I''ve just set up a program, got my vertex buffers working, everything I have is rendering great. Right now I''ve just got 2 simple objects in it, so before there are hundreds, I''d like to make sure I choose the right method. My question is whether it''s better to transform the vertices myself for each object in the vertex buffer beforehand (this would require a lot of updating in the vertex buffer), or leave the vertex buffer alone, and just screw with the world matrix for each object to achieve a similar effect. My problem with the second method is that it doesn''t seem to me that the drawprimitive calls should be split, which would be kinda necessary in this method....sorry if I''m a bit unclear, but can anyone answer?