Planet-sized batched dynamic adaptive meshes (P-BDAM)

Abstract

We describe an efficient technique for out-of-core management and interactive rendering of planet sized textured terrain surfaces. The technique, called P-Batched Dynamic Adaptive Meshes (P-BDAM), extends the BDAM approach by using as basic primitive a general triangulation of points on a displaced triangle. The proposed framework introduces several advances with respect to the state of the art: thanks to a batched host-to-graphics communication model, we outperform current adaptive tessellation solutions in terms of rendering speed; we guarantee overall geometric continuity, exploiting programmable graphics hardware to cope with the accuracy issues introduced by single precision floating points; we exploit a compressed out of core representation and speculative prefetching for hiding disk latency during rendering of out-of-core data; we efficiently construct high quality simplified representations with a novel distributed out of core simplification algorithm working on a standard PC network.

Interactive Out-of-Core Visualisation of Very Large Landscapes on Commodity Graphics Platform

“…We summarize here the main concepts behind BDAM. Please refer to the original papers for further details [1,2].…”

“…In this way, each mesh composed by a collection of small patches arranged as a correct bintree triangulation still generates a globally correct triangulation. This simple edge error property is exploited, as explained in the original paper [2], to design a distributed out-of-core high quality simplification algorithm that concurrently builds all patches.…”

“…We use both a data layout aimed at optimizing memory coherence and a cache managed using a LRU strategy for caching the most recent textures and patches directly in graphics memory. Since the disk is, by far, the slowest component of the system, we have further optimized the external memory management component with mesh compression and speculative prefetching (see [2] for details).…”

Displacement patches for view-dependent rendering

“…Most GPU-based level-of-detail algorithms were designed for height fields and terrain data sets [25][26][27][28][29][30][31]. Little work has been done to deal with general 3D models.…”

Seamless patches for GPU-based terrain rendering

“…Various approaches cache triangulated regions in texture memory [Cig03a,Cig03b,Lar03a,Lev02a], while others exploit the geometric locality to maximize the efficiency of the cache [Hop99a]. Terrains usually compensate small geometric details by textures and as a result, they are often accompanied by huge texture maps.…”