On Tuesday, 14 January 2020 04:47:32 UTC, Infinite Reality wrote: > > What is the principle of Depth Partition? >
Depth buffer precision is dependent upon the near/far distance ratio, the lower this ratio the lower the precision and more z fighting you have - where pixels of polygons that are close together start to be rendered in the wrong order, so pixels from polygons that are further away get rendered instead of ones that are slightly nearer. The depth partition technique breaks the view into distinct partitions, this might be just two partitions or many, usually two is enough. The partitions abut so the far distance of near partition equals the near distances of the far partition. The partitions are rendered before the nearer ones with the depth buffer being cleared before the next nearest partition is rendered, the colour buffer is kept for each pass. The distance of the junction between the two partitions is not the mid distance between the overall near/far, but is computed to keep the near/far ratios and hence precision the same for each partition. The OSG's depth partition class computes this distance for you. One alternate to depth partitioning is to use a non linear depth buffer as suggested by Maxim. This requires custom shaders to work, but as this is normal for modern applications anyway this isn't much extra work. Have a look online for more details on these approaches. -- You received this message because you are subscribed to the Google Groups "OpenSceneGraph Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to osg-users+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/osg-users/a6956422-79a1-4ed8-863c-71da3d1811d5%40googlegroups.com.