Novel method for asynchronous pipeline topographic drawing

Abstract

The invention discloses a novel method for asynchronous pipeline topographic drawing and belongs to the technical field of digital topographic drawing. Drawing pipelines of the method sequentially comprise three steps of data preparation, CPU operation and GPU operation, wherein the method comprises three parallel drawing pipelines at the same time; and the three pipelines perform different steps at the same time. The data preparation step comprises requiring data of a local disk or network, receiving data and organizing data. The CUP operation step comprises building a tree and performing geometrically gridding operation. The method can be used for topographic drawing and has the advantages of quickness, delay elimination, smooth frame rate, controllability and the like.

Description

technical field [0001] The invention relates to terrain rendering, in particular to a new asynchronous pipeline terrain rendering method, which belongs to the technical field of digital terrain rendering. Background technique [0002] Traditional terrain rendering algorithms (such as ROAM or other algorithms) are usually synchronous pipeline terrain rendering methods. The rendering pipeline, or a complete rendering process of one frame, goes through the following steps in sequence: [0003] Prepare data→build tree→grid→draw [0004] For the above-mentioned first step - data preparation, the terrain rendering system with a general fixed area size does not need to be carried out, but for massive terrain data, since the terrain area drawn is not fixed, it should be obtained from the data source (local or network) query New data, but due to reasons such as disk I / O delay and network delay, it may take a long time to wait for the data to be ready, which will slow down the frame ...

AI Technical Summary

Problems solved by technology

However, for the huge amount of terrain data that cannot be loaded into memory at one time and wider network applications, the currently requested data (elevation data or texture data) may not be available temporarily, making it necessary to wait before building trees, and the speed is natural will come down

Although some strategic processing can be done, such as replacing the data with a default value before it arrives, this improves the performance but loses the image quality, because there may be many places that are flat, but suddenly there are unfavorable situations such as mountains.

[0006] In addition, the time consumed by tree building and geometric meshing depends on the size of the terrain created, the recursive level and the number of triangles. It is difficult for existing terrain algorithms to control the number of triangles per frame, let alone control the tree building time per frame. With the frame rate, even an adaptive algorithm like ROAM, which uses dynamic splitting and merging to control the number of triangles, it is difficult to make the frame rate completely stable

For large-scale terrain and high-precision and complex grids, the process of tree building and gridding may be very long, which will increase the waiting time of subsequent steps and subsequent drawing pipelines.

However, if the mesh complexity is reduced, or the terrain scale is reduced, there will be a loss of display quality

Images