confusing triangle winding in terrain patches rendering

[ghosthunter]

Recently I have studied the source code of pyrogenesis engine. And there're few confusing implementation.

For example in the function "CPatchRData::BuildIndices" I can see indices in clockwise order - [0,0] -> [1,0] -> [0,1] -> [1,0] -> [1,1] -> [0,1]

        for (ssize_t j = 0; j < PATCH_SIZE; j++)

        {

            for (ssize_t i = 0; i < PATCH_SIZE; i++)

            {

                if (texgrid[j][i] == tex)

                {

                    bool dir = terrain->GetTriangulationDir(px+i, pz+j);

                    if (dir)

                    {

                        indices.push_back(u16(((j+0)*vsize+(i+0))+base));

                        indices.push_back(u16(((j+0)*vsize+(i+1))+base));

                        indices.push_back(u16(((j+1)*vsize+(i+0))+base));

 

                        indices.push_back(u16(((j+0)*vsize+(i+1))+base));

                        indices.push_back(u16(((j+1)*vsize+(i+1))+base));

                        indices.push_back(u16(((j+1)*vsize+(i+0))+base));

                    }

                    else

                    {

                        indices.push_back(u16(((j+0)*vsize+(i+0))+base));

                        indices.push_back(u16(((j+0)*vsize+(i+1))+base));

                        indices.push_back(u16(((j+1)*vsize+(i+1))+base));

 

                        indices.push_back(u16(((j+1)*vsize+(i+1))+base));

                        indices.push_back(u16(((j+1)*vsize+(i+0))+base));

                        indices.push_back(u16(((j+0)*vsize+(i+0))+base));

                    }

                }

            }

        }

However the default triangle winding of OpenGL is counter-clockwise. Anyway in the Atlas editor the terrain is stil visible with front faces up. 

I wonder whether such implementation is rectified with projection matrix and orientation matrix, Or this is just my misunderstanding in some point?

Hope that some expert can provide help. I really appreciate that!

 

[vladislavbelov]

Hi! And sorry for the long reply.

On 10/8/2026 at 1:59 PM, ghosthunter said:

I wonder whether such implementation is rectified with projection matrix and orientation matrix

Yes, it's in sync with the projection matrix and culling mode.