Since its founding in 1969, SIGGRAPH has been the place to learn and share advancements in computer graphics and interactive techniques. My colleagues on the team working in graphics and simulation, reminisced on their time at SIGGRAPH in the 90s when working at studios like Pixar, and I got the chance to return to SIGGRAPH 2022 in Vancouver (locale of my 1st SIGGRAPH) after 10+ years.
What I’ve found to be unique about SIGGRAPH (stands for ‘Special Interest Group on Computer Graphics and Interactive Technique’) is the blend of thought leadership within art, design, and technology - all coming together to focus on storytelling. This year, SIGGRAPH introduced topics from artificial intelligence and the metaverse to accelerated computing and digital realities, very fitting to Duality’s focus.
Building the Open Metaverse
It first started as a Birds of a Feather session (more informal), continued on as a podcast with conversations with technical experts, and then at SIGGRAPH 2022 a full-day course programming was put together to cover the current state, Metaverse Standards, and potential future directions. Kudos to Marc Petit (Epic Games) and Patrick Cozzi (Cesium) for organizing!
Takeaways
- The idea of Composability (a great read from Jon Radoff)
The long-term thinking behind what’s being built for the metaverse. Composability being the most powerful creative force and the power of iteration. - What does geospatial have to do with the metaverse? Everything.
As OGC members, it was like world’s colliding meeting Nadine at SIGGRAPH. This year especially, it was clear that SIGGRAPH wasn’t just about creating hyper-realistic films or games. But how might the areas of 3D-first computing, game engine ecosystems, the evolution of content creation, scaling users and worlds, and interoperability come together to solve real world problems? Some for example, climate change and emergency response!
- Standardization is the work of a community. The constant necessary loop of standards ⇌ collaboration.
With the two current dominant standards of glTF and USD, we are committed to supporting this organic evolution and finding the best ways to work with both. In the last three years, we’ve adopted dozens of open standards without definitively attaching to any single one. So here we continue as open source enthusiasts - and we take that as a responsibility to better the emerging and evolving standards of our industry.
glTF - or, Graphics Language Transmission Format, a JSON based format developed and maintained by the Khronos Group. This is the most web-centric format. It is fairly streamlined and interoperable - having an extension system supported by a large set of platforms. It is also relatively easy to parse. GLTF 2.0 was recently released as the ISO/IEC international standard, furthering its claim to be the “JPEG of 3D.”
USD - or, Universal Scene Description is a format developed by Pixar. More complex than GLTF, USD carries more emphasis on rendering, while maintaining strong typing and complex structures, often making it the more powerful standard to utilize all of today’s modern features with better control. Ultimately, USD can be viewed as more user friendly, but this does mean that USD requires a dedicated parser. USD has gained significant traction in large part due to adoption by industry giants like Apple (.usdz) and NVIDIA.
What's next
We have an idea and direction for this 3D-first computing landscape and how Digital Twins are visualized (supported meshes and materials) but what about the ontology and physics of them? Our team and the broader community are actively working on these areas and we look forward to sharing more with you.
