This presentation by Vulkan's Strategic Officer outlines the API's strategic evolution, shifting from its original focus on direct AAA game development to addressing its current reality as a complex, multi-purpose portability layer. The speaker identifies key challenges facing the ecosystem, including growing API complexity, developer confusion, and hardware fragmentation.
The core of the new strategy is the Vulkan Roadmap, an industry alignment tool designed to guide hardware vendors and platform partners toward a common future. The roadmap is not a developer profile but a long-term plan to achieve two primary goals: first, to combat fragmentation by systematically raising the minimum feature requirements and closing the capability gap between different classes of hardware over time. Second, a new key objective is to make Vulkan "a joy to use" by improving usability without sacrificing performance. This will be achieved by influencing future hardware design to be less reliant on complex, low-level API calls for speed, and by providing clearer deprecation paths for legacy features. The talk concludes with a Q&A session clarifying that there are no immediate plans for a "Vulkan 2.0," as many of its desired goals can be accomplished within the existing 1.x framework.
Vulkan's Strategic Evolution: A Roadmap for Simplicity and Performance
0:01:04 Original Vision for Vulkan 1.0: Vulkan was initially created as a better fit for modern hardware than OpenGL, designed with a very explicit, performance-first philosophy primarily targeting AAA games.
0:02:08 Vulkan's Current Reality: The ecosystem has evolved. Vulkan is now heavily used as a portability and layering target (e.g., via DXVK for Steam, Angle for Android) rather than a direct API for many games. This has led to increased complexity, fragmentation, and developer confusion about the "right" way to perform tasks.
0:05:19 The Vulkan Roadmap's Purpose: The roadmap is an industry-facing tool to set a clear direction for hardware vendors and platform partners. It is not a developer profile to target directly, but a mechanism for accountability and planning to ensure the ecosystem moves forward together.
0:08:46 Goal 1: Tackling Fragmentation: A key goal of the roadmap is to reduce the massive feature gap between low-end (e.g., Android Baseline) and high-end hardware. This is done by planning future minimum requirements, raising the feature baseline over time, and aligning with partners like Google (Android) and Valve (Proton) on future standards.
0:15:20 Goal 2: Making Vulkan a Joy to Use: A significant new goal for the Vulkan working group is to improve the API's usability and make it a "joy to use" without compromising its high-performance nature. The strategy is to use long-term planning to influence future hardware, making the "easy way" to use the API also the "fast way."
0:19:25 Enabling Better Feature Design: The long planning cycles of the roadmap allow for wider feedback and a more considerate design process. This helps create superior, cross-vendor features by combining the best aspects of existing extensions, such as the effort to merge Graphics Pipeline Library and Shader Object into a unified KHR extension.
0:21:20 A Clear Deprecation Strategy: Vulkan aims to provide clear guidance for developers by signposting modern, preferred features (e.g., Dynamic Rendering) as replacements for legacy ones (e.g., Render Pass Objects). This will be done without removing old features or breaking backward compatibility.
0:25:19 Addressing Legacy Devices: The problem of old, feature-poor devices that no longer receive updates cannot be fixed directly. The strategy is to prevent this from happening in the future by using the roadmap to ensure new platforms and devices are built on a higher, more consistent feature baseline.
0:27:32 The Stance on "Vulkan 2.0": The speaker states there are no current plans for a Vulkan 2.0, as there is no consensus on what it should be. The cost of breaking backward compatibility is high, and many desired improvements (like simplification) can be achieved within the 1.x framework.