Intel Confidential — Do Not Forward

From Virtual to Reality How high fidelity visualization based on Autodesk Rapid RT technology is accelerating product design decisions

Peter Rundberg, Principal Engineer, Autodesk

Mehmet Adalier, Director, Intel Corporation

INFORMATION IN THIS DOCUMENT IS PROVIDED IN CONNECTION WITH INTEL PRODUCTS. NO LICENSE, EXPRESS OR IMPLIED, BY ESTOPPEL OR OTHERWISE, TO ANY INTELLECTUAL PROPERTY RIGHTS IS GRANTED BY THIS DOCUMENT. EXCEPT AS PROVIDED IN INTEL'S TERMS AND CONDITIONS OF SALE FOR SUCH PRODUCTS, INTEL ASSUMES NO LIABILITY WHATSOEVER AND INTEL DISCLAIMS ANY EXPRESS OR IMPLIED WARRANTY, RELATING TO SALE AND/OR USE OF INTEL PRODUCTS INCLUDING LIABILITY OR WARRANTIES RELATING TO FITNESS FOR A PARTICULAR PURPOSE, MERCHANTABILITY, OR INFRINGEMENT OF ANY PATENT, COPYRIGHT OR OTHER INTELLECTUAL PROPERTY RIGHT.

A "Mission Critical Application" is any application in which failure of the Intel Product could result, directly or indirectly, in personal injury or death. SHOULD YOU PURCHASE OR USE INTEL'S PRODUCTS FOR ANY SUCH MISSION CRITICAL APPLICATION, YOU SHALL INDEMNIFY AND HOLD INTEL AND ITS SUBSIDIARIES, SUBCONTRACTORS AND AFFILIATES, AND THE DIRECTORS, OFFICERS, AND EMPLOYEES OF EACH, HARMLESS AGAINST ALL CLAIMS COSTS, DAMAGES, AND EXPENSES AND REASONABLE ATTORNEYS' FEES ARISING OUT OF, DIRECTLY OR INDIRECTLY, ANY CLAIM OF PRODUCT LIABILITY, PERSONAL INJURY, OR DEATH ARISING IN ANY WAY OUT OF SUCH MISSION CRITICAL APPLICATION, WHETHER OR NOT INTEL OR ITS SUBCONTRACTOR WAS NEGLIGENT IN THE DESIGN, MANUFACTURE, OR WARNING OF THE INTEL PRODUCT OR ANY OF ITS PARTS.

Intel may make changes to specifications and product descriptions at any time, without notice.

All products, dates, and figures specified are preliminary based on current expectations, and are subject to change without notice.

Intel processors, chipsets, and desktop boards may contain design defects or errors known as errata, which may cause the product to deviate from published specifications. Current characterized errata are available on request.

Any code names featured are used internally within Intel to identify products that are in development and not yet publicly announced for release. Customers, licensees and other third parties are not authorized by Intel to use code names in advertising, promotion or marketing of any product or services and any such use of Intel's internal code names is at the sole risk of the user.

Intel product plans in this presentation do not constitute Intel plan of record product roadmaps. Please contact your Intel representative to obtain Intel’s current plan of record product roadmaps.

Performance claims: Software and workloads used in performance tests may have been optimized for performance only on Intel® microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems, components, software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. For more information go to http://www.Intel.com/performance

Intel, Intel Inside, the Intel logo, Centrino, Intel Core, Intel Atom, Pentium, and Ultrabook are trademarks of Intel Corporation in the United States and other countries

Legal

Driving Industrial Innovation From Vision to Reality

Fundamental Discovery to Gain Fundamental Insights

Technical Computing Continues Its Rapid Growth To Compete, You Must Compute

Source: IDC: Worldwide Technical Computing Server 2013–2017 Forecast; Other brands, names, and images are the property of their respective owners.

