- What’s new with RenderMan?
RenderMan version 22 is focused on artist productivity, achieving remarkable levels of interactivity through significant new core technology, improved data management, and advanced new features drawn from Pixar film productions. Major updates to the core architecture allow RenderMan to support “live rendering” at every stage of the pipeline, providing fluid feedback for fast artist iteration. The ability to call on RenderMan’s full light transport computations throughout the filmmaking pipeline, authoring to a single set of assets, will enable new kinds of collaboration and will transform traditional pipelines. RenderMan’s acclaimed scalability, handling huge scenes executing on huge batch render farms, benefits this new engineering too – driven by production requirements at customer studios. RenderMan version 22 also includes advanced lighting technology used internally on Pixar’s own animated feature films, including Finding Dory, Cars 3, Coco, and Incredibles 2. Pixar research and development also provided significant new path tracing and sampling techniques for producing superior imagery faster. Version 22 will also introduce Pixar’s ground-breaking USD (Universal Scene Description) into RenderMan, integrating this innovative and evolving pipeline technology directly into the renderer.Learn More
- Have the artist interfaces changed?
Yes! The RenderMan artist interfaces have been updated in three ways: Idea Iteration Speed, Artistic Capabilities, and Interface Clarity. RenderMan provides deep new versions of our integrations into artist applications such as Autodesk’s MAYA, Foundry’s Katana, and SideFX Houdini. We evaluate other applications frequently and prioritize work on the ones that Pixar and other studios need most. Direct manipulation of parameters and objects are instantly reflected in RenderMan’s new live rendering modes. With full support for single pass workflows, RenderMan delivers an optimized shading and lighting pipeline, simplifying the process of managing and rendering complex shots. Since geometry creation, deletion, and edits are all supported live mode, along with a live renderer that is “always on” – reflecting artist edits to shader parameters, light parameters and positions, and camera parameters. RenderMan provides a large catalog of ready-to-use materials and light rigs, called the RenderMan Preset Browser. Items in the catalog can be used as-is or can be copied and modified, and artists can create their own catalog entries. The Preset Browser entries are sharable between RenderMan artist applications, and between users and shows. New results from Pixar productions and research give artists great controls such as the artistic capabilities of RenderMan’s Marschner Hair Shader and others. The control options themselves have been rearranged or thoroughly rewritten for clarity and to expose useful simple controls first, with advanced settings still accessible. All of these features combine to provide an artist experience that is both professional and enjoyable to use.
- What technology is at the heart of the material catalog?
RenderMan provides a very sophisticated set of fundamental shader components and ray tracing algorithm controls, often referred to as “uber shaders”. Each of these carefully factored components can have highly articulated input parameters providing studios with plenty of ways to attain a distinct look as well as ways to achieve any effect. The Preset Browser provides a way to save the values of these many parameters as “presets” – collections of parameter settings and shader networks that work together to make just the right polished steel or aged oak for your project. You can start with the preset catalog that we provide, making subtle changes as you like, or you can create entirely new materials or light rigs from scratch – then save and share those with your crew.
- Is RenderMan version 22 a new rendering architecture?
RenderMan version 22 retains the key outward facing RIS architectural factoring and composability features introduced over the last several releases. Many scenes built for version 21 will render with no change in version 22. However, version 22 introduces significant new internal changes to system architecture, APIs, and data handling that enable its remarkable new interactive live rendering capabilities. Old API external interfaces still exist, but there are also new ones are geared towards modern hardware and application structures. RenderMan has been constantly evolving since its introduction more than 30 years ago to address the ever-changing artistic demands of Pixar itself and the productions at other major studios. Sometimes that evolution was accomplished through architectural changes, and sometimes through R&D leading to new features - often a combination of the two. Version 22 is both, having pioneering new features and algorithms combined with highly efficient, forward looking new internal systems.
- How does the new system architecture help artists?
We have rearchitected our Maya, Katana, and Houdini integrations to use the new APIs that version 22 has been built around. These APIs allow us to create a high bandwidth, high throughput pipe between the various artist applications and technology such as USD with RenderMan itself. This fast and wide data pipe is what gives version 22 the ability to provide artists with a high level of interactivity. It also allows us to more easily integrate with other applications in the future.
- Does the new system architecture mean that RIB is gone?
No. RIB is still around and is an excellent way to feed RenderMan version 22 in batch rendering. In RenderMan 21, RIB was used to drive RenderMan from Maya. Where in version 22 the new system architecture is used. In RenderMan version 22 RIB isn’t as prevalent as RenderMan version 21, but it is still available. In fact, one of the batch render options from Maya is to create a set of RIBs and to then pass it to the farm for rendering.
- Does RenderMan retain its traditional customizability?
