@ThisIsTian

PhD excited about Real-time rendering, Game and Machine/Deep Learning @UMBC. Rendering Programmer @EpicGames. Opinions are my own.

NC, USA
Joined December 2015
Thin Geometry Detection has been quietly keeping Nanite foliage stable under TSR since it shipped - fixing the "boiling" on thin geometry, auto-enabled with Nanite foliage. Finally wrote up how it works. @UnrealEngine
3
14
70
6,469
Tiantian Xie retweeted
We’re publishing a deeper technical report on why generation is needed for the next leap toward photorealism. VRAM and compute limit both the scene abstraction and the number of rays we can afford. Previous DLSS reconstructs from that representation—so even perfect reconstruction is still bounded by what the scene contains. DLSS 5 is the first DLSS to generate beyond that abstraction and break through that ceiling, while staying grounded in the game and controlled by the artist. We also compare it with offline generative models and renderers at similar scaling cost. research.nvidia.com/labs/adl…
28
41
16
404
126,184
Tiantian Xie retweeted
I worked on Lumen Lite which is now out in UE 5.8. It's based on Irradiance Fields with Probe Occlusion, but with automatic probe placement. It is used when the engine is set to Medium GI and Reflections quality. dev.epicgames.com/documentat…
28
56
1
474
30,330
Heads-up: if you use Nanite foliage with Before-DoF translucency (glass, some VFX), those need special handling through the Temporal Responsiveness node or they will create artifacts over the foliage. Prefer After-DoF when possible, as TSR handles it automatically.
1
5
533
Tiantian Xie retweeted
Thanks for tuning in to The Witcher 4 — Unreal Engine 5 tech demo and joining us during the State of Unreal 2025! #UnrealFest The full presentation will be uploaded and shared with everyone later today, but in the meantime, here's four screenshots straight from the demo to tide us all over. 😍
1,276
8,336
2,367
58,531
5,533,096
We want Unreal Engine 5.6 to scale to large and rich worlds while preserving graphic quality and delivering 60 FPS. What better world to do this in than that of The Witcher? Tap in: epic.gm/state-of-unreal-live…
86
265
20
2,725
134,857
Tiantian Xie retweeted
Alan Wolfe, William Donnelly, and Henrik Halén, Importance-Sampled Filter-Adapted Spatio-Temporal Sampling jcgt.org/published/0014/01/0…
20
54
4,343
I'm excited to share our latest paper (work with Matt Pharr and @inversepixel ), "Improved Stochastic Texture Filtering Through Sample Reuse" accepted to I3D'2025. arxiv.org/abs/2504.05562 research.nvidia.com/labs/rtr… 1/
4
51
1
210
13,968
Tiantian Xie retweeted
Surely you have heard of Perlin, right? This year, he will come to I3D to give us his presentation "AI and the coming XR Singularity". More info at i3dsymposium.org/2025/keynot… Don't miss it! Register now for the conference: i3dsymposium.org/2025/regist… #I3D2025
1
3
304
Tiantian Xie retweeted
Portsmouth, Kutz, and Hill, EON: A practical energy-preserving rough diffuse BRDF. jcgt.org/published/0014/01/0…
44
1
124
10,096
Tiantian Xie retweeted
I just pushed a new paper to arXiv. I realized that a lot of my previous work on robust losses and nerf-y things was dancing around something simpler: a slight tweak to the classic Box-Cox power transform that makes it much more useful and stable. It's this f(x, λ) here:
34
258
16
2,134
236,401
New dithering method dropped I call it Surface-Stable Fractal Dithering and I've released it as open source along with this explainer video of how it works. Explainer video: youtube.com/watch?v=HPqGaIMV… Source repository: github.com/runevision/Dither… #gamedev #vfx
59
296
17
2,312
91,167
My new blog post explains spectral radiometric quantities, photometry and basics of spectral rendering. This is part 2/2 in a series on radiometry. Learn what it means when a light bulb has 800 lumen and how your renderer can account for that. momentsingraphics.de/Radiome…
3
57
1
215
10,770
Oldies but goldies: J Allen, Short Time Spectral Analysis, Synthesis, and Modification by Discrete Fourier Transform, 1977. Popularized the use of the spectrogram as the baseline tool for audio processing. en.wikipedia.org/wiki/Spectr… en.wikipedia.org/wiki/Short-…
2
39
1
194
11,797
Re-writing a non-convex optimization as a difference of convex (DC) leads to a simple iterative optimization method. Sometimes referred to as "convex-concave procedure" in the ML literature. lita.univ-lorraine.fr/~lethi… ceremade.dauphine.fr/~carlie…
1
66
2
467
48,466
Classic 3D videogame shadow techniques 30fps.net/pages/videogame-sh…
3
27
105
6,855
Tiantian Xie retweeted
Max, Efficient Motion Blurred Spheres Using Texture Mapping. jcgt.org/published/0013/02/0…
6
20
2,135