Governments & Research Commercial/Industrial New Users – New Uses

Business Transformation Big Data Analytics Enabling Data Driven Science

Better Products

Faster Time to Market

Reduced R&D

From Diagnosis to personalized treatments

quickly

Genomics Clinical Information

Transforming the world of data and information into KNOWLEDGE

“My goal is simple. It is complete understanding of the universe, why it is as it is and why it exists at all” Stephen Hawking

From Vision to Reality: Designed Without Prototypes!

Future Intel® Xeon® Processor E5-2600 V2 family Intel® True Scale Fabric

Vehicle Images © Audi, used by Permission; Other brands and names are the property of their respective owners

"For Audi to stay on the forefront of automotive design, we required a new way to visualize our designs. Working with

Intel and Autodesk, we have been able, for the first time, to adopt Real Time Predictive Rendering to interactively see our car

design concepts with high fidelity visualization. This helps us reduce the costs of development by eliminating expensive prototype turns“

-- Audi

New Intel® Xeon Phi™ Coprocessor Products

6

Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems, components, software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. For more information go to http://www.intel.com/performance 1. Claim based on calculated theoretical peak double precision performance capability for a single coprocessor. 16 DP FLOPS/clock/core * 60 cores * 1.053GHz = 1.0108 TeraFlop/s.

7100 Family: Highest Performance

Most Memory 3100 Family:

Performance & Value 5100 Family:

High Density Form Factor

7120P 7120X 3120P 3120A 5120D

Next Intel® Xeon Phi™ Processor: Knights Landing

All products, computer systems, dates and figures specified are preliminary based on current expectations, and are subject to change without notice.

Designed using Intel’s cutting-edge

14nm process

Not bound by “offloading” bottlenecks

Standalone CPU or PCIe coprocessor

Leadership compute & memory bandwidth

Integrated on-package memory

Embree 2.0: An Advanced Ray Tracing Toolkit • Embree is an open source high fidelity visualization toolkit for

developers of ray tracing applications

• Easy to integrate, Rapid Prototyping

• Highly efficient and scalable Ray Tracing features and capabilities

• Compatible with present and future compute platforms

• Available as Open Source http://embree.github.com (Apache 2.0 license)

• Key features of Embree 2.0

• Support for Latest Intel® Xeon® Processor Family and Intel® Xeon Phi™ Coprocessor Products

• Integration with Intel SPMD Program Compiler (ISPC)

• Support for “Ray Packets” (4, 8, or 16 rays per packet)

• Two-level Hierarchies and fast BVH Builders

Embree 2.0 Modes of Operation

Rendering Engine (C++, scalar)

Embree Single Ray Kernels (C++,intrinsics)

Embree C++ API

Rendering Engine (C++, intrinsics)

Embree Packet Kernels (C++, intrinsics)

Rendering Engine (ISPC, vectorized)

Embree Packet Kernels (C++, intrinsics)

Embree C++ API Embree ISPC API

Single Ray C++ Mode • Easy to integrate • Use existing renderer

Ray Packet C++ Mode • Difficult to maintain • Highest performance • Re-write existing renderer

ISPC Mode • Easy to maintain • Good performance • Re-write existing renderer

SIGGRAPH 2013

Real-time Visualization at

Audi

Image courtesy of Audi AG

SIGGRAPH 2013

Digital Data Control Model - DKM

Visualization of the entire car…

…and final decision with the managing board.

Visual quality validation of design, sensation and surface

SIGGRAPH 2013

DKM – Rendering Algorithms

2000 - 2009 GPU Rasterization Simplest render mode Only direct light is taken into account

2009 - 2013 Ray Tracing with Ambient Occlusion Simulates how objects occlude some of the incoming light Only achieves a darkening effect where light does not reach

2013 Path Tracing Simulates global illumination and all lighting effects All material interactions are taken into account by

simulating the life of each path of light bouncing around in the scene

Very time consuming and computationally intensive

SIGGRAPH 2013

DKM Example

Gap with path tracing

Gap with ambient occlusion shadows

Slight differences in algorithms can cause big optical differences. The cost of higher accuracy is significantly longer render times.

