works fine windows 7 X64
Nope, it crashes with nVidia for me on either OS
edit:
also, these curious functions
void shittify(uint4 B[8])
{
uint4 tmp[4];
tmp[0] = (uint4)(B[1].x,B[2].y,B[3].z,B[0].w);
tmp[1] = (uint4)(B[2].x,B[3].y,B[0].z,B[1].w);
tmp[2] = (uint4)(B[3].x,B[0].y,B[1].z,B[2].w);
tmp[3] = (uint4)(B[0].x,B[1].y,B[2].z,B[3].w);
#pragma unroll
for(uint i=0; i<4; ++i)
B[$i] = EndianSwap4(tmp[$i]);
tmp[0] = (uint4)(B[5].x,B[6].y,B[7].z,B[4].w);
tmp[1] = (uint4)(B[6].x,B[7].y,B[4].z,B[5].w);
tmp[2] = (uint4)(B[7].x,B[4].y,B[5].z,B[6].w);
tmp[3] = (uint4)(B[4].x,B[5].y,B[6].z,B[7].w);
#pragma unroll
for(uint i=0; i<4; ++i)
B[i+4] = EndianSwap4(tmp[$i]);
}
void unshittify(uint4 B[8])
{
uint4 tmp[4];
tmp[0] = (uint4)(B[3].x,B[2].y,B[1].z,B[0].w);
tmp[1] = (uint4)(B[0].x,B[3].y,B[2].z,B[1].w);
tmp[2] = (uint4)(B[1].x,B[0].y,B[3].z,B[2].w);
tmp[3] = (uint4)(B[2].x,B[1].y,B[0].z,B[3].w);
#pragma unroll
for(uint i=0; i<4; ++i)
B[$i] = EndianSwap4(tmp[$i]);
tmp[0] = (uint4)(B[7].x,B[6].y,B[5].z,B[4].w);
tmp[1] = (uint4)(B[4].x,B[7].y,B[6].z,B[5].w);
tmp[2] = (uint4)(B[5].x,B[4].y,B[7].z,B[6].w);
tmp[3] = (uint4)(B[6].x,B[5].y,B[4].z,B[7].w);
#pragma unroll
for(uint i=0; i<4; ++i)
B[i+4] = EndianSwap4(tmp[$i]);
}