To cap or not to cap?

Univox

Hi, I'm a total newbie so I apolgize if my question seems a little ignorant.

With cylindrical objects, is it advantageous from a system perspective to close the object with a cap poly, or is it just one more poly that's not needed?

What I mean is, say you have a 8 sided cylinder with one end hidden in another object. Are these 8 vertices more taxing to render in a game engine than if they were capped? Or does it not make any difference and I'm being ridiculous?

Thanks, I don't want to just create models that look good, I want to try to be as efficient as possible also.

KDR_11k

Unless you have autoLODs, no. A cylinder is only one tristrip, a capped cylinder is probably three strips or something.

malcolm

I agree, a tri strip is rendered differently depending on what console or pc you are working on, the ps2 strips polygons completely different than the xbox, also depending on how you uv map the cylindar you could actually end up with a tri strip for every triangle in your cylindar, as well the way you triangulate the object can cause this too. My favourite rule of thumb is, more polygons always equals more strip verts no matter how you build the model. Please delete any geometry you will never see.