Game Development Reference
In-Depth Information
q14 = a*d;
q22 = b*b;
q23 = b*c;
q24 = b*d;
q33 = c*c;
q34 = c*d;
q44 = d*d;
valid = TRUE;
v.x = 0.0; v.y = 0.0; v.z = 0.0;
deleted = FALSE;
return TRUE;
}
To return the quadric error metric of a pair, we take the sum of the vertex
quadrics and multiply by the contraction target and its transpose. This
utility function assumes that the sum of the vertex quadrics is stored in the
member variables. We pass the contraction target as a VECTOR
reference.
double CQuadric::QEM(VECTOR &v)
{
double vx, vy, vz;
vx = q11*v.x + q12*v.y + q13*v.z + q14;
vy = q12*v.x + q22*v.y + q23*v.z + q24;
vz = q13*v.x + q23*v.y + q33*v.z + q34;
return (vx*v.x + vy*v.y + vz*v.z + 1.0);
}
When creating an array of valid pairs we first assign every edge. This is
done by iterating through each polygon and adding each edge of a polygon
to the valid pairs array. If the pair is already present in the array then the
algorithm moves on without adding this pair. If the pair is not present then
the current pair is added to the valid pairs array. After adding the pair a
counter qpTotal is incremented. If this value exceeds the array size, which
is stored in the variable qpMax , then the array is doubled in size and the
original array copied to the new memory allocation. This dynamic allocation
of memory can provide a useful technique for array expansion. Valid pairs
include vertices whose distance apart is within a certain boundary. One of
the parameters used for the function CreateValidPairsArray is distance .
Search Nedrilad ::




Custom Search