DOCSLIB.ORG

Flexible and Efficient AI Solutions 2 Powervr Vision & AI Portfolio Combination of Hardware, Software, and Ecosystem Integrations

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Flexible and Efficient AI Solutions 2 Powervr Vision & AI Portfolio Combination of Hardware, Software, and Ecosystem Integrations 运用GPU + NNA 实现设备端异构计算 2018年11月 www.imgtec.com A history of Imagination Dreamcast Focuses Intel / Apple Over 11B Acquired by Listed in London Console strategy on invest Acquire MIPS shipment Canyon Bridge mobile 1985 1994 1995-1998 1999 2002 2006 2013 2017 2017/11 Set up: Computer First Develops MBX 1B GPU Launch of Spin off Focus on 3D GPU affordable IP mobile shipment first NNA MIPS video with NEC DAB business GPU hardware radios model shipping © Imagination Technologies Flexible and efficient AI Solutions 2 PowerVR Vision & AI portfolio Combination of hardware, software, and ecosystem integrations Ecosystem Open source tools & machine learning frameworks Drivers & SDKs & Debugging & Software Low-level high-level Compilers & analysis APIs APIs tuning tools tools Neural Hardware Network GPU Dedicated Accelerator Processors © Imagination Technologies Flexible and efficient AI Solutions 3 PowerVR GPU Efficient compute across a wide range of markets . Flexible range of solutions . XE/XM/XMP Series: Cost efficiency focussed . Performance/mm2 . XT Series: Power efficiency focussed . Performance/mW . Highly scalable and configurable . Adapt to range of market requirements . Feature set, performance characteristics . Compute versus Graphics balance . System focus . Optimised to work with PowerVR NNA © Imagination Technologies Flexible and efficient AI Solutions 4 PowerVR Neural Network Accelerator (NNA) Complete neural network inferencing solution Typical SoC Image sensors Video ISP Codec . Brand new architecture Peripherals PowerVR PowerVR 2NX HDMI NN Accelerator I/ O . Provides the best performance for even the smallest Fabric area budget Wireless 2NX Driver DDR Main CPU GPU Audio . Complete solution for NN acceleration Flash . Offload up to 100% of NN processing . Optimised for low bit depth . Minimises the system bandwidth load © Imagination Technologies Flexible and efficient AI Solutions 5 Flexible AI offerings: PowerVR NNA Fastest performance with PowerVR NNA System Memory Bus PowerVR NNA • Fastest performance • Lowest cost • Supported feature set © Imagination Technologies Flexible and efficient AI Solutions 6 Flexible AI offerings: PowerVR GPU Fully flexibility with PowerVR GPU PowerVR GPU ALU Cluster TA Scheduler ISP Compute Data Master ALU Cluster MMU/System System Level Memory Cache Interface System Memory Bus PowerVR GPU • Fast performance • Low cost • Fully flexibility © Imagination Technologies Flexible and efficient AI Solutions 7 Flexible AI offerings: PowerVR NNA + GPU Fastest Performance and fully flexibility PowerVR GPU On Chip Memory ALU Cluster TA Scheduler ISP Compute On Chip Memory Data Master ALU Cluster MMU/System System Level Memory Cache Interface System Memory Bus PowerVR NNA PowerVR GPU On Chip Memory • Reduces bandwidth • Fastest performance • Fast performance • Lowers latency • Lowest cost • Low cost • Shared by other cores • Supported feature set • Fully flexibility © Imagination Technologies Flexible and efficient AI Solutions 8 Moving beyond hardware . AI has no open-standard API . Popular frameworks are desktop first . Requires rich software environment to compliment hardware © Imagination Technologies Flexible and efficient AI Solutions 9 Software for AI www.imgtec.com PowerVR AI Software Overview Unified solution across NNA and GPU Neural Network Framework Offline . PowerVR AI Toolkit PowerVR AI Toolkit Extract Analyse Map Debug . Simplifies deployment workflow NN Model Optimised Weights . Easy debugging and network analysis NN Model Weights Application . User optimisation networks possible On Device Android NN HAL IMG DNN API . Supported APIs: IMG DNN & Android NN NNA Driver GPU Driver . Seamless operability between GPU & NNA H/W PowerVR NNA PowerVR GPU . Straight forward development . IMG DNN API available on GPU today © Imagination Technologies Flexible and efficient AI Solutions 11 IMG DNN API Cross core low-level interface for easy integration for applications or SDKs Common interface Target selectable Model importable Simplifies Maximises Easiest development flexibility deployment © Imagination Technologies Flexible and efficient AI Solutions 12 Android Neural Network API support Google’s brand new interface for AI acceleration Introduced in 8.1 Extended in P Available for NNA and GPU © Imagination Technologies Flexible and efficient AI Solutions 13 Summary Imagination provides flexible, efficient solutions for dedicated AI processing . Complimentary range of hardware offerings . Select the best combination to match any requirements . Highly efficient Vision and AI IP . Efficient mapping of neural network models . Highly optimised inference execution . Single common NN API . Integrate once and utilise all our hardware . Android supported . Perfect for mobile developers © Imagination Technologies Flexible and efficient AI Solutions 14 Thank You! www.imgtec.com .