Yes. RenderMan offers a complete set of ready to use rendering and artist tools, but it also retains its storied ability to be profoundly customized by experts who want to develop unique looks or unique pipelines around the RenderMan core. A fundamental aspect of the RIS architecture is the ability to replace components or even entire renderer subsystems with custom plug-ins or scriptable elements, such as for BxDFs, Patterns, Projections, Integrators, LightPathExpressions, and more. All of the traditional aspects of custom geometric proceduralism are still supported as well, such as programmatic generation of particles, hair, or foliage.
- Will the new APIs that work with the new system architecture be published?
We are planning to publish the new APIs at a later date. They will not be published with RenderMan version 22. We are keeping them private so that we can iterate on them to make sure they are properly structured and fully featured before we release them publicly in a subsequent version of RenderMan. A few of our closest partners will be helping us in this process.
- How are new shaders and materials developed?
RenderMan supplies a small number of very powerful “uber shaders” such as PxrSurface, PxrVolume, and PxrMarschnerHair whose parameters can be set to produce a huge number of unique and artistically tuned looks and effects. Primarily, new materials are defined by creating shader networks built from a library of these and other of functional nodes. Open Shading Language (OSL) scripting is supported as an easy and powerful way to create procedural patterns. BxDF shaders can be written directly in C++ to implement sophisticated light transport interactions with materials. One of the huge strengths of RenderMan is the factoring of the entire renderer in to functional units that can be varied independently, or even replaced with custom code. Even the primary path tracing algorithm itself, called the integrator, can be selected from a menu of alternatives – from general purpose unidirectional path tracers to sophisticated industry leading bidirectional solutions, there are even tracers that visualize geometry characteristics while developing or debugging other shading network components.
- Will my custom RenderMan 21 BxDFs, OSL patterns, procedurals, and other plugins continue to work in RenderMan version 22?
Yes, with minor changes. We are constantly working to make RenderMan a better product. This includes adopting new technologies, adopting new techniques, and making RenderMan evermore performant. In order to keep RenderMan capable of handling ever growing scene complexity and production requirements, we need to make changes to the underlying APIs. RenderMan version 22 is no different. Our overall RIS architecture has not changed, so any custom BxDFs, OSL patterns, C++ patterns, integrators, procedurals, or other plugins should continue to work. However, your C++ plugins are likely to require a few minor changes to work in version 22.
- What is USD?
Pixar has evolved the Universal Scene Description (USD) project over the years from several internal production projects. It is a system for authoring, composing, and reading hierarchically organized scene description. USD comprises a set of modules that scalably encode and interchange static and time-sampled 3D geometry and shading data between artist applications, to archives, and to renderers. Domain-specific schema modules define the geometry and shading encoding atop USD's domain-agnostic core. Pixar now manages USD as open source project, releasing it to the industry as a whole in order to promote a sensible tool ecosystem that will benefit all studios. It also can simplify interchange between studios when they are collaborating on projects. USD allows artists to control non-destructive variations on base assets, and to transmit sophisticated shading networks and lighting information between tools, in addition to raw geometry. All of which and more make it valuable. The RenderMan team works closely with the USD developers at the studio to ensure that RenderMan provides first-class support for USD features as they evolve.
- How scalable is RenderMan CPU thread usage?
RenderMan takes full advantage of modern CPU architectures and core counts. Raw performance is always strongly dependent on scene specifics and whatever custom shading, geometry, or geometry nodes are supplied by studios themselves. It can also be affected by file i/o speed and memory speed and availability. RenderMan compute speed can scale nearly linearly with the number of available CPU cores on typical commodity studio desktops and render farms, and it is even tested frequently on systems having 100+ physical cores. Hyperthreading of these cores, doubling the thread count, also often adds measurable benefits. RenderMan subsystems can also take full advantage of SIMD hardware extensions, including “wide SIMD” instructions sets such as AVX-512.
- What is rendering noise, and why is denoising important?
RenderMan’s advanced form of physically-based path tracing simulates real light transport and can produce realistic lighting effects, as well as providing intuitive lighting and camera controls to artists. Computing the true physics of light isn’t practical or necessary for production work. So instead path tracers sample the environment with statistical probes that bounce light among objects and the camera. Light energy is altered by surface materials, atmospheric phenomena, reflections, refractions, lens distortions, motion blur, and more. RenderMan makes high quality images that look great on the big screen by averaging together billions of samples, converging toward the real solution. If sampling ends too soon or the paths aren’t distributed well, then some light energy or blockers won’t be detected, and these “errors” or incomplete measurements are called noise. The effect is usually speckled artifacts because some pixels remain too bright or too dark. Computing to full convergence prevents noise but can take a long time so production rendering often cuts off early, leaving some visible noise. Denoising then is a post process that smooths over that noise using estimates. It is difficult because simply averaging neighbor pixels produces blurry images. Instead sharp edges and fine detail must be retained, such as highlights on different strands of hair, and some objects really are speckled! RenderMan provides a best-in-class tunable denoiser developed by Walt Disney Animation Studios, Disney Research Zurich, and Pixar. It does such a good job on complex production images that actual rendering time (convergence) can often be stopped earlier than before, prior to denoising, shortening overall renders time. Denoising is a very active area of research, and the RenderMan team tracks potential next-generation solutions.
