Monday 2/27/17 - Room 2006, West Hall

  • 10:00AM - 11:00AM | Advanced Graphics Tech: D3D12 and Vulkan done right & wave programming in D3D12 and Vulkan
    D3D12 and Vulkan Done Right:
    Gareth Thomas (AMD) will discuss how with great power comes great responsibility and it is down to the game developer to feed the GPU correctly on explicit APIs. This talk focuses on key areas which AMD and NVIDIA have both found to require careful consideration.
    Wave Programming in D3D12:
    David Lively (AMD) and Holger Gruen (NVIDIA) will provide background on the new SIMD-level programming constructs supported in D3D and Vulkan. This talk will provide insight on managing portability given variations of SIMD-width between vendors, and diving into some practical cases for shader optimization.

  • 11:20AM - 12:20PM | Advanced Graphics Tech: How to Thrive on the Bleeding Edge Whilst Avoiding Death by 1,000 Paper Cuts
    A DirectX 12 development story told by Jurjen Katsman CEO of Nixxes, PC developers of AAA games 'Deus Ex: Mankind Divided' and 'Rise of the Tomb Raider'. During this talk, Jurjen will share the expectations set, challenges faced and lessons learned by his teams during their vast experience with DirectX 12 development.

  • 1:20PM - 2:20PM | Advanced Graphics Tech: "Async Compute: Deep Dive" & "Raster Ordered Views and Conservative Rasterization"
    Async Compute: Deep Dive:
    Alex Dunn (NVIDIA) and Stephan Hodes (AMD) will discuss how Asynchronous Compute is an important feature of new APIs and an essential know-how for developers using these APIs. In this talk, They'll give a wide range of advice, from high to low level, including case studies, live examples, and overview of available tools, to make a game engine benefit from asynchronous compute.
    Raster Ordered Views and Conservative Rasterization:
    Rahul Sathe (NVIDIA) and Evgeny Makarov (NVIDIA) will discuss how conventional MSAA rasterization can result in a lot of overhead throughout the rendering pipeline. They show how hardware MSAA can be combined with ROVs instead of conventional render targets during rasterization for better AA at reduced cost. They also demonstrate a technique that uses conservative rasterization to place the sample(s) in a fully programmable way. They use raster order views to ensure that pixels along the shared edges of the triangles generated by clipper are handled correctly.

  • 2:40PM - 3:40PM | Advanced Graphics Tech: Moving to DirectX 12: Lessons Learned

    This talk will cover the high- and low-level changes that had to be made in the engine to make it a better fit for DirectX 12 API. This session is about the architectural changes that had to be made, and will share an experience in achieving the speed-of-light performance of the engine with the aid of DirectX 12.

  • 4:00PM - 5:00PM | Advanced Graphics Tech: "Cinematic Depth of Field" & "Advanced Particle Simulation in Compute"
    Cinematic Depth of Field:
    Karl Hillesland (AMD) will explore a solution for high quality yet practical depth of field that leverages compute to overcome some of the previous bottlenecks commonly associated with this technique.
    Advanced Particles, Fluid and Cloth using DirectX Compute:
    Hammad Mazhar (NVIDIA) and Richard Tonge (NVIDIA) discuss how particles have many uses, from low to high complexity: non-colliding sparks, shell casings colliding against the ground, particle-particle collision for rigid debris, particle grids for cloth, up to fluid dynamics. Now that Async Compute allows us to utilize otherwise wasted GPU cycles, it is attractive to move more and more of these effects to D3D Compute. Hammad and Richard provide a sequence of successively more complicated D3D compute particle effects. They start from a non-interacting particle system, adding effects until they reach unified debris, cloth and fluid simulation, noting the incremental GPU cost of each.