blog.yui.codes thought dump revolving around art+tech

a lighting/env work in progress

I had this vision for a really long time and had collected all sorts of references over the course of several months. Coming from Singapore, I wanted to do something that was like a greenhouse, but in a solarpunk world. Obviously since I am focusing on lighting here, I decided to rework an existing UE5 map to the best of my abilities rather than overscope and start modeling, texturing and everything.


I also did another version which was a moody, humid abandoned greenhouse vibe.

(read more...)


lessons in lighting

Recently, I took a dive into lighting in UE5 over the course of 10?ish weeks under the guidance of Peter Tran. WARNING: Image heavy post…just for this post, I’m adding post truncation to the site.

First Attempt

For my first delve into studying lighting, I took the old Elemental scene and did two lighting scenes. The goal was primarily to do something that looked different with it and to familiarize myself with the tools available to me in UE5. In particular, I experimented a bit with volumetric lighting and getting some nice light shafts. I also played with the skybox a lot for the first outside image and it was very interesting to try to get a nice sense of depth without losing immersion by manually tweaking blur to emulate depth of field. In this case, I think I would have preferred the mountains to be less clear while not losing too much detail on the clouds, even if that isn’t super physically right I suppose. For the second image, in hindsight, the bottom right corner is a little too dark, but I liked the mood of this scene.

Second Wind

This time, I spent a ton of time brainstorming on what kind of project I could do. I browsed through Artstation and decided to pick my favorite artists and do a top-down lighting like in Dota2 or League of Legends.

(read more...)


notes on energy drift in time integrators

Final Project Idea

Energy drift - usually damping - is substantial for numerical integration schemes that are not symplectic, such as the Runge-Kutta family.

Symplectic integrators usually used in molecular dynamics, such as the Verlet integrator family, exhibit increases in energy over very long time scales, though the error remains roughly constant. These integrators do not in fact reproduce the Hamiltonian mechanics of the system; instead, they reproduce a closely related “shadow” Hamiltonian whose value they conserve many orders of magnitude more closely.

Paper Presentation - FEPR

This week we have paper presentations and I talk about this paper to the class, Fast Energy Projection for Real-time Simulation of Deformable Objects by Dinev, Liu et. al. Reference links are at the bottom.

### Disclaimer Tiantian Liu made a much better video explaining the paper. You can find that in the references as well. My presentation is very much tailored to the class and what we know collectively, which is not that much.

The link to the slides is http://bit.ly/yuifepr

References

Final Project Presentation