I am building a rendering engine, and need to take advantage of multiple vertex streams. I have two vertex buffers, one with x and z coords, and one with y(height) and normal data. My two vertex formts are struct xGridVertex { D3DXVECTOR2 pos; //XZ coords FLOAT tu, tv; }; #define D3DFVF_GRIDVERTEX (D3DDECLTYPE_FLOAT2|D3DFVF_TEX1) struct xSectionVertex { FLOAT height; // Y coord D3DXVECTOR3 normal; }; #define D3DFVF_SECTIONVERTEX (D3DDECLTYPE_FLOAT1|D3DFVF_NORMAL) and my D3DVERTEXDECLARATION9 is static D3DVERTEXELEMENT9 vertex_description[] = { {0, 0, D3DDECLTYPE_FLOAT2, D3DDECLMETHOD_DEFAULT, D3DDECLUSAGE_POSITION, 0}, {0, 8, D3DDECLTYPE_FLOAT2, D3DDECLMETHOD_DEFAULT, D3DDECLUSAGE_TEXCOORD, 0}, {1, 0, D3DDECLTYPE_FLOAT1, D3DDECLMETHOD_DEFAULT, D3DDECLUSAGE_POSITION, 1}, {1, 4, D3DDECLTYPE_FLOAT3, D3DDECLMETHOD_DEFAULT, D3DDECLUSAGE_NORMAL, 0}, D3DDECL_END() }; Now my main question is... How do I instruct the pipeline to convert the xy data in stream 0 into xz data, and then combine the solitary height value in stream 1 as the y component? Secondly, is this the correct way to render such a situation? hr = pDevice->SetVertexDeclaration(vertex_dec); hr = pDevice->SetStreamSource(0, vb_grid, 0, sizeof(xGridVertex)); hr = pDevice->SetStreamSource(1, vb_heights, 0, sizeof(xSectionVertex)); hr = pDevice->SetIndices(ib_section_grid); D3DXMATRIXA16 matWorld; D3DXMatrixIdentity(&matWorld); pDevice->SetTransform(D3DTS_WORLD, &matWorld); hr = pDevice->DrawIndexedPrimitive(D3DPT_TRIANGLESTRIP, 0, 0, 32*32, 0, 32*32*2);

Quote:Original post by Zipster
I believe you'll need to write a simple vertex shader in order to accomplish what you want. At least, if you have any special usages (as you do with a second position element).

Thank you very much, I had a feeling that might be the solution, but I was trying the old fixed function method.

Vertex shader will give me more flexibility anyway. Thanx.