AMD Introduces World's Most Powerful 16- Core

Total Page:16

File Type:pdf, Size:1020Kb

AMD Introduces World's Most Powerful 16- Core November 7, 2019 AMD Introduces World’s Most Powerful 16- core Consumer Desktop Processor, the AMD Ryzen™ 9 3950X – AMD Ryzen™ 9 3950X rounds out 3rd Gen Ryzen desktop processor series, arriving November 25 – – New AMD Athlon™ 3000G processor to provide everyday users with unmatched performance per dollar, coming November 19 – SANTA CLARA, Calif., Nov. 07, 2019 (GLOBE NEWSWIRE) -- Today, AMD announced the release of the highly anticipated flagship 16-core AMD Ryzen 9 3950X processor, available worldwide November 25, 2019. AMD Ryzen 9 3950X processor brings the ultimate processor for gamers with effortless 1080P gaming in select titles1 and up to 2X more energy efficient processing power compared to the competition2 as the world’s fastest 16- core consumer desktop processor3. In addition, AMD also announced a significant performance uplift4 coming for mainstream desktop users with the new AMD Athlon 3000G, arriving November 19, 2019. “We are excited to bring the AMD Ryzen™ 9 3950X to market later this month, offering enthusiasts the most powerful 16-core desktop processor ever,” said Chris Kilburn, corporate vice president and general manager, client channel, AMD. “We are focused on offering the best solutions at every level of the market, including the AMD Athlon 3000G for everyday PC users that delivers great performance at an incredible price point.” AMD Ryzen 9 3950X: Fastest 16-core Consumer Desktop Processor Offering up to 22% performance increase over previous generations5, the AMD Ryzen 9 3950X offers faster 1080p gaming in select titles1 and content creation6 than the competition. Built on the industry-leading “Zen 2” architecture, the AMD Ryzen 9 3950X also excels in power efficiency3 with a TDP7 of 105W. As the fastest 16-core consumer desktop processor available3, AMD Ryzen 9 3950X processor offers incredible performance best experienced with a liquid cooling solution. AMD is recommending the use of an AIO solution with a minimum 280mm radiator. A list of AMD recommended coolers can be found on AMD.com to ensure enthusiasts can maximize the potential of the Ryzen 9 3950X processor. MODEL CORES/ BOOST8/ TOTAL TDP7 PLATFORM PCIe® Gen SEP (USD)9 AVAILABILITY THREADS BASE CACHE (MB) (WATTS) 4.0 LANES FREQUENCY (processor + (GHZ) AMD X570) AMD Ryzen™ 16/32 Up to 4.7/3.5 72 105W AM4 44 (36 $749 Nov 25, 2019 9 3950X useable) AMD Athlon 3000G: Versatile Everyday Processor, Unlocked AMD also announced a significant upgrade coming for mainstream desktop users with the new AMD Athlon 3000G processor, arriving November 19, 2019. The new AMD Athlon 3000G with Radeon™ Graphics brings reliable computing experiences to a wide range of users, from day-to-day needs to high-definition PC gaming, offering faster framerates when gaming and enhanced productivity performance over competition4 at an incredible price point. The Athlon 3000G is the first “Zen”-based Athlon processor that is unlocked for overclocking potential, delivering the only unlocked processor in its segment10. MODEL CORES/ PROCESSOR GRAPHICS TDP7 (WATTS) PLATFORM SEP (USD)9 AVAILABILITY THREADS FREQUENCY (GHZ) AMD Athlon™ 2/4 3.5 Radeon™ Vega 35W AM4 $49 Nov 19, 2019 3000G 3 AMD AM4 Platform Developments To ensure the best experiences for all users, this month, AMD released AGESA version 1004 to its motherboard ecosystem. Containing more than 150 updates, AGESA 1004 offers significant improvements for the AM4 platform focused primarily on stability. Key improvements include AMD X570 stability and compatibility with add-in devices, PCIe® device support and stability, and interoperability of PCIe®, USB, SATA, and device reset capabilities, alongside performance enhancements with fastest core utilization and further boost frequency optimizations for the AMD Ryzen™ 9 3900X. More information on the specific updates AGESA 1004 can be found here. Supporting Resources Lean more about AMD Ryzen 9 3950X Desktop Processor Learn more about AMD Athlon Desktop Processors Become a fan of AMD on Facebook Follow AMD on Twitter About AMD For 50 years AMD has driven innovation in high-performance computing, graphics and visualization technologies ― the building blocks for gaming, immersive platforms and the datacenter. Hundreds of millions of consumers, leading Fortune 500 businesses and cutting- edge scientific research facilities around the world rely on AMD technology daily to improve how they live, work and play. AMD employees around the world are focused on building great products that push the boundaries of what is possible. For more information about how AMD is enabling today and inspiring tomorrow, visit the AMD (NASDAQ:AMD) website, blog, Facebook and Twitter pages. The information contained herein is for informational purposes only, and is subject to change without notice. Timelines, roadmaps, and/or product release dates shown in herein are plans only and subject to change. “Zen” and “Zen 2” are codenames for AMD architectures, and are not product names. GD-122. 1 Testing by AMD Performance Labs as of 9/10/2019 using an AMD Ryzen™ 9 3950X processor, Core i9-9920X. Games tested at 1920x1080 with high in-game quality preset. Results may vary. RZ3-74 2 Testing by AMD performance labs using an AMD Ryzen™ 9 3950X, Core i9-9920X and Core i9-9900K, measuring wall power during Cinebench R20 nT. Results may vary. RZ3-76​ 3 Testing by AMD Performance Labs on 09/15/2019, comparing the AMD Ryzen 9 3950X (AMD’s fastest 16-core) to the Intel Core i9-9960X (Intel’s fastest 16-core), using the Cinebench R20 single-core benchmark score and Cinebench R20 multi-core benchmark score to measure single-core and multi-core performance for each processor. Performance results may vary. RZ3-72 4 Testing done by AMD performance labs on 9/15/2019. Gamer performance as represented by the following games tested using Windows 10 at 720p and low settings: Fortnite, Rocket League, Counter Strike: Global Offensive. Application performance as represented by the following applications tested using Windows 10: Cinebench R20, PCMark 10, Lame MP3 Encoder, Adobe Premiere and Adobe Photoshop. Results may vary with configuration. All Intel Pentium G5000 series processors have locked multipliers, but the Athlon 3000G which competes in the same entry-level segment has an unlocked multiplier. ATG-11 5 Testing by AMD Performance Labs as of 6/03/2019 utilizing 3rd Gen AMD Ryzen™ processors: 3950X, 3900X, 3800X, 3700X, 3600X, 3600 and Ryzen™ 7 2700X in Cinebench R20 1T. Results may vary. RZ3-25 6 Testing by AMD Performance Labs as of 9/10/2019 using an AMD Ryzen™ 9 3950X processor, Core i9-9920X, and Core i9-9900K in: DaVinci Resolve, Adobe Premiere, Cinebench R20, Handbrake 1.1.1, LLVM Compile Time, POV-Ray 3.7, and V-Ray. Results may vary. RZ3-75 7 Though both are often measured in watts, it is important to distinguish between thermal and electrical watts. Thermal wattage for processors is conveyed via thermal design power (TDP). TDP is a calculated value that conveys an appropriate thermal solution to achieve the intended operation of a processor. Electrical watts are not a variable in the TDP calculation. By design, electrical watts can vary from workload to workload and may exceed thermal watts. GD-109 8 Max boost for AMD Ryzen PRO Processors is the maximum frequency achievable by a single core on the processor running a bursty single-threaded workload. Max boost will vary based on several factors, including, but not limited to: thermal paste; system cooling; motherboard design and BIOS; the latest AMD chipset driver; and the latest OS updates. GD-150 9 AMD Suggested Retail Price in USD. Price subject to change. 10 Overclocking AMD processors, including without limitation, altering clock frequencies / multipliers or memory timing / voltage, to operate beyond their stock specifications will void any applicable AMD product warranty, even when such overclocking is enabled via AMD hardware and/or software. This may also void warranties offered by the system manufacturer or retailer. Users assume all risks and liabilities that may arise out of overclocking AMD processors, including, without limitation, failure of or damage to hardware, reduced system performance and/or data loss, corruption or vulnerability. GD-106 Contact: Sophia Hong AMD Communications (512) 917-9998 [email protected] Laura Graves AMD Investor Relations (408) 749-5467 [email protected] Source: Advanced Micro Devices.