- Is RenderMan implementing new open source standards?
Yes. RenderMan includes support for many industry standard tools and formats including: USD, OSL, VDB, Cryptomatte, OIIO, Alembic, PTEX, OpenEXR, SeExpr, OpenSubdiv, and others. We also track ongoing developments in projects like MaterialX, Embree, and Optix closely.
- Are there different versions of RenderMan for different studios?
No. RenderMan exists in one standard version for everyone and the technology that Pixar develops for RenderMan has always been delivered to the entire RenderMan community through the productization efforts of the RenderMan team. It is RenderMan’s famous underlying flexibility that allows individual studios to customize the renderer via their own, shaders, plug-ins and other pipeline tools.
- What 3D applications does RenderMan support?
Pixar provides artist-friendly, deeply integrated RenderMan interface plug-ins for Autodesk's Maya®, The Foundry's KATANA®, and Side Effect’s Houdini®. Please refer to renderman.pixar.com for details of version compatibility for each new release.
- When will the RenderMan version 22 for Houdini plug-in be available?
Customers who are already using the version 21 RenderMan for Houdini will be able to continue using it with version 22.0, immediately when it becomes available. A completely new RfH with deep integration for Houdini 16.5 and 17 and new live rendering capabilities will become available early in the RenderMan version 22 family of point releases, with an initial release currently target before the end of 2018. Updates for new RenderMan and Houdini features will be added in future major and minor releases directly from the RenderMan download site, like plug-ins for the other artist applications.
- What about support for other applications like Cinema4D and Blender?
We track the applications that are most important to productions at Pixar, ILM, and our community of commercial customer studios, and we frequently evaluate additional deep integrations within other Digital Content Creation applications. An important aspect of RenderMan version 22 has been re-engineering our live rendering interfaces specifically to allow easier, faster, and better integrations into new artist tools. We understand the power and capabilities of applications such as Cinema4D and Blender, and the strong interest in them, particularly from the Non-Commercial RenderMan community. We are pursuing integrations for these applications, and others, sometimes through third parties, as commercial priorities allow. Please contact us at firstname.lastname@example.org and we can help you find the best solution for your requirements.
- Is Tractor included in the latest version of RenderMan?
Yes. New purchases of RenderMan include a matching number of Tractor licenses. Tractor is also available separately at $100 per license.
The Future Of RenderMan
- What is RenderMan XPU, how does it use GPU hardware?
Pixar has been using and developing various special purpose GPU renderers internally for several years. Building on that experience, the experimental RenderMan XPU project is a flexible and modern approach to combining CPU+GPU computations for faster rendering when powerful GPU processors are available – though it can always fully render with the CPU only if necessary. Pixar has committed to commercializing the RenderMan XPU technology in some future release, currently targeting the RenderMan version 23 timeframe, sometime in 2019. The initial feature set and product packaging are still being defined, and the product capabilities will of course receive ongoing extensions and updates in later releases. Indications from early internal testing show very compelling results.
- What is the roadmap for USD and RenderMan?
Inside Pixar, the RenderMan and USD teams work closely with each other, both on studio filmmaking projects as well as on crafting useful next-generation pipeline tools for the animation and VFX communities. Both teams also draw significant inspiration and feedback from those communities, and are committed to making the tools work well together in many contexts. RenderMan is committed to native support of USD assets, and USD is committed to native support of RenderMan as the preeminent first class rendering delegate.
- What is the relationship between RenderMan and other R&D groups within Disney?
There are a number of research initiatives within the Disney Studio organization including Pixar, Walt Disney Feature Animation (WDAS), Industrial Light & Magic (ILM), Disney Research in Zurich, and others. Each of these groups is separate so they pursue different approaches to rendering and solving production problems in different ways. No other rendering R&D effort can compare with the resources and depth of experience embodied in the Disney organization and its decades of experience defining the state of the art in animation and film rendering. RenderMan is the conduit through which tools and techniques that derive from this R&D effort will be channeled into the public domain. Pixar and these other Disney groups describe their significant contributions regularly at commercial and technical research venues around the world.
- Have prices changed?
Yes, RenderMan prices are periodically adjusted to account for the competitive market and to actively sustain the commercial aspects of creating and supporting the industry’s leading rendering solution for a huge variety of studios and platforms.
