Embedded Files
The Internet was built on the promise of distributed computing and the open exchange of ideas and data. As we move from hypertext and hypermedia to a fully immersive metaverse built on interactive 3D content, the challenges and opportunities are exponentially greater[2]. In order to support the emerging media of tomorrow, it is crucial to be able to bridge the gap between the promise of these breakthroughs and the feasibility of widespread access to them. Until now, no system scales rendering speed across many dimensions of work in order to allow content creators to tap into the vast pool of graphics cards from an online network.
Problem
We are on the cusp of a technological transformation of our very view of reality, affecting everything from computation to physics. Everything is becoming more virtual, from the explosion of mobile content over the past decade to engineers producing new realities, augmented and virtual, that allow us to immerse ourselves in new computer-generated worlds. As entertainment companies adopt these new ways of producing new visual effects, content creators and editors find themselves facing new dimensions of complexity. Larger and more complex jobs spanning thousands of frames across time (for animations) and space (for VR walkthroughs) require external servers and additional resources. Complexity of rendering is also exponentially increasing due to higher frame resolution and frame rate (e.g. UHD 8K@240 fps is 256 x the work of HD 720p30). Further raising complexity are increases in views per frame (e.g stereo rendering doubles workload to support left and right viewpoints). In addition to increasing complexity, the use of computer generated imagery (CGI) is accelerating. Over the past decade, the amount of visual effects (VFX) scenes per Film and TV productions is accelerating has grown dramatically, while Architecture, Design, Marketing and Engineering firms increasingly rely on advanced CGI and photorealistic rendering to support new forms of visualization.
Already within the public GPU cloud an increasingly large number of market segments are competing for limited High Performance GPU compute resources. Centralized data centers have been unable to expand fast enough or price competitively to meet demand due to high upfront CAPEX and rapid GPU hardware depreciation. Because centralized investment in GPU infrastructure risks underutilization when demand falls below capacity, pricing has remained high - or requires large upfront commitments to offset monetization risk. For independent artists wanting to produce virtual and augmented reality experiences, local GPU render farms are prohibitively expensive and impractical to install. As a result, they have to choose between costly GPU cloud rendering services or building expensive local GPU server infrastructure.
Yet, the current system harbors many inefficiencies. Most developers’ GPUs remain idle when they are not rendering their own work, reducing the productivity potential of local GPU infrastructure. Further, excess GPU supply from proof-of-work cryptocurrency mining has led to an arms race dynamic in which increasing compute resources are dedicated to mining fixed (or regressive) block rewards. The rise of ASIC (Application Specific Integrated Circuit) and FPGA (Field Programmable Gate Array) mining, in which hardware is designed exclusively to solve blockchain hash functions, has only exacerbated this trend. The result is that the use of GPU’s for proof-of-work blockchain mining consumes tremendous power with fewer productive returns, leading to waste or shutdown in certain cases. With the rise of more computationally efficient blockchain protocols, there is an opportunity to more productively use latent GPU compute resources to create the next generation of 3D and holographic media.
In distribution, not only do artists face exponential rendering increases - with the push to higher frame rates, higher resolution, and more realistic images - but it is also difficult to monetize 3D works across a fragmented app ecosystem. Holographic media requires new layers of 3D metadata, asset interoperability, and granular Digital Rights Management (DRM) when compared to less interactive media formats like 2D rasterized imagery. Even today, studios or independent content creators have difficulty tracking the media they create. As the Internet grows, it becomes almost impossible to effectively track digital assets and to prove an infringement through copyright and trademark laws. As a result, business models have forced users into all or nothing subscription paywalls, costly downloads, or intrusive data mining for ads. Despite innovations in eCommerce, producers and consumers have few options for micro-consumption.
The lack of a 3D media monetization framework is leading to a market failure depressing the growth of immersive virtual and augmented reality production. Developers and content creators are often forced into walled-garden app stores, resulting in platform lock-in and increased intermediary costs, harming both consumers and producers. Further, app store fragmentation limits artists ability to monetize complex 3D works across the widest audience, making it less economically viable to produce more ambitious immersive works.
Ultimately, pressures on GPU cloud rendering as well as technical limitations in distributing and monetizing complex holographic works are constraining the growth of holographic media. This next generation of media, which is more costly to produce and more complex in its distribution, is most efficient through an open smart contract based micro transaction framework for effective monetization.
RNDR Solution
Most artists’ GPUs remain idle when they are not rendering their own work. By utilizing the RNDR Token network ecosystem, developers can choose to monetize their idle GPU power by performing renders for RNDR Tokens. The RNDR network provides a mechanism for idle GPU power from around the world to be efficiently allocated for complex 3D rendering tasks, providing a scalable peer-to-peer GPU cloud computing network[3]. Through this process, RNDR minimizes GPU waste, provides Artists with affordable and fast on-demand parallelized rendering, and enables efficient dual use of compute resources. Developers with idle GPUs can contribute their spare capacity to the network and earn tokens, which they can re-use when rendering demands exceed local infrastructure and additional scale is needed.
The blockchain based rendering network facilitates efficient, reliable, and remunerative rendering of time-stamped tasks on a peer-to-peer basis. As node operators successfully complete scenes on the network, they build trust needed to perform higher tier rendering tasks. Nodes work by accurately carrying out the rendering instructions encoded by a user into a 3D scene graph. When a user confirms that the work was successfully executed, the node becomes more trusted. The network refers to this process as Proof-of-Render where trustworthiness is built on rendering 3D scenes rather than the proof-of-work mining used by other blockchain protocols.[4]
Blockchain technology has also evolved to now store, validate and time-stamp complex mixes of technical specifications, schedules, accounts, regulations, protocols, standards, and property rights. This technology can handle digital rights management, needed for complex digital assets that can be routinely copied and for which time-stamped proof of authorship is crucial. Recently, the Ethereum blockchain has enabled tokens, which allow for immediate and more complex transactions to be executed using smart contracts[5]. Inside each RNDR transaction backed by an ORBX package, there is a hash of all assets in the system used to build the source render graph for the render job. Within the ORBX module, application and plugins can be launched, enabling users to access remote 3D experiences as ORBX.js or HTML 5 live streams.[6] This means flow of RNDR tokens that utilize and render each asset can be spread back to creators of these assets (e.g. 3D models, texture packs, industrial scans, toolchain services) whenever they are transactionally connected back to their source render job (either once, or recurring).[7]
With RNDR, physically correct rendering tasks are completed quickly and efficiently in a blockchain based peer-to-peer network with no error or delay and with securely protected property rights. GPUs that are otherwise wasted on proof of work mining are used to render ultra-high resolution images, generating more productivity from these compute resources. RNDR, powered by the blockchain, uses the act of creation — rendered by the laws of physics and light — to open the door to a new model of holographic computing.
Report abuse