Recommended publications
  • AMD Ryzen™ PRO & Athlon™ PRO Processors Quick Reference Guide
    AMD Ryzen™ PRO & Athlon™ PRO Processors Quick Reference guide AMD Ryzen™ PRO Processors with Radeon™ Graphics for Business Laptops (Socket FP6/FP5) 1 Core/Thread Frequency Boost/Base L2+L3 Cache Graphics Node TDP Intel vPro Core/Thread Frequency Boost*/Base L2+L3 Cache Graphics Node TDP AMD PRO technologies COMPARED TO 4.9/1.1 Radeon™ 6/12 UHD AMD Ryzen™ 7 PRO 8/16 Up to 12MB Graphics 7nm 15W intel Intel Core i7 10810U GHz 13MB 4750U 4.1/1.7 GHz CORE i7 14nm 15W (7 Cores) 10th Gen Intel Core i7 10610U 4/8 4.9/1.8 9MB UHD GHz AMD Ryzen™ 7 PRO Up to Radeon™ intel 4/8 6MB 10 12nm 15W 4.8/1.9 3700U 4.0/2.3 GHz Vega CORE i7 Intel Core i7 8665U 4/8 9MB UHD 14nm 15W 8th Gen GHz Radeon™ AMD Ryzen™ 5 PRO 6/12 Up to 11MB Graphics 7nm 15W intel 4.4/1.7 4650U 4.0/2.1 GHz CORE i5 Intel Core i5 10310U 4/8 7MB UHD 14nm 15W GHz (6 Cores) 10th Gen AMD Ryzen™ 5 PRO 4/8 Up to 6MB Radeon™ 12nm 15W intel 4.1/1.6 3500U 3.7/2.1 GHz Vega8 CORE i5 Intel Core i5 8365U 4/8 7MB UHD 14nm 15W th GHz 8 Gen Radeon™ AMD Ryzen™ 3 PRO 4/8 Up to 6MB Graphics 7nm 15W intel 4.1/2.1 4450U 3.7/2.5 GHz CORE i3 Intel Core i3 10110U 2/4 5MB UHD 14nm 15W (5 Cores) 10th Gen GHz AMD Ryzen™ 3 PRO 4/4 Up to 6MB Radeon™ 12nm 15W intel 3.9/2.1 3300U 3.5/2.1 GHz Vega6 CORE i3 Intel Core i3 8145U 2/4 4.5MB UHD 14nm 15W 8th Gen GHz AMD Athlon™ PRO Processors with Radeon™ Vega Graphics for Business Laptops (Socket FP5) AMD Athlon™ PRO Up to Radeon™ intel Intel Pentium 4415U 2/4 2.3 GHz 2.5MB HD 610 14nm 15W 300U 2/4 3.3/2.4 GHz 5MB Vega3 12nm 15W 1.
