Microsoft Describes Next Xbox SoC

发布时间:2017-08-23 00:00
作者:Ameya360
来源:Rick Merritt
阅读量:1190

  Microsoft described Scorpio, the SoC inside its Xbox X One. The device, detailed at Hot Chips, aims to balance cost and performance for 4K gaming in a console shipping in November.

  Scorpio packs 7 billion transistors with 6 TFlops graphics performance in a 359 mm2 chip made in a 16FF+ TSMC process. The chip, designed in partnership with AMD, also packs eight x86 cores running at 2.3 GHz and sharing 4 Mbytes L2 cache.

  Scorpio uses 12 GBytes of GDDR5 memory rather than 32 MBytes embedded SRAM used in a prior generation or a high-bandwidth memory (HBM) module which engineers considered. “We were a little nervous making a consumer product with data moving around at 6.8 GHz,” said John Sell, a Microsoft distinguished engineer, referring to the GDDR5 data rate.

  The single large block of memory enables much simpler tools for application developers, Sell said. It also emulates the embedded SRAM for compatibility with apps written for the prior console.

  Microsoft kicked the tires of the kinds of HBM modules AMD pioneered in GPUs in 2015 and Nvidia uses now on Volta. “But for a consumer product HBM2 is too expensive and inflexible…its memory bandwidth is not as granular, and we would be locked into [an HBM] module,” Sell said.

  It’s too early to tell what the next generation trade-offs will look like, but both a Jedec GDDR6 and an HBM3 are in the works. A major cost issue for HBM is a lack of test coverage and thus relatively low yields, he said.

  Other memory alternatives may include modules using lower cost organic substrates than the silicon interposer in HBM. Intel has its own alternative called EMIB for its foundry customers. However, Microsoft aimed to reuse many blocks, particularly some custom analog designs from the existing Xbox One S that also was made in a TSMC 16FF+ process.

  The 16FF+ process “is quite mature now — we actually used it for the Xbox One S — but this is still one of TSMC’s largest 16FF+ chips,” he said.

  The majority of Scorpio consists of 40 graphics processing units, four spares and shader arrays running at a core frequency of 1.172 GHz.

  AMD’s Zen x86 core was not ready in time for the SoC which uses eight older AMD CPUs. However, they are newer and faster units than on the closest rival console, Sell said.

  The SoC supports HEVC encode and decode at 60 Hz. It also includes eight custom audio processors enhanced for spatial audio. A separate south bridge includes I/O for PCIe, USB and a SATA hard disk. The entire system will sell for about $500, including an integrated Blu-ray drive.

(备注:文章来源于网络,信息仅供参考,不代表本网站观点,如有侵权请联系删除!)

