|
| 1 | +### Mardini 2025 - Desert Dunes |
| 2 | + |
| 3 | +Today is day 29 of [Mardini 2025][mardini-2025]. The prompt for today is "SUNSET", |
| 4 | +in the LOPS category. |
| 5 | + |
| 6 | +You can do a sunset scene over just about anything, and chances are it'll look decent. |
| 7 | +Funnily enough, this is actually the second time making a desert scene, although [the |
| 8 | +first time][day-62] I was following along with a tutorial and specifically learning how |
| 9 | +to use heightfields and the old COPS context. This time I focused less on both of those |
| 10 | +(although I did use them to generate the terrain and textures) but I focused most of my |
| 11 | +energy on: |
| 12 | + |
| 13 | + 1. A noise- and trigonometry-based sand ripples (the few inches tall wind-driven |
| 14 | + ripples.) |
| 15 | + 2. Running another MPM simulation for the footsteps in the sand with (sorta) realistic |
| 16 | + sand collapsing/resetting. |
| 17 | + 3. The render and postprocessing. |
| 18 | + |
| 19 | +New stuff: |
| 20 | + |
| 21 | + - Another attempt at MPM, this time with the "sand" material type. I also was a bit |
| 22 | + smarter, and chose a project that doesn't require perticularly good results, since |
| 23 | + I'm just taking the end result and flattening it to the topmost layer of sand |
| 24 | + so I can update the original mesh and attempt to blend between the simulation and |
| 25 | + original mesh height. Even if the results are a little messy, the blending and |
| 26 | + a moderate blur should fix the issues. And it turned out all right, although |
| 27 | + you can definitely still see where the blending begins. |
| 28 | + - A lot of postprocessing in the slap comp, in ways I've definitely never used it |
| 29 | + before. HSV adjust, Chroma key, selective blur, selective glow, |
| 30 | + 3x3 convolve, etc. I also used a slope direction + streak blur COP for the first |
| 31 | + time to generate the sand texture, although it's a very subtle effect and not |
| 32 | + visible in the final render. I found that my slap comp behaves significantly |
| 33 | + differently when I render 1080p versus 4k. Presumably other render sizes would |
| 34 | + have their own strange render characteristics. At some point I'll look into which |
| 35 | + specific nodes are causing the discrepancy, but not today. I'm beat. |
| 36 | + |
| 37 | +My entry post is [here][entry-post]. |
| 38 | + |
| 39 | +[mardini-2025]: https://www.sidefx.com/community-main-menu/contests-jams/mardini-2025/ |
| 40 | +[day-62]: https://brandonslade.me/houdini-practice/62 |
| 41 | +[entry-post]: https://www.sidefx.com/forum/topic/100340/?page=1#post-442097 |
0 commit comments