    [Show full text]
  • Amd(Amd.Us)18Q1 点评 2018 年 07 月 30 日
    海外公司报告 | 公司动态研究 证券研究报告 AMD(AMD.US)18Q1 点评 2018 年 07 月 30 日 作者 AMD 7 年最佳,10 年翻身,重申买入,TP 上调至 何翩翩 分析师 23 美元 SAC 执业证书编号:S1110516080002 [email protected] 业绩超预期,7 年来最佳盈利季 雷俊成 分析师 SAC 执业证书编号:S1110518060004 AMD 18Q2 实现 7 年来最佳盈利季度,non-GAAP EPS 0.14 美元,营收 17.6 [email protected] 亿美元同比大涨 53%,均超过华尔街预期的 EPS 0.13 美元和营收 17.2 亿美 马赫 分析师 SAC 执业证书编号:S1110518070001 元。计算与图形业务同比大涨 64%至 10.9 亿美元好于市场预期的 10.6 亿, [email protected] 但受 Q2 区块链相关贡献进一步减弱带来该业务环比跌 3%。挖矿业务本季 董可心 联系人 营收占比从上季的 10%降低为 6%,公司进一步看淡下半年需求。EESC 业务 [email protected] 同比涨 37%至 6.7 亿美元,好于预期的 6.61 亿,EPYC 逐步进入放量阶段, 公司维持到年底会实现中单位数份额的预测。Q2 毛利率提升至 37%,Q3 指引营收 17 亿美元,同比增长 7%,略低于市场预期的 17.6 亿,毛利率提 相关报告 升至约 38%;全年指引营收增速保持 25%,我们认为公司指引基于 17Q3 的 1 《AMD(AMD.US)点评:EPYC“从 高基数较为保守,且区块链影响作为一次性业务逐渐消弭也会进一步减少 零到一”终实现,7nm 产品周期全方位 业绩不确定性,我们看好 EPYC 会在下半年至 Q4 迎来关键放量。 回归“传奇”;TP 上调至 22 美元,重 服务器市场 AMD 与 Intel“荣辱互见” 申买入》2018-06-20 2 《AMD(AMD.US)点评:公布 7nm 服务器市场 AMD 与 Intel“荣辱互见”,EPYC 服务器随着 Cisco、HPE 适配 GPU 加入 AI 计算抢滩战,Ryzen+EPYC 以及超级云计算客户的需求能见度提高,Q2 出货量和营收均环比提高超 50%,目前与 AMD 合作的 5 个云计算巨头成主要推动力。我们认为 AMD “双子星”仍是中流砥柱;TP 上调至 将继续通过单插槽服务器高核心数和低功耗打造性价比优势,下半年加速 18 美元,重申买入》2018-06-08 市场渗透蚕食 Intel 份额,进入明年则等待 7nm 的第二代 EPYC 面市,面对 3 《AMD(AMD.US)18Q1 点评:2018 已将 10nm Cannon Lake 量产时点延后至明年的 Intel,AMD 将终于实现制 开门红,业绩指引均超预期,Ryzen 继 程反超,加速量价齐升。“从零到一”抢占 20 亿美元以上的市场份额。 续扎实闪耀,EPYC 仍待升级放量,重 反观 Intel Q2 数据中心业务收入 55.5 亿美元,虽然在整体行业高景气度下 申买入》2018-04-30 同比增长 27%,但仍低于市场预期的 56.3 亿美元。业绩发布会上 Intel 更为 4 《2017 扭亏为盈业绩迎拐点,2018 明确消费级 10nm 产品会到 19 年下半年节日旺季才推向市场,让市场情绪 厚积待薄发,Ryzen+EPYC 继续双星闪 愈加悲观的同时也给了 AMD 足够的时间窗口。 耀,重申买入》2018-02-01 Ryzen 继续攻城略地,进一步打开笔记本市场 5 《AMD(AMD.US)点评:合作英特
    [Show full text]
  • Software-Based Undervolting Faults in AMD Zen Processors Fehler in AMD Zen Prozessoren Durch Software-Basierte Unterspannung
    Software-based Undervolting Faults in AMD Zen Processors Fehler in AMD Zen Prozessoren durch Software-basierte Unterspannung Bachelorarbeit im Rahmen des Studiengangs IT-Sicherheit der Universität zu Lübeck vorgelegt von Anja Rabich ausgegeben und betreut von Prof. Dr. Thomas Eisenbarth mit Unterstützung von Luca Wilke Lübeck, den 31. August 2020 Abstract Dynamic Voltage and Frequency Scaling (DVFS) is a powerful performance enhance- ment method used by modern processors, allowing them to scale voltage or frequency as needed based on the power requirements of the CPU. This not only saves power, but also prevents processors from overheating. However, the continued integration of soft- ware interfaces giving a user direct access to this functionality has been shown to be a potential security risk, allowing a privileged adversary to indirectly tamper with sensitive computations. This thesis summarizes the results of various papers showing that using DVFS features, unsuitable voltage/frequency values can be set for the processor leading to hardware faults and calculation errors which can be used to undermine the integrity of Trusted Execution Environments (TEE). Results are partially replicated for Intel’s TEE implementation SGX, followed by extending the same