The All New FractalForums is now in Public Beta Testing! Visit FractalForums.org and check it out!
|
|
Fractal coast | ||||||
Previous Image | Next Image | ||||||
Description: The "water" is a simple modelling using a Fresnel model and waves. Tint / transparency and wave period depends slightly of the underwater topology. https://www.facebook.com/pages/Materia-Numerica/230106587144790 http://www.materianumerica.net Stats: Total Favorities: 0 View Who Favorited Filesize: 347.99kB Height: 750 Width: 1500 Discussion Topic: View Topic Keywords: Perlin juliabox rocks water coast glsl Posted by: Chaos_Ink September 04, 2014, 01:12:38 PM Rating: by 3 members. Image Linking Codes
|
||||||
0 Members and 1 Guest are viewing this picture. |
|
Comments (7) | |
Chillheimer | June 11, 2015, 08:35:48 PM alexl: I use 2D gaussian-blur (in two passes) as a post-processing. (please use "quote comment instead of "report comment" in the future. reporting is to complain about a comment to a moderator) |
Chaos_Ink | September 10, 2014, 08:03:11 PM Sorry for the misunderstanding. I have a background in physics, not in computer graphics, therefore by the expression "by hand" I meant anything that is not derived from the ground model (here Snell-Fresnel). As I wrote, blur reflections could be modeled by an integration (or a summation if you prefer) over a (wide) cone of slightly displaced observer vectors. I think it may lead to a different result from random reflected vectors. Anyway water has so many subjective aspects. The blur effect has also a secondary effect of providing a pseudo-physical scale since this blurring occurs at larger distance to the observer. In addition I'm not sure I want to model a pseudo-nature since this "fractal" modelling leads to interesting structures by themselves. |
alexl | September 10, 2014, 06:18:42 PM "by hand"? Do you think that each frame from 14 minute video has been processed in Photoshop by hand? Oh, no! It is just one pass of rendering (render to texture -> blur filter ->render to texture). I tried to apply the mathematical correct modeling of reflections (with random small waves), but it looks worse than a simple blur. Here are few samples of generative water (only gaussian blur): http://www.sanbasestudio.com/art/img/ex1425.jpg http://www.sanbasestudio.com/art/img/ex1440.jpg http://www.sanbasestudio.com/art/img/ex1511.jpg |
Chaos_Ink | September 07, 2014, 08:09:03 AM Oki! Yes, artistic blur car be achieved with gaussians but in truth I'm not so fond of "by hand" post-processing. After a little thinking, blur reflections come from an uneven surface, tiny bumps, particles, etc that perturb the reflected vector. This scattering can be modeled, I guess, by a type of Monte-Carlo integration over the observer vector. This means the computation of many reflected rays within the observer cone. This adds some calculation time of course but perhaps I'll try it later. Anyhow the Snell-Fresnel model apply! Have a nice day! |
alexl | September 06, 2014, 04:15:57 PM I use 2D gaussian-blur (in two passes) as a post-processing. |
Chaos_Ink | September 05, 2014, 08:24:25 PM Thank you for your comment! Where does this blur effect come from? Is it from a kind of particle diffusion or ? For now I prefer to keep to a (simple) Snell-Fresnel modelling I have some understanding of. Water is in truth very complex to model and can display many different subjective aspects. |
alexl | September 05, 2014, 07:21:06 PM Just a tip: the reflection in the water should be slightly blur. Otherwise, the water looks like a glass. Sample: |
Powered by SMF Gallery Pro