Load more

Recommended publications
  • EEMBC and the Purposes of Embedded Processor Benchmarking Markus Levy, President
    EEMBC and the Purposes of Embedded Processor Benchmarking Markus Levy, President ISPASS 2005 Certified Performance Analysis for Embedded Systems Designers EEMBC: A Historical Perspective • Began as an EDN Magazine project in April 1997 • Replace Dhrystone • Have meaningful measure for explaining processor behavior • Developed business model • Invited worldwide processor vendors • A consortium was born 1 EEMBC Membership • Board Member • Membership Dues: $30,000 (1st year); $16,000 (subsequent years) • Access and Participation on ALL Subcommittees • Full Voting Rights • Subcommittee Member • Membership Dues Are Subcommittee Specific • Access to Specific Benchmarks • Technical Involvement Within Subcommittee • Help Determine Next Generation Benchmarks • Special Academic Membership EEMBC Philosophy: Standardized Benchmarks and Certified Scores • Member derived benchmarks • Determine the standard, the process, and the benchmarks • Open to industry feedback • Ensures all processor/compiler vendors are running the same tests • Certification process ensures credibility • All benchmark scores officially validated before publication • The entire benchmark environment must be disclosed 2 Embedded Industry: Represented by Very Diverse Applications • Networking • Storage, low- and high-end routers, switches • Consumer • Games, set top boxes, car navigation, smartcards • Wireless • Cellular, routers • Office Automation • Printers, copiers, imaging • Automotive • Engine control, Telematics Traditional Division of Embedded Applications Low High Power
    [Show full text]
  • Imagination Technologies Group Plc Annual Report 2006
    Imagination Technologies Group plc Annual Report 2006 The data and projections shown on pages 1-24 are the product of consolidated partner data, analyst information and Imagination Technologies research. Imagination Technologies, the Imagination Technologies logo, PowerVR, the PowerVR logo, Metagence, the Metagence logo, Ensigma, the Ensigma logo, PURE Digital, IMGworks, CodeScape, META, MTX, MBX, MBX Lite, SGX, UCC, MiniEngine, PocketDAB, ReVu, the Bug, the Bug logo, Legato, EVOKE-1, TEMPUS, CHRONOS, OASIS, PURE ONE, SONUS-1XT and Élan are trademarks or registered trademarks of Imagination Technologies Limited. All other logos, products, trademarks and registered trademarks are the property of their respective owners. Copyright © 2006 Imagination Technologies Limited, an Imagination Technologies Group plc company. 4 colour print + matt lamination on 1 side, 215x280mm (h) trim size with 4mm spin, 300gsm board fold, gather and perfect bind 19/06/2006 Contact Oscar Tse / IMG Publications / 01923 260511 / [email protected] Introduction Improvements in technology mean that more and more functions can be integrated onto just one silicon chip. Whilst in the past, half a dozen companies might each have made a specialised chip to go into an advanced product like a TV or mobile phone, now it is more likely that one large semiconductor company will make a single chip that contains most, if not all, of that specialised knowledge. That’s where silicon IP, or ‘intellectual property’ comes in. Semiconductor companies can licence parts of a chip design from companies that specialise in those advanced technologies, instead of developing the technology themselves. They are especially likely to do this in new market areas.
    [Show full text]
  • Power Aware Tactical Computing
    Power Aware Tactical Computing Song J Park1, Dale R Shires1, Brian J Henz1, James A Ross2, David A Richie3, and Jordan J Ruloff2 1U.S. Army Research Laboratory, APG, MD 2Dynamic Research Corp., Andover, MA 3Brown Deer Technology, Forest Hill, MD Abstract - Power consumption has become a chief supports and maintains high performance computing (HPC) impediment in the advancement of digital computing. To resources. The ARL DSRC provides state-of-the-art improve performance amid power limitations, accelerators computational solutions for the DoD research and are being applied to everyday systems. In particular, due to development community. Among the systems available at the mix of popularity and raw computational power, ARL is Harold system, which consists of 10,752 cores with graphics processing units (GPUs) have extended the the processing capability of 120 trillion floating-point applicability of digital computing in a multitude of sectors operations per second (TFLOPS). A predecessor, JVN from ubiquitous smartphones to environmentally responsible system, decommissioned in 2009, had 2048 cores with the supercomputing. Operating within the mobile power theoretical peak of 14.7 TFLOPS. Shifting the focus to constraint, the usefulness of a high performance graphics single-precision, the peak processing power for the Harold processor system in a tactical environment is explored in this system is 240 TFLOPS and 29.5 TFLOPS for the JVN study. A line-of-sight optimization algorithm serves as a system. Given that a single graphics card, Radeon HD 6990, compute-intensive application with characteristics relating is rated at 5.1 TFLOPS for peak single-precision arithmetic, to a tactical scenario.
    [Show full text]
  • IMG PPT Template
    Imagination Technologies Group Overview May 2010 Company Overview . Leading Semiconductor IP Supplier . POWERVR™ graphics, video, display processing . ENSIGMA™ receivers and communications processors . META™ processors – SoC centric real-time, DSP, Linux . Licensees: Leading Semis and OEMs . #3 Silicon Design IP provider * . Innovative Consumer Product Manufacturer . PURE digital radio, internet connected audio . Established technology powerhouse . Founded:1985 . Listed:1994-London Stock Exchange:IMG . Employees: more than 650 worldwide * Source: Gartner IP Suppliers Report, March 2010 UK Headquarters R&D Sales 2.2 © Imagination Technologies Ltd. Group Revenues £m 65 60 PURE 55 Technology 50 45 40 35 30 25 20 15 10 5 0 04/05 05/06 06/07 07/08** 08/09 * 13 months to April ‘08 3 © Imagination Technologies Ltd. Everything, Everywhere The connected “four screens”: TV, PC, Mobile & In-Car The Cloud Mobile devices have to deliver the internet, PC capabilities, TV, music, navigation and more… 4 © Imagination Technologies Ltd. Our Technology Business SoC Technologies & Solutions Markets Mobile IMGworks Phone Customer’s 3rd Party IP Internal IP Customised IP Multimedia ENSIGMA META POWERVR Communications Hand-held Processor/DSP Graphics Multimedia POWERVR Display META POWERVR Audio Video SoC Home Consumer & Entertainment Display Audio Mobile Computing OpenGL MeOS Radio OpenGL-ES De-interlace Nucleus TV OpenVG Linux Frame rate In-Car Mobile TV DirectX conversion Android Electronics WiFi etc Audio Bluetooth Noise SW Stacks reduction etc MPEG1/2/4 etc H264 etc Emerging VC1 Markets etc 5 © Imagination Technologies Ltd. The Virtuous Technology Cycle… Semiconductor Partners Creates SoCs & Sells Chips SUNPLUS Makes products Creates & & sells them to Licenses IP consumers… OEMs Emerging trend with OEM partners Consumers demand more, better, cheaper…everywhere Consumers …which drives demand for more advanced & efficient IP … 6 © Imagination Technologies Ltd.