methodology to AMD’s Zen Pro- cessors, on which there is currently no information. Results show that undervolting is an unlikely attack vector. iii Zusammenfassung Dynamische Spannungs- und Frequenzskalierung (engl. DVFS) ist ein in modernen Prozessoren vorhandener Leistungs- und Stromverwaltungsmechanismus, womit Span- nung und Frequenz der CPU je nach Bedarf skaliert werden können. Somit wird nicht nur Strom gespart, sondern auch zusätzlich verhindert, dass der Prozessor überhitzt. Die zunehmende Integration von Softwareschnittstellen zu diesen Mechanismen die dem Nutzer Einstellungsmöglichkeiten anbieten, haben sich zunehmend als potenzielle Sicherheitslücke erwiesen.
    [Show full text]
  • High Performance Linpack Benchmark on AMD EPYC™ Processors
    High Performance Linpack Benchmark on AMD EPYC™ Processors This document details running the High Performance Linpack (HPL) benchmark using the AMD xhpl binary. HPL Implementation: The HPL benchmark presents an opportunity to demonstrate the optimal combination of multithreading (via the OpenMP library) and MPI for scientific and technical high-performance computing on the EPYC architecture. For MPI applications where the per-MPI-rank work can be further parallelized, each L3 cache is an MPI rank running a multi-threaded application. This approach results in fewer MPI ranks than using one rank per core, and results in a corresponding reduction in MPI overhead. The ideal balance is for the number of threads per MPI rank to be less than or equal to the number of CPUs per L3 cache. The exact maximum thread count per MPI rank depends on both the specific EPYC SKU (e.g. 32 core parts have 4 physical cores per L3, 24 core parts have 3 physical cores per L3) and whether SMT is enabled (e.g. for a 32 core part with SMT enabled there are 8 CPUs per L3). HPL performance is primarily determined by DGEMM performance, which is in turn primarily determined by SIMD throughput. The Zen microarchitecture of the EPYC processor implements one SIMD unit per physical core. Since HPL is SIMD limited, when SMT is enabled using a second HPL thread per core will not directly improve HPL performance. However, leaving SMT enabled may indirectly allow slightly higher performance (1% - 2%) since the OS can utilize the SMT siblings as needed without pre-empting the HPL threads.
    [Show full text]
  • AMD's Early Processor Lines, up to the Hammer Family (Families K8
    AMD’s early processor lines, up to the Hammer Family (Families K8 - K10.5h) Dezső Sima October 2018 (Ver. 1.1) Sima Dezső, 2018 AMD’s early processor lines, up to the Hammer Family (Families K8 - K10.5h) • 1. Introduction to AMD’s processor families • 2. AMD’s 32-bit x86 families • 3. Migration of 32-bit ISAs and microarchitectures to 64-bit • 4. Overview of AMD’s K8 – K10.5 (Hammer-based) families • 5. The K8 (Hammer) family • 6. The K10 Barcelona family • 7. The K10.5 Shanghai family • 8. The K10.5 Istambul family • 9. The K10.5-based Magny-Course/Lisbon family • 10. References 1. Introduction to AMD’s processor families 1. