i wonder how this affects the variance / noise characteristics of a raytraced image. on one hand, it adds an additional dimension to the lighting integral being calculated, and one could imagine that having a whole range of camera rays for each pixel in the image would increase noise since we are adding radiance contributions from many such camera rays (and by extension the variance associated with each one), but on the other hand the blurring effect would possibly make noise less visually problematic?
i wonder how this affects the variance / noise characteristics of a raytraced image. on one hand, it adds an additional dimension to the lighting integral being calculated, and one could imagine that having a whole range of camera rays for each pixel in the image would increase noise since we are adding radiance contributions from many such camera rays (and by extension the variance associated with each one), but on the other hand the blurring effect would possibly make noise less visually problematic?