Forward+ decal rendering [wayback-archive]
- goals: avoid additional geometry, and don’t increase draw call count
- extend the light structure to also support decals
- cull the decals, sort them and apply them in the object shader just as lights
- still has a few open problems, for example mip selection
- possible solution: calculate the UV gradients manually as in this example
Depth proxy transparency rendering [wayback-archive]
- Depth proxy: 2 channel textures that contains min and max of transparent entity
- using this structure for various applications
- Particles Self-Shadowing
- Volumetric Effects-Particles Sorting
- Particles-Transparent Meshes Sorting
Little Lightmap Tricks [wayback-archive]
- no pixel gaps are required between charts, if they are required then the UV calculation is wrong
- if all pixels in a chart are similar, can squash them down to a single pixel and point the UVs to it
- deduplicate nearly identical charts
- store similar colors for unused pixels, increase DXT quality
- explanation of the scheme used for the wii version of Call Of Duty: Modern Warfare
It’s All About The Data [wayback-archive]
- how to optimize data transformation from API agnostic command buffers to API specific calls
- think about data layout
- combine state settings into few and bigger commands
Unreal Dev Day Montreal Presentations Released [wayback-archive]
- UE4 Performance and Profiling
- Paragon Character Texturing Pipeline
- Creating Complex In-Game Effects
- Lighting with Unreal Engine Masterclass
- Fortnite Trailer Pipeline
Aftermath 1.3 Update [wayback-archive]
- no need to rename the .exe anymore
- can return resource information
- reduced overhead (memory and CPU performance)
- samples released by the Xbox Advanced Technology Group
- this includes source code for Xbox One samples
Designing a Next-Gen GPU API for the Web
- API to supersede WebGL
- abstraction on Vulkan, Metal, D3D12 but embracing web constraints
- comparison of similar concepts in all 3 APIs
Design Patterns for Low-Level Real-Time Rendering - Video
- was discussed in Issue 10. Now the video is released
Register Cache: Caching for Warp-Centric CUDA Programs [wayback-archive]
- software abstraction that uses inter-thread shuffling to replace shared memory with thread register usage
- use case: replace shared memory with registers to cache kernel inputs
Dealing with geometry in Vulkan [wayback-archive]
- how to upload data into buffers for dynamic and static cases
