need help with D3DXVec3TransformCoord on second mesh

when i got to use this in render loop , Mesh2 gets the same converted x,y,z as Mesh 1 , both meshs are placed next to each other so i dont know why they get the same values after D3DXVec3TransformCoord
do you have to create a new matrix where you draw the mesh?
http://img136.imageshack.us/my.php?image=shipsbi5.jpg

Code:

---

void CreateBoundingBox2(LPD3DXMESH Mesh2){
 
  if(Mesh2) //if its valid
  {
  BYTE* pVertices=NULL;
  HRESULT hr=Mesh2->LockVertexBuffer(D3DLOCK_READONLY, (LPVOID*)&pVertices);
 
  D3DXComputeBoundingBox((D3DXVECTOR3*)pVertices,
                              Mesh2->GetNumVertices(),
                                                  D3DXGetFVFVertexSize(Mesh2->GetFVF()),
                                                  &minBounds2,
                                                  &maxBounds2);
  Mesh2->UnlockVertexBuffer();
  }

 
//put each corner to worldspace
// We have min and max values, use these to get the 8 corners of the bounding box
m_objectBounds2[0] = D3DXVECTOR3( minBounds2.x, minBounds2.y, minBounds2.z ); // xyz
m_objectBounds2[1] = D3DXVECTOR3( maxBounds2.x, minBounds2.y, minBounds2.z ); // Xyz
m_objectBounds2[2] = D3DXVECTOR3( minBounds2.x, maxBounds2.y, minBounds2.z ); // xYz
m_objectBounds2[3] = D3DXVECTOR3( maxBounds2.x, maxBounds2.y, minBounds2.z ); // XYz
m_objectBounds2[4] = D3DXVECTOR3( minBounds2.x, minBounds2.y, maxBounds2.z ); // xyZ
m_objectBounds2[5] = D3DXVECTOR3( maxBounds2.x, minBounds2.y, maxBounds2.z ); // XyZ
m_objectBounds2[6] = D3DXVECTOR3( minBounds2.x, maxBounds2.y, maxBounds2.z ); // 
m_objectBounds2[7] = D3DXVECTOR3( maxBounds2.x, maxBounds2.y, maxBounds2.z ); // XYZ

// Transform the 8 corners of our object space bounding box into world space
//sprintf(data , "%s %f\n" , "m_objectBounds[0].x", m_objectBounds[0].x);
//WriteConsole(Console,data, sizeof(data) , &written, NULL);

//convert
for( int i = 0; i < 8; i++ ){
  D3DXVec3TransformCoord( &worldBounds2[i], &m_objectBounds2[i], &matTrans );
}





Sleep(50);
}

---

Why are you doing this manually?

The correct way to compute bounding boxes is to transform and then re-compute the bounding box. Otherwise you will end up with some bounding volumes that are not accurate. You cannot create the bounding volume and then just transform it.

D3DXComputeBoundingBox() might interest you.

right so now in BeginScene i have

Code:

---

D3DXMatrixTranslation(&matTrans , Left , 0.0f , Forward);
        d3ddev->SetTransform(D3DTS_WORLD, &(matTrans));
    CreateBoundingBox(meshSpaceship);
        d3ddev->SetRenderState(D3DRS_AMBIENT, D3DCOLOR_XRGB(50, 0, 0));
       


    for(DWORD i = 0; i < numMaterials; i++)    // loop through each subset
    {
        d3ddev->SetMaterial(&material[i]);    // set the material for the subset
        meshSpaceship->DrawSubset(i);    // draw the subset
               
    }

---

Uhh in the code above (first post), you were writing the values before transforming them. Don't be offended it that isn't the problem (but gotta get the obvious stuff out of the way).

As Bob said your previous code could have worked but in the end your algo would not have worked. Due to floating point roundoff and all of the transformations you would have put your original bounding volume through your system would not have worked.

Transform then recreate the bounding volume. I recommend you render your volumes so you will see what begins to happen after several transforms.

i am now struggling to figure out how the collisions is done ,
ive got 2 meshs loaded to screen and have there Bounding Min Max and converted to world space , is it something like if(WorldBounds[0].x > WorldBounds2[0].x)