| Microsoft DirectX 8.1 (pixel shader versions 1.0, 1.1, 1.2, 1.3, 1.4) | 
Conditionally chooses between src1 and src2, based on the comparison src0 > 0.5.
cnd dest, src0, src1, src2
| Argument | Description | Registers | Version | |||
|---|---|---|---|---|---|---|
| vn | cn | tn | rn | |||
| dest | Destination register | x | 1.0 | |||
| x | x | 1.1, 1.2, 1.3 | ||||
| x | 1.4 | |||||
| src0 | Source register | r0.a | 1.0, 1.1, 1.2, 1.3 | |||
| src1, src2 | x | x | x | x | 1.0, 1.1, 1.2, 1.3 | |
| src0, src1, src2 | x | x | 1.4 phase 1 | x | x | x | 1.4 phase 2 | 
To learn more about registers, see Registers.
For versions 1.0 to 1.3, src0 must be r0.a. Version 1.4 has no such restriction.
// Version 1.1 to 1.3
if (r0.a > 0.5)
  dest = src1
else
  dest = src2
// Version 1.4 compares each channel separately.
for each component in src0
{
   if (src0.component > 0.5)
     dest.component = src1.component
   else
     dest.component = src2.component
}
These examples show a four-channel comparison done in a version 1.4 shader, as well as a single-channel comparison possible in a version 1.1 shader.
// Version 1.4 compares all four components. ps.1.4 def c0, -0.5, 0.5, 0, 0.6 def c1 0,0,0,0 def c2 1,1,1,1 cnd r1, c0, c1, c2 // r0 contains 1,1,1,0 because, // r1.x = c2.x because c0.x ≤ 0.5 // r1.y = c2.y because c0.y ≤ 0.5 // r1.z = c2.z because c0.z ≤ 0.5 // r1.w = c1.w because c0.w > 0.5 // Version 1.1 to 1.3 compares against the replicated alpha channel // of r0 only. ps.1.1 def c0, -0.5, 0.5, 0, 0.6 def c1 0,0,0,0 def c2 1,1,1,1 mov r0, c0 cnd r1, r0.a, c1, c2 // r1 gets assigned 0,0,0,0 because // r0.a > 0.5, therefore r1.xyzw = c1.xyzw
// This example compares two values, A and B, to each other. // This example assumes A is loaded into v0 and B is loaded into v1. // Both A and B must be in the range of -1 to +1, and since the // color registers (vn) are defined to be between 0 and 1, // the restriction happens to be satisfied in this example. // The shader is shown below. ps.1.0 // version instruction sub r0, v0, v1_bias // r0 = A - (B - 0.5) cnd r0, r0.a, c0, c1 // r0 = ( A > B ? c0 : c1 ) // The result in r0 is c0 if A > B. Otherwise, the result in r0 is c1.