Reliabilty of the virtual model- a question of precision

SIGGRAPH 2013

DKM Example

Gap with path tracing

Gap with ambient occlusion shadows

Slight differences in algorithms can cause big optical differences. The cost of higher accuracy is significantly longer render times.

Reliabilty of the virtual model- a question of precision

SIGGRAPH 2013

Goal: Reduction of physical cars needed for decision process > cost reduction, process speed up

• No final decisions by management board • Only used as support for the physical

based processes

2000 2010 2005 2013

2000 • PC based OpenGL Visualization • Simple materials • No shadows or reflections • No heuristic experience of the car • Very limited data sets

Development Path for Visualization @ Audi

SIGGRAPH 2013

Goal: Reduction of physical cars needed for decision process > cost reduction, process speed up

2005 • PC based OpenGL Visualization • Advanced materials • Improved light representation • No real-time shadows, calculation in

advanced is necessary • Simplified reflexions

• First use for final decisions by management board

• Often used for preliminary decisions

2000 2010 2005 2013

Development Path for Visualization @ Audi

SIGGRAPH 2013

Goal: Reduction of physical cars needed for decision process > cost reduction, process speed up

2010 • HPC based Real-time Raytracing

Visualization • Real-time shadows • Real-time reflections and refractions • Fully interactive navigation

• Most of the final decisions are based on the virtual model

• Physical models only necessary as detail models

2000 2010 2005 2013

Development Path for Visualization @ Audi

SIGGRAPH 2013

2000 2010 2005 2013

Goal: Reduction of physical cars needed for decision process > cost reduction, process speed up

2013 • HPC based Real-time Raytracing

Visualization with full Global Illumination • Physical correct lights and materials • More Simulation than Visualization • Fully interactive navigation

• Significant reduction of physical models and costs

Development Path for Visualization @ Audi

SIGGRAPH 2013

Technology - Facts and Figures

• Autodesk Opticore Studio Professional 2013

• Autodesk RapidRT Real-time Raytracing technology

Software • BARCO Projector Galaxy 4K

32 • 32.000 Lumen • Powerwall 6,0 m x 2,4 m

20 x 8 feet • 6,7 Mio. pixels

• NEC HPC Cluster with 360 nodes • Intel® Xeon® Processor E5-2670 8-

Core Sandy Bridge EP (dual socket)

• Hyperthreading 11.520 cores • 64 GB memory per node • Intel MPI Library • Cluster Network with 40 Gb/s

QDR Intel True Scale Infiniband

Hardware Powerwall

SIGGRAPH 2013

RapidRT Overview

SIGGRAPH 2013

RapidRT is a physically based path tracer. High quality final frame rendering. Interactive viewport rendering. Fully dynamic – arbitrary scene modifications with instant

feedback. Goal: consistent appearance with final frame result.

Scalable from laptops to large HPC clusters. Provides the same appearance across all platforms.

RapidRT Overview – Rendering Component

SIGGRAPH 2013

RapidRT is available in the following Autodesk products: Inventor Revit Navisworks Showcase Opticore Studio 3dsMax (internal prototype) Maya (internal prototype)

RapidRT Overview – Client Products

SIGGRAPH 2013

Interactive Draft Final Frame

>30 FPS Production movie quality

RapidRT today

Rendering Use Cases

RapidRT - Unified renderer across the full spectrum enables consistent look and feel for all use cases

SIGGRAPH 2013

Physically Based Rendering

SIGGRAPH 2013

Photorealistic rendering vs. physically based rendering. Physically based lights (1/r2 falloff, no fake shadows). Physically based materials (energy conserving, BSDFs). Physically based light transport (global illumination).

Clear trend in the graphics industry towards physically based rendering over the last few years. Film Games Recent renderers (iray, Arnold, Maxwell, Vray RT, Octane, Indigo,

LuxRender, Cycles, etc).

Physically based rendering

SIGGRAPH 2013

Simplifies and shortens workflows. Few expert parameters needed. Predictable results. Example: changing environments or lighting situations.

Saves a lot of time and money. Compute time much cheaper than artist time. Difference is increasing at a rapid rate as computation speed is

growing faster than ever – especially through the cloud. Bonus: often more realistic renderings. Note: still a lot of room for artistic freedom.

Physically based rendering – Advantages

SIGGRAPH 2013

Physically Based Noise-Free Interactive Rendering in RapidRT

SIGGRAPH 2013

Ultimate goal: render converged images at interactive frame rates using physically based algorithms. Not there yet (for most scenes). In the meantime, provide an alternative for interactive

rendering that: Uses reasonable approximations to the fully path traced result. Gives the user a good understanding of what the final image will

look like. Delivers noise-free images at 7-10 FPS. Is fully automatic – the same scene setup as for PBR.

Noise-Free Interactive Rendering – Motivation

SIGGRAPH 2013

RapidRT on Intel® Xeon Processor Clusters

SIGGRAPH 2013

Very high performance SSE and AVX instructions for allow for high parallelism Large on chip caches Intel® Xeon Processor E5-2697 (12 core Ivy Bridge) is the one of

the fastest processors we have seen Large memory Very mature tools Compiler/Debugger – Intel Composer XE Profiling tools – Intel VTune Amplifier XE, Intel TAC

Rendering on Intel® Xeon Processors

SIGGRAPH 2013

Very different from render farms End to end frame time in the tens of ms range at very

large resolutions. Leverage HPC tools and equipment to scale to huge

systems at low latencies IntelMPI Intel True Scale Infiniband

Clustered Interactive Rendering

SIGGRAPH 2013

Close to linear scalability up to ~1000 compute nodes (16000 cores). Close to converged 4k

images at ~10 fps

Clustered Interactive Rendering - Scalability

SuperMUC

SIGGRAPH 2013

RapidRT on Intel® Xeon Phi™ Coprocessor

SIGGRAPH 2013

Extreme parallelism ~60 cores per chip 16 wide SIMD instructions 4 threads per core

Large memory (16 GB) – Intel® Xeon Phi™ Coprocessor 7120 The same mature tools as on Intel Xeon processors Compiler/Debugger – Intel Composer XE Profiling tools – Intel VTune Amplifier XE, Intel TAC

Intel Embree 2.0 reference code for efficient ray traversal

Rendering on Intel® Xeon Phi™ Coprocessor

SIGGRAPH 2013

Advantages over GPU ray tracing CPU programming model and >95% shared code with CPU Advanced algorithms that don’t work on streaming architectures Stable performance all the way from simple scenes to complex

interior scenes with complex materials and many lights Large memory allows for large scenes and production rendering

Rendering on Intel® Xeon Phi™ Coprocessor

SIGGRAPH 2013

Examples

SIGGRAPH 2013

RapidRT in Inventor

Noise-free mode (interactive) Path Tracing (converged)

SIGGRAPH 2013

RapidRT in Revit

Noise-free mode (interactive) Path Tracing (converged)

SIGGRAPH 2013

RapidRT in 3dsMax

Noise-free mode (interactive) Path Tracing (converged)

SIGGRAPH 2013

RapidRT in Maya

Path Tracing (converged) Noise-free mode (interactive)

SIGGRAPH 2013

Thank you

More info? A demo? Visit us at the Intel booth!

Autodesk is a registered trademark of Autodesk, Inc., and/or its subsidiaries and/or affiliates in the USA and/or other countries. All other brand names, product names, or trademarks belong to their respective holders. Autodesk reserves the right to alter product and services offerings, and specifications and pricing at any time without notice, and is not responsible for typographical or graphical errors that may appear in this document.

Intel Confidential — Do Not Forward