- What is the new price of RenderMan?
RenderMan now costs $595 per license, providing access to either the artist interface or the batch renderer. For larger requirements over 25 licenses, and especially over 250 licenses, attractive studio pricing and payment packages are available. Usage based cloud licensing is available through various cloud compute vendors separately. Rentals and special short-term burst capacity options are also available.
- Is maintenance on RenderMan available and what does it cover?
Yes. Maintenance is $250 per RenderMan license for commercial users and includes full support and upgrades throughout the renewable 1-year term of the agreement.
- Are RenderMan subscription licenses available?
Several licensing models are available. Traditional perpetual licenses are what many studios continue to choose. These licenses allow you to run the version of RenderMan available at purchase time for as long as you like, for an up-front purchase price. By purchasing recurring annual maintenance for those initial licenses, they can be upgraded to new major releases immediately as they become available. License rentals provide access to the current release for a short term, then the licenses expire. For studios needing a large number of licenses, RenderMan is also available in fixed-term bundles with annual billing on an attractive discount schedule.
- Are RenderMan licenses floating or node-locked?
All commercial RenderMan licenses are available as floating, providing the flexibility to deploy RenderMan anywhere on your network. However, node locked licenses are available if required.
Non-Commercial Licensing FAQ
- When is the free Non-Commercial RenderMan available?
The free Non-Commercial version of RenderMan is typically released shortly after each major commercial release. Timing of the NCR release depends only the time needed to properly package the product for the non-commercial downloads, and to prepare the website and licensing infrastructure. The Non-Commercial RenderMan download contains the same renderer and features as the main flagship commercial product.
- How does Pixar define “non-commercial” use?
Simply any usage of RenderMan that does not involve direct commercial use to generate profits. Example non-commercial use-cases include evaluations, personal learning and familiarization, student usage, any type of experimentation, research, and the development of tools and plug-ins that compliment the RenderMan.
- Is Non-Commercial RenderMan appropriate for Educational Institutions?
In general, no. Non-Commercial RenderMan is intended for individual use. For classroom deployment or other educational use cases, please contact email@example.com discuss whether other options such as floating license grants may apply.
- Is the free non-commercial RenderMan restricted in any way?
No. There is only one RenderMan and the free non-commercial RenderMan is exactly the same as the commercial version. There are no watermarks, no time limits, and no reduced functionality. The only limitation is that upon acceptance of the EULA at installation, the software must only be used for non-commercial purposes. Technical support for Non-Commercial RenderMan is supplied by the active and highly technical community itself.
- Why is Pixar making RenderMan free for non-commercial use?
This is a strategic and long-term commitment by Disney / Pixar to the advancement and dissemination of the production industries most advanced rendering technologies and the interchange of assets in common formats. First, RenderMan going forwards will be the conduit through which applicable rendering technologies developed within Disney / Pixar research will be channeled into the public domain to establish a common platform for production, research and development, trials and experimentation, learning, and other applications. Second, it is Pixar’s belief that limitations on software access have become a brake on the development of the production industry, and that universal access and a set of common standards and practices can only stimulate greater growth and development. Third, existing trial and evaluation methods of providing access through watermarked or time-expiry licenses are unsatisfying for proper evaluation. The resources and technology now being invested in RenderMan are of superior quality and will continue to anticipate the needs of film production imagery as Pixar has over the past 30 years. Providing RenderMan free for non-commercial usage represents the commitment of Disney / Pixar that RenderMan is the long-term film rendering standard.
- Do I need to call Pixar directly to access free Non-Commercial RenderMan?
No. Non-Commercial RenderMan can be obtained from the RenderMan website by automatic download after registration.
- When version 22 of Non-Commercial RenderMan is released, will I still be able to download version 21?
No, Non-Commercial RenderMan downloads are updated to the latest available version in order to keep the research community up to date, simplify community support, and to ensure that tool developers have a consistent target platform.
- Are floating licenses available?
No. Only node-locked versions of the free Non-Commercial RenderMan are available.
- What if I want to use Non-Commercial RenderMan for commercial work?
Simply contact us at firstname.lastname@example.org, and we will register you as a commercial user and create a custom commercial licensing package for your requirements.
- Can images created in Non-Commercial RenderMan be published?
Yes, as long the images are not sold for profit or used for commercial advertising purposes. Please include the following “rendered with RenderMan” credit to indicate Non-Commercial RenderMan usage.Get Logo
- Can plug-ins and tools developed with Non-Commercial RenderMan be sold commercially?
Yes. Non-Commercial RenderMan cannot be used directly for commercial use, but we encourage you to use RenderMan as a development platform for your own tools, technologies, and complementary solutions that are made available to the community commercially.