Why are shadowmap frustums so large in spotlights?

When I visualize shadowmap frustums in the editor, I see they are HUGE, especially when compared to the related spotlight’s attenuation radius. The problem is not their width or aperture, they fit quite tightly around the attenuation radius volume, but the depth. I would expect the frustums to set their zfar at the attenuation radius volume far end, but they continue for thousands of unreal units, I’d say they are 3 to 5 times as deep as their related attenuation volume.

Since this is including tons of geometry that is never going to cast a shadow it can have a great performance impact. Is there a reason for this that I’m unaware of or is it a bug?

Hi Gatherer -

The Shadow Frustum that is drawn in when you enable visualization is the region that the light’s shadowmap covers, but the actual culling of shadow casters is calculated by the Cone.

Thank You

Eric Ketchum