    [Show full text]
  • Renesas Technology to Release SH7776 (SH-Navi3), Industry’S First Dual-Core Soc with Built-In Image Recognition Processing Function for Car Information Terminals
    Renesas Technology to Release SH7776 (SH-Navi3), Industry’s First Dual-Core SoC with Built-in Image Recognition Processing Function for Car Information Terminals ⎯ Achieves excellent performance of 1,920 MIPS and a single-chip solution for next- generation car information systems ⎯ Tokyo, January 19, 2009 — Renesas Technology Corp. today announced the SH7776 (SH-Navi3), a dual-core system-on-chip (SoC) device with on-chip enhanced graphics functions and a high- performance image recognition processing function for the next generation high-performance car information terminals that evolved from car navigation systems. The SH7776 (SH-Navi3) integrates two CPU cores on a single chip and achieves superior processing performance of 1,920 million instructions per second (MIPS), twice that of comparable earlier single-chip SoC products from Renesas Technology. Sample shipments will begin in April 2009 in Japan. Renesas Technology has the top share*1 in the worldwide market for microprocessors for car navigation systems, and the SH-Navi Series,*2 a single-chip SoC device implementing functions essential for high-performance car navigation systems such as 3-D graphics, is already used in many high-end products. Car navigation systems will be evolving into “next-generation car information centers,” which play the central role in car information control to assure convenience, safety, and environmental considerations while driving. The new SH7776 dual-core SoC device delivers on a single chip the performance and functions essential to such a “next-generation car information center.” It enables display of colorful and realistic 3-D graphics use of a multimedia and information communication as well as navigations, display of 2-D/3-D graphics use of a graphical user interface (GUI)*3 for enhanced user-friendliness, and image recognition for functions such as lane detection and preceding vehicle tracking.
    [Show full text]
  • GPU Developments 2017T
    GPU Developments 2017 2018 GPU Developments 2017t © Copyright Jon Peddie Research 2018. All rights reserved. Reproduction in whole or in part is prohibited without written permission from Jon Peddie Research. This report is the property of Jon Peddie Research (JPR) and made available to a restricted number of clients only upon these terms and conditions. Agreement not to copy or disclose. This report and all future reports or other materials provided by JPR pursuant to this subscription (collectively, “Reports”) are protected by: (i) federal copyright, pursuant to the Copyright Act of 1976; and (ii) the nondisclosure provisions set forth immediately following. License, exclusive use, and agreement not to disclose. Reports are the trade secret property exclusively of JPR and are made available to a restricted number of clients, for their exclusive use and only upon the following terms and conditions. JPR grants site-wide license to read and utilize the information in the Reports, exclusively to the initial subscriber to the Reports, its subsidiaries, divisions, and employees (collectively, “Subscriber”). The Reports shall, at all times, be treated by Subscriber as proprietary and confidential documents, for internal use only. Subscriber agrees that it will not reproduce for or share any of the material in the Reports (“Material”) with any entity or individual other than Subscriber (“Shared Third Party”) (collectively, “Share” or “Sharing”), without the advance written permission of JPR. Subscriber shall be liable for any breach of this agreement and shall be subject to cancellation of its subscription to Reports. Without limiting this liability, Subscriber shall be liable for any damages suffered by JPR as a result of any Sharing of any Material, without advance written permission of JPR.
    [Show full text]
  • The Supercomputer in Your Pocket
    14 The supercomputer in your pocket Over the past decade, the computing landscape has shifted from beige boxes under desks to a mix of laptops, smartphones, tablets, and hybrid devices. This explosion of mobile CPUs is a profound shift for the semiconductor industry and will have a dramatic impact on its competitive intensity and structure. Harald Bauer, The shift toward mobile computing, at the expense Two major contests will play out for semi- Yetloong Goh, of tethered CPUs, is a major change that has conductor vendors competing in the mobile Joseph Park, raised the competitive metabolism of the semi- arena: the clash between vertical-integration Sebastian Schink, and conductor sector. Mobile, now the central and horizontal-specialization business models, Christopher Thomas battleground of the technology industry, is having and the clash between the two dominant an intense impact on the larger semiconductor architectures and ecosystems, ARM and x86. landscape. Mobile-computing processor require- Each of these battles will be explored in ments now drive the industry, setting design detail below. requirements for transistor structures, process generations, and CPU architecture. It’s the More and more smartphones are as capable as must-win arena for all the semiconductor sector’s the computers of yesteryear. Laptops have top companies. Over the next half decade, leading displaced desktops as the most popular form players will spend hundreds of billions of dollars factor for PCs, and, thanks to the success on R&D. of Apple’s iPad, tablets have stormed into the 15 marketplace. PC original equipment manu- assumes a doubling of processing power every facturers (OEMs) aren’t waiting to lose consumer 18 months, holds the same path for the next few share of wallet to tablets.
