You are seeing the new design of my blog, I launched a new theme for my blog a few days ago and still doing some tweaks to optimize this theme. If you visit my blog...
The color of water is influenced by a very large number of factors, especially by sky color and light conditions, so it’s radically different on a sunny day vs. an overcast day.
From the abstract: Describes realtime physics-based simulation and realistic rendering of rivers using Smoothed Particle Hydrodynamics. Claims their method of generating the surface is far faster than the Marching Cubes. approach, and is well suited for rendering.
The demo video shows very tiny shallow streams of water, which look rather unstable and artificial. The larger area demos (for a valley and flooding) also looks unstable and less realistic than a simple water heightfield.
open-source implementation of the 2D Kass et al. shallow water simulation, with downloadable demo, with accompanying academic paper (pdf)
‘In the simulation we use Jacobi iteration for solving linear systems. For the visualization we use a method based on the refraction and reflection of light on the water surface.’
requires an Nvidia Geforce 6200 or newer (it didn’t run on my ATI Radeon 9800)
thesis project of Claes Johanson, Sweden, building on previous ‘projected grid’ work
includes binaries and DX9 source for a method of using the GPU to create an animated water surface, using “a grid mesh whose vertices are even-spaced, not in world-space, but in post-perspective camera space.”
result is very smooth, with great reflection/refraction appearance
Damien Hinsinger, Fabrice Neyret, Marie-Paule Cani published the concept of a projected grid for realtime water, which restricts computations to the visible part of the ocean surface, adapts the geometric resolution to the viewing distance and only considers the visible waves wavelengths.
demonstrates the realistic rendering of clean and foggy water, Fire and Sky (OpenGL)
A tutorial about the water effect is included
uses GL_ARB_vertex_program and GL_ARB_fragment_program
Joe Dart’s RoamAlaska water demo
an open-source implementation which follows the Tessendorf paper for the basic approach, uses a take-off on Gottfried Chen’s algorithm, and Paul Bourke’s FFT function
Jeff says “This methods simplifies the full navier-stokes fluid flow equations for shallow water. It uses a height map to represent the water surface. I have implemented both the 2D and 3D versions of this system and it is very suitable for realtime performance on the scale I need.”
A software toolkit for digital terrain and river network analysis. Was a free, academic tool, developed by Scott Peckham at University of Colorado, then a commercial product from RSI, then spun off as its own company Rivix LLC. As of 2008, it is $500/$900 a license.
Includes extensive DEM tools, including formats and patching together multiple DEM files.
hydrologic operations including extracting drainage network patterns
RealFlow, “the most sophisticated fluid simulator on the market”, uses a particle-based approach which looks fairly good, but is naturally very non-realtime
Finelight Visual Technology has several commercial and licensable tools and APIs for water simulation and rendering, SeaMonster, WaveTools V1 and V2.
IST Visual Systems Lab (VSL) claimed to have implemented a physically based, realtime hydrology model in 1993
Campbell, Charles. “Fluid Dynamics Methodologies for Computer Graphics”, M.S. Thesis, University of Central Florida, Orlando, Florida, 1991.
Jim Chen. “Physically-Based Modeling and Real-Time Simulation of Fluids”, Ph.D. Dissertation, University of Central Florida, Orlando, Florida, May 1995.
“A freely available 3D-GIS topographic parameterization and hydrologic analysis model. The model automatically delineates watershed and subcatchment boundaries, generalizes geophysical parameters and provides in-depth analysis tools to examine and compare the hydrologic properties of subcatchments.”
might be useful for operating on DEMs as well, e.g. using simulated water erosion to provide detail and accuracy beyond the 30m/10m grid data
Stava, O., Benes, B., Brisbin, M., and Krivanek, O., Siggraph/Eurographics Symposium on Computer Animation 2008 video
Interactive, because the three erosion algorithms (force-based and dissolution-based hydraulic erosion, and direct material transportation through sediment slippage) are GPU-based and run at least 20 fps for scenes with grid resolution of 2048×1024.
describes a simplified model of hydraulic erosion with a heightfield grid
Load Balancing in Parallel Implementation of Hydraulic Erosion Visual Simulations, 2002
shows how the task of simulating erosion can be distributed among several computers (or CPUs) with message-passing between for the state of their boundaries