Greetings, Unfortunately I have no reply on this message, and I am interested in what are the points of my project ideas I should pay special attention for?
-----Original Message----- From: SilverMax [mailto:[email protected]] Sent: Thursday, April 02, 2009 12:07 AM To: 'CS developers and users list' Subject: Re: [CsMain] GSoC project ideas -----Original Message----- From: Jorrit Tyberghein [mailto:[email protected]] Sent: Monday, March 30, 2009 11:08 PM To: CS developers and users list Subject: Re: [CsMain] GSoC project ideas Can you tell us a bit about these methods from that ACM Symposium article and how you think you might try to implement these in Crystal Space? Greetings, On Mon, Mar 30, 2009 at 7:50 PM, SilverMax <[email protected]> wrote: > Hello, my name is Max Efimov. > > I would like to participate in the project in environmental effects area. I > have 3 year of C++ experience and familiar with some 3D technologies. > Unfortunately I have no experience in programming on CrystalSpace but have > some experience on Irrlicht Engine (I believe it is similar things :). > > I have the following ideas about environmental effects implementation: > > 1. Make a flexible framework (if there is not) to allow user just choose > what he want to do with the environment > > 2. Realise the set of algorithms and methods described by NVIDIA, ATI and > ACM Symposium on Interactive 3D Graphics and Games (I3D) - 2008 > > 3. Test all the methods at maximum abilities. > > I just imagine in common what to do but I have no start point to suggest > some more ideas. Would You like to help me with it? > > ---------------------------------------------------------------------------- -- > > _______________________________________________ > Crystal-main mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/crystal-main > Unsubscribe: > mailto:[email protected]?subject=unsubscribe > > Greetings, I have read the ACM Symposium article "Interactive multiple anisotropic scattering in clouds" and now I have some project ideas listed below: 1. To include a new unit that represents a system of classes managing the calculations and render (a framework). 2. To perform a class that manages shape (by defining a mesh), type, density and other parameters of clouds (a user controlled class). The class manages mesh loading or generating based on heightmap, sets initial parameters, applies a Hypertexture to a mesh and perform other first order actions. Also it is important to use a apply a texture directly to the mesh and another texture to represent as a Hypertexture with a depth of h (cloud's core and boundary respectively). This method allows to apply any fractal real time noise (like scalar 3D Perlin noise) to the Hypertexture. Fractal noise allows to simulate inhomogeneous nature of cloud's boundaries. Because of the noise is quite computational expensive (over 80% of computations) this feature should be optional. 3. To perform an algorithm for collecting rays from a light source and simulate light transport between droplets. We can achieve this by dividing the cloud into 8 diapasons of scattering orders according to supposed droplet size and densities inside the cloud. There are some approaches that have their advantages and disadvantages (computation efficiency and visualisation realism) but the most efficient method is to use phase function Mie for computation a raytrace anisotropic behaviour with different orders of scattering. This method is quite computational expensive but allows to visualise strong features of real clouds like backscattering, sharp boundaries, glow, fogbow. 4. To perform an algorithm that provides to define a collector area. Collector area is a place on a cloud slab where the light rays accumulates. Then all the rays passes throught 8 orfers of different scattering (low-order scattering is single and other are multiple). Then all of the scattered rays are directed to point P (by Transport function) that corresponds to the view area. It allows to visualise scattered rays by sending their parameters to shader that computes an appearance of a cloud. 5. To perform an algorithm that provides real time computation of cloud's position and it's boundary deviation. We can simulate boundary deviation by recomputing the noise of Hypertexture every frame with initial parameters of the current noise state. And cloud's position and velocity depends on the wind's strength. 6. To manage reducing calculations when the cloud node is not in a view and make recalculations if the angle of view or camera position (or target) is changed in the scene (if a cloud node is in a view). 7. If time allows, append an effect of rain and snow. To add to the framework a class that performs placing of area particle emitter with influence of reduced gravity, wind velocity, droplet size distribution (DSD). This allows to simulate a rain or snow behaviour in distant approximation. -- Project Manager of Crystal Space (http://www.crystalspace3d.org) and CEL (http://cel.crystalspace3d.org) Support Crystal Space. Donate at https://sourceforge.net/donate/index.php?group_id=649 Personal page: http://users.telenet.be/jorritTyberghein/ ---------------------------------------------------------------------------- -- _______________________________________________ Crystal-main mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/crystal-main Unsubscribe: mailto:[email protected]?subject=unsubscribe ---------------------------------------------------------------------------- -- _______________________________________________ Crystal-main mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/crystal-main Unsubscribe: mailto:[email protected]?subject=unsubscribe ------------------------------------------------------------------------------ _______________________________________________ Crystal-main mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/crystal-main Unsubscribe: mailto:[email protected]?subject=unsubscribe