在线留言询价

相关阅读
Microsoft First In-House AI Chip “Maia” Produced by TSMC’s 5nm
  On the 15th, Microsoft introducing its first in-house AI chip, “Maia.” This move signifies the entry of the world’s second-largest cloud service provider (CSP) into the domain of self-developed AI chips. Concurrently, Microsoft introduced the cloud computing processor “Cobalt,” set to be deployed alongside Maia in selected Microsoft data centers early next year. Both cutting-edge chips are produced using TSMC’s advanced 5nm process, as reported by UDN News.  Amidst the global AI fervor, the trend of CSPs developing their own AI chips has gained momentum. Key players like Amazon, Google, and Meta have already ventured into this territory. Microsoft, positioned as the second-largest CSP globally, joined the league on the 15th, unveiling its inaugural self-developed AI chip, Maia, at the annual Ignite developer conference.  These AI chips developed by CSPs are not intended for external sale; rather, they are exclusively reserved for in-house use. However, given the commanding presence of the top four CSPs in the global market, a significant business opportunity unfolds. Market analysts anticipate that, with the exception of Google—aligned with Samsung for chip production—other major CSPs will likely turn to TSMC for the production of their AI self-developed chips.  TSMC maintains its consistent policy of not commenting on specific customer products and order details.  TSMC’s recent earnings call disclosed that 5nm process shipments constituted 37% of Q3 shipments this year, making the most substantial contribution. Having first 5nm plant mass production in 2020, TSMC has introduced various technologies such as N4, N4P, N4X, and N5A in recent years, continually reinforcing its 5nm family capabilities.  Maia is tailored for processing extensive language models. According to Microsoft, it initially serves the company’s services such as $30 per month AI assistant, “Copilot,” which offers Azure cloud customers a customizable alternative to Nvidia chips.  Borkar, Corporate VP, Azure Hardware Systems & Infrastructure at Microsoft, revealed that Microsoft has been testing the Maia chip in Bing search engine and Office AI products. Notably, Microsoft has been relying on Nvidia chips for training GPT models in collaboration with OpenAI, and Maia is currently undergoing testing.  Gulia, Executive VP of Microsoft Cloud and AI Group, emphasized that starting next year, Microsoft customers using Bing, Microsoft 365, and Azure OpenAI services will witness the performance capabilities of Maia.  While actively advancing its in-house AI chip development, Microsoft underscores its commitment to offering cloud services to Azure customers utilizing the latest flagship chips from Nvidia and AMD, sustaining existing collaborations.  Regarding the cloud computing processor Cobalt, adopting the Arm architecture with 128 core chip, it boasts capabilities comparable to Intel and AMD. Developed with chip designs from devices like smartphones for enhanced energy efficiency, Cobalt aims to challenge major cloud competitors, including Amazon.
2023-11-17 16:00 阅读量:1643
Microsoft Teams Up with Taiwan’s AI LAB to Jointly Announce AI Gene Analysis Platform
Taiwan Microsoft and Taiwan’s Artificial Intelligence Laboratory (AI LAB) have jointly announced the TaiGenomics Artificial Intelligence gene analysis platform. Integrated with Microsoft’s Azure Cloud Computing, the platform conducts gene sequencing, analysis, and diagnosis effectively through the use of AI to assist medical specialists in the rapid prediction of potential diseases caused by genetic variations.Taiwan AI LAB Founder Ethan Tu stated that Taiwan has an advantage in the development of precision medical care, and in addition to having a solid foundation in the communications industry, Taiwan’s National Health Insurance Database, medical records, and medical electronic imaging also provide robust databases for genetic testing.This collaboration with Microsoft utilizes the advantages of Azure to develop TaiGenomics, which analyzes the necessary space and cloud computing resources necessary for the analysis of medical big data. It is anticipated that this collaboration will enable TaiGenomics to develop Taiwan’s strengths and drive a new ecology through integration with the medical treatment ecosystem. Furthermore, there is an opportunity to begin this endeavor in Taiwan and then extend it to the rest of the world.General Manager of Microsoft Taiwan Sun Jikang stated: AI has always been Microsoft’s goal, and we anticipate that the joint development of the TaiGenomics gene analysis platform with Taiwan’s AI LAB will serve as a connection with other team research services and experts to expand the precision medical ecosystem.In order to integrate with Azure cloud computing, massive amounts of data are processed for gene detection using AI, which drastically reduces data comparison time and gene testing costs. It also assist medical specialists in the rapid prediction of potential diseases. As for the resources to establish TaiGenomics, they will all be placed in cloud storage in order to solve the problem of massive data needed for genetic testing.
2019-01-03 00:00 阅读量:1128
Citrix partners Microsoft to simplify office network deployments
Microsoft, MediaTek Team on IoT SoC
  MediaTek is working with Microsoft to deliver this year the first Azure Sphere chip, the MT3620 that the partners expect to drive IoT innovation with built-in security and connectivity.  Azure Sphere is designed for highly secured, connected MCU-powered devices at a price that the companies expect will make enterprise-class security affordable for an array of cloud-connected gadgets.  MediaTek and Microsoft have worked together to develop a specialized chipset with a Wi-Fi connected controller built around a processor designed to run Azure Sphere’s IoT operating system that includes support for Microsoft’s latest security protocols. The chipsets will be sold as part of the Azure Sphere solution, enabling customers to connect their MCU-powered products and devices with protection and security provided by Microsoft.  "Microsoft had a vision we also believed in. Bringing trust and security into IoT solutions from chip to cloud, to make connected devices practical to sell and manage for businesses and consumers," said Jerry Yu, a MediaTek vice president, in a press statement.  MediaTek is the reference chip partner for the Azure Sphere Partner Program.  Cost and Security  The partners expect a significant drop in the cost of connectivity will allow billions of MCU-powered devices, from household appliances and health monitors to children’s toys and industrial equipment to become IoT devices.  Yet each connection is vulnerable to attacks. IoT won’t become a mass market until devices, data and cloud are secure. That’s the fundamental problem the two companies aim to solve with their latest project.  With more than 9 billion microcontroller-powered devices entering the market annually, Microsoft and MediaTek see as critical the need to ensure every connected device, regardless of price point, has the highest level of security embedded.  "In the next decade we’ll see the democratization of connectivity to billions of devices,” said Microsoft Managing Director Galen Hunt. "Azure Sphere brings together the best of Microsoft’s expertise in cloud, software, and silicon - resulting in a unique approach to security that starts in the silicon and extends to the cloud."  The MediaTek MT3620 is the first Azure Sphere-certified chip, which the companies expect will be the foundation for a new generation of secure intelligent edge devices and solutions.  With Azure Sphere, Microsoft aims to build a new ecosystem, consisting of silicon vendors, ODMs and OEMs from a broad range of industries.  Recognizing the opportunities and risks associated with the growing number of IoT devices and applications, a group of companies have banded together to ensure that security standards are developed and upheld, according to Microsoft.  MediaTek is now sampling chips with lead customers and expects to ramp production by the third quarter of 2018.
2018-04-18 00:00 阅读量:1085
  • 一周热料
  • 紧缺物料秒杀
型号 品牌 询价
BD71847AMWV-E2 ROHM Semiconductor
TL431ACLPR Texas Instruments
RB751G-40T2R ROHM Semiconductor
CDZVT2R20B ROHM Semiconductor
MC33074DR2G onsemi
型号 品牌 抢购
STM32F429IGT6 STMicroelectronics
ESR03EZPJ151 ROHM Semiconductor
BP3621 ROHM Semiconductor
BU33JA2MNVX-CTL ROHM Semiconductor
IPZ40N04S5L4R8ATMA1 Infineon Technologies
TPS63050YFFR Texas Instruments
热门标签
ROHM
Aavid
Averlogic
开发板
SUSUMU
NXP
PCB
传感器
半导体
相关百科
关于我们
AMEYA360微信服务号 AMEYA360微信服务号
AMEYA360商城(www.ameya360.com)上线于2011年,现 有超过3500家优质供应商,收录600万种产品型号数据,100 多万种元器件库存可供选购,产品覆盖MCU+存储器+电源芯 片+IGBT+MOS管+运放+射频蓝牙+传感器+电阻电容电感+ 连接器等多个领域,平台主营业务涵盖电子元器件现货销售、 BOM配单及提供产品配套资料等,为广大客户提供一站式购 销服务。