Abstract

Abstract. As smart city applications increase today, the importance of web-based representations of data that serve as inputs for these applications, such as 3D city models and digital twins, is growing. Since 2011, GPU hardware has greatly advanced, and detailed 3DCMs and digital twins have increased the size of the data to be displayed, which in turn increases performance requirements. WebGL, which couldn’t fully harness the capabilities and power of modern GPUs, began to struggle to meet the increasing performance demands over time and become a bit outdated. Consequently, a new graphic API named WebGPU was developed by W3C as a response to limitations of WebGL and went live with Chrome 113. Now, we have a brand-new graphic API called WebGPU which harness the full power of modern GPUs and more performant than WebGL. As WebGPU is a new graphic API, its potential enhancements and what it can bring over WebGL in the terms of WebGIS have not been examined yet. Hence, the main idea and contribution of this work is to investigate WebGPU in real-world use cases for WebGIS applications and discuss what it brings over WebGL. For this purpose, a side-by-side performance comparison has been made. The comparisons have been made in the terms of API differences and performance. And finally, an experiment has been carried out how much data can be rendered can be rendered at minimum 60fps in both APIs. The experiments show that WebGPU is more low-level and way more performant than WebGL and it has a lot to offer in the terms of WebGIS applications.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call