    [Show full text]
  • Openvg BOF 5 Imagination Technologies
    POWERVRPOWERVR MBXMBX && SGXSGX OpenVGOpenVG SupportSupport andand ResourcesResources Kristof Beets 3rd Party Relations Manager - Imagination Technologies [email protected] © Copyright Khronos Group, 2006 - Page 1 ImaginationImagination TechnologiesTechnologies Our Mission: Deliver complementary, flexible, high performance silicon Intellectual property (IP) cores that are cost-effective for high volume System On Chip (SoC) applications, and enable industry-leading product innovation Key Facts: • Founded: 1985 • Listed: 1994 (London Stock Exchange) • Employees: more than 400 worldwide …of which >80% are engineers • Revenues: £48.1M ($94.8M) (FY 2006/07) • Offices: HQ: UK Sales: Japan, US, Taiwan, Korea, China R&D: UK, India Our goal is to be one of the top 3 global semiconductor Design IP providers © Copyright Khronos Group, 2006 - Page 2 POWERVRPOWERVR MBXMBX FamilyFamily • OpenGL ES 1.x Compliant • OpenVG 1.0 Support • Family Members - POWERVR MBX - POWERVR MBX Lite • Tile Based Deferred Rendering - Market Proven Advanced Tiling Algorithms - Order-independent Hidden Surface Removal POWERVR MBX: The de-facto standard for - Lowest silicon area, bandwidth and power mobile graphics acceleration - Excellent system latency tolerance >50 POWERVR 3D-enabled phones shipping • High Quality Performance Texture Filtering POWERVR MBX • POWERVR Texture Compression: 2bpp and 4bpp Family • High Quality, High Performance Anti-Aliasing OpenGL ES1.1 • Internal True Color Direct3D Mobile • DOT3 Per-pixel Lighting OpenVG 1.0 • Optional POWERVR VGP Triangles/Sec 1.7M … 3.7M - Dedicated programmable Vertex Processing Unit Pixels/Sec 135M … 300M - Allows high polygon throughput - Advanced features: Skinning, Curved Surfaces, Lighting Performance quoted at 100MHz for MBX, MBX Lite. SoC achievable performance quoted, Performance scales with clock speeds up to 200MHz and beyond.
    [Show full text]
  • NBAY 6120 Lecture 6 Donald P
    Computer Graphics Software & Hardware NBAY 6120 Lecture 6 Donald P. Greenberg March 16, 2016 Recommended Readings for Lecture 6 • Mike Seymour. “The State of Rendering, Part 1,” fxguide.com, July 15, 2013. FXGuide. • Mike Seymour. “The State of Rendering, Part 2,” fxguide.com, July 17, 2013. FXGuide. Why Is It Important? • 99% of our information intake is pictorial through our eyes • Educational Modules • Entertainment • Games • Advertising • Medical • Computer Aided Design • Data Visualization Ivan Sutherland 1963 General Electric - 1967 Program of Computer Graphics, Cornell University DPG - 1967 Program of Computer Graphics, Cornell University Cornell in Perspective Film 1972 Program of Computer Graphics, Cornell University Gouraud Flat Polygon Shading 1972 Each polygon is shaded based on a single normal. Gouraud Thesis Gouraud Smooth Shading 1972 Each pixel is shaded by interpolating intensities computed in each of the polygon’s vertices. Gouraud Thesis Phong Shading 1974 Model Model • Environment Geometry & topology Camera Material properties >Color, reflectance, textures Perspective >(Cost, strength, thermal properties) Raster Operations Image • Lighting Storage Geometry & position Intensity, spectral distribution Display Direction, spatial distribution Model Camera • Viewer Position Camera • Viewer direction Perspective • Field of view Raster Wide angle Operations Telephoto Image Storage • Depth of focus Near Display Far Perspective Model Transformation Camera • Perspective transformation Matrix multiplication (4 x 4) Perspective
    [Show full text]
  • 3D Augmented Reality Mobile Application Prototype for Visual Planning Support
    3D Augmented Reality Mobile Application Prototype for Visual Planning Support Arnau Fombuena Valero Master’s of Science Thesis in Geoinformatics TRITA-GIT EX 11-010 School of Architecture and the Built Environment Royal Institute of Technology (KTH) Stockholm, Sweden November 2011 Abstract The aim of this thesis is to implement a prototype of a 3D Augmented Reality mobile application. Using 3D is becoming more and more common for professionals as well as users. A good example of that is Google Earth and its 3D view. Implementing a mobile application that takes advantage of 3D and Augmented Reality may be very useful for planning new constructions in both urban and non-urban areas allowing to visualize how the construction will be in the future and how it will interact with its surrounding environment. There is a great potential for such kind of applications. An example could be the modification of a certain area of a city; allowing the inhabitants of that city to preview the project and, hopefully, avoiding unnecessary conflicts related to that project. In non- urban areas this application is also very useful for helping decision making by visualizing, on site, how the project will be and its impact on the environment. In order to preview a future construction there is the need to have a 3D model. Therefore, a 3D format for that model is necessary. Since COLLADA is a 3D standard interexchange format it is used in this thesis. Together with COLLADA, the computer graphics imagery and gaming technology called OpenGL ES 2.0 is used. Using COLLADA and OpenGL ES 2.0 combined with the properties of the views’ layers, the camera input, the sensors in the mobile device and the positioning technologies permit obtaining successful results displaying a 3D object in an Augmented Reality mobile prototype application.
