I'm having trouble trying to implement adaptive subdivision in SceneKit. The results I'm expecting should be similar to what's seen in the 2017 SceneKit demo for WWDC of the Fox mesh. I would like it to work in ARKit. Do I have to specify my pointOfView.camera as any factor in the process of initializing the subdivision, or do I have to calculate the length away from the camera to model at all times adjusting my subdivisionLevel in a SceneKit renderer delegate method?
let tessellator = SCNGeometryTessellator()
tessellator.edgeTessellationFactor = 3.0
tessellator.insideTessellationFactor = 3.0
newNode.geometry?.tessellator = tessellator
tessellator.smoothingMode = .pnTriangles
tessellator.isAdaptive = true
tessellator.maximumEdgeLength = 50
modelNode.geometry?.wantsAdaptiveSubdivision = true
modelNode.geometry?.subdivisionLevel = 1
See Apple documentation or the video (at 23:00).