 | 1 | x | ||||||
| 单位化 | 3 | x | x | |||||
| 幂运算 | 3 | x | x | |||||
| 倒数 | 1 | x | ||||||
| reapeat | 3 | x | x | |||||
| 从主函数或子程序做return处理 | 1 | x | x | |||||
| 子程序的结束 | 1 | x | ||||||
| 平方根的倒数 | 1 | x | ||||||
| Sign | 3 | x | x | |||||
| Sin和cos | 8 | x | x | |||||
| Less than compare | 1 | x | ||||||
| 减法 | 1 | x | ||||||
| 版本 | 0 | x | ||||||
部分函数细节
mova
语法
mova dst, src
dst必须是地址寄存器，a0
示例
将浮点数赋值到整数寄存器，转换要使用凑整函数
if(dest is an integer register)
{
int intSrc =RoundToNearest(src);
dest = intSrc;
}
else
{
dest = src;
}
在2_x或以上版本，地址寄存器是部分向量，写入任何通道都可以
mova a0.xz, r0
sge
语法
sgedst, src0, src1
逐通道比较 src0与src1，如果src0大于等于src1返回1否则返回0
算法
dest.x = (src0.x>= src1.x) ? 1.0f : 0.0f;
dest.y = (src0.y>= src1.y) ? 1.0f : 0.0f;
dest.z = (src0.z>= src1.z) ? 1.0f : 0.0f;
dest.w = (src0.w>= src1.w) ? 1.0f : 0.0f;
sgn
语法
sgn dst, src0, src1
返回src0的符号
src1, src2为临时寄存器保存中间计算过程，为未定义
算法
for each component in src0
{
if(src0.component < 0)
dest.component= -1;
else
if(src0.component == 0)
dest.component = 0;
else
dest.component = 1;
}
slt
语法
slt dst, src0, src1
与sge相反，逐通道比较src0与src1，如果src0小于src1返回1否则返回0
算法
dest.x = (src0.x < src1.x) ? 1.0f : 0.0f;
dest.y = (src0.y < src1.y) ? 1.0f : 0.0f;
dest.z = (src0.z < src1.z) ? 1.0f : 0.0f;
dest.w = (src0.w < src1.w) ? 1.0f : 0.0f;
-----wolf96 2017/1/3