    [Show full text]
  • New Igr Study Forecasts the Number of Active Volte Subscribers Worldwide to Grow at a CAGR of 165 Percent Over Five Years
    Contact iGR Iain Gillott (512) 263-5682 [email protected] FOR IMMEDIATE RELEASE New iGR study forecasts the number of active VoLTE subscribers worldwide to grow at a CAGR of 165 percent over five years Study also discusses impact of VoLTE on consumers in each region AUSTIN, Texas, March 20th, 2015 – VoLTE, or Voice over LTE, is very important to mobile operators because this technology enables them to eventually terminate their legacy, circuit- switched voice 2G/3G network infrastructure in favor of consolidating voice traffic on the LTE network platform. Gradually, VoLTE will allow many operators to refarm spectrum away from 2G/3G to LTE, which will significantly lower voice infrastructure costs. In the last 12 months, many operators, principally in the U.S., Japan and Asia Pacific, have deployed VoLTE and started offering a range of services, including HD voice. The Apple iPhone 6 models and the Samsung Galaxy S4, 5, and 6, among others, all support VoLTE, and this strong support for VoLTE among the most popular smartphone models will only serve to increase the penetration of VoLTE users as more networks are launched. iGR expects the deployments to continue and has forecasted the number of active VoLTE subscribers to grow at a CAGR of 165 percent between 2014 and 2019. “With the many new VoLTE deployments of the last year, VoLTE has started to impact not only mobile operators, but also consumers," said Iain Gillott, president and founder of iGR, a market research consultancy focused on the wireless and mobile industry. “Consumers are seeing the benefit of many new multimedia products and services that mobile operators can offer as they move from legacy voice solutions to the new platform.” iGR’s new market study, Global VoLTE Market Forecast, 2014-2019: A Growing Global Service, provides an overview of VoLTE, a discussion of its benefits, an updated status on the market, and profiles of the major vendors.
    [Show full text]
  • Renderering Roblox
    Rendering Roblox Vulkan Optimisations on PowerVR Presented By: Arseny Kapoulkine David Hamill 1 What is Roblox? • Online multiplayer game creation platform • All content is user generated • Windows, macOS, iOS, Android, Xbox One • 100M+ MAU, 2.5M+ CCU Graphics scaling • Substantial differences between devices • PowerVR SGX543: Apple iPad 2 (2011), Samsung Galaxy S4 (2013) • PowerVR GM 9446: Oppo Reno Z (2019) • Developers make the content once • It’s our job to run it well* Full spectrum design • How do you implement a graphics feature? • Build for the lowest common denominator and upscale? • Build for the highest end GPU and optimize? • Build a feature for the lowest common denominator and upscale? • Pro: Consistent behavior • Con: Poor quality • Build a feature for the highest end GPU and optimize? • Pro: Great quality on high-end • Con: Poor performance on low-end, inconsistent behavior, battery/thermal limits Full spectrum design • How do you implement a graphics feature? • Build for the lowest common denominator and upscale? • Build for the highest end GPU and optimize? • Sometimes we pick one and iterate • Prefer to design for the entire spectrum from the beginning Quality vs performance Quality vs performance • Advanced settings give users lots of control! However… • Suboptimal performance for users who can’t tune these well • … or know about them! Especially on mobile. • Performance is often counter-intuitive • CPU vs GPU bottlenecks • Example: denser grass is faster • Different games have different performance characteristics
    [Show full text]