Tools
Search
Import
Library
Explore
Videos
Channels
Figures
Atmrix
About
Tools
Search
Import
Library
Explore
Videos
Channels
Figures
Atmrix
About
Go Back
TS
Tim Sweeney
04/30/25
@ Lex Fridman
If you render two triangles per pixel, the image can be indistinguishable from one rendered with thousands of triangles per pixel, but rendering less than that will show visible artifacts.
Video
LF
Tim Sweeney: Fortnite, Unreal Engine, and the Future of Gaming | Lex Fridman Podcast #467
@ Lex Fridman
04/30/25
Related Takeaways
TS
Tim Sweeney
04/30/25
@ Lex Fridman
Rendering a perfect single frame is easier than rendering a series of frames in motion that look perfect, as earlier algorithms often resulted in visible transitions between different triangle counts.
TS
Tim Sweeney
04/30/25
@ Lex Fridman
Rendering objects at extreme close-up and far away distances requires generating the right amount of geometry to be indistinguishable from reality without excess.
TS
Tim Sweeney
04/30/25
@ Lex Fridman
The Nyquist limit states that to render a texture at a certain resolution, you never need more than twice the pixels than in the texture displayed on the screen.
TS
Tim Sweeney
04/30/25
@ Lex Fridman
One breakthrough in graphics was designing a pipeline that avoids the triangle rasterization hardware in GPUs, allowing for better results through pixel shader calculations and ray tracing.
TS
Tim Sweeney
04/30/25
@ Lex Fridman
The optimization of rendering in Unreal Engine was a significant challenge, focusing on reducing the cost of rendering pixels to improve performance.
TS
Tim Sweeney
04/30/25
@ Lex Fridman
Nanite technology in Unreal Engine minimizes visible transitions in geometry, making them statistically present but not perceptible to the viewer.
TS
Tim Sweeney
04/30/25
@ Lex Fridman
Getting human faces right in computer graphics involves the interplay of numerous systems; if any aspect is incorrect, it can lead to an uncanny valley effect, making the representation feel off.
TS
Tim Sweeney
04/30/25
@ Lex Fridman
Rendering hair in computer graphics is complex because accurately simulating every strand would be computationally prohibitive; instead, approximations are used to capture the overall effect of hair.
TS
Tim Sweeney
04/30/25
@ Lex Fridman
The human eye's focus and attention mechanism determines how much detail should be shown in a scene, requiring high resolution for areas of focus while allowing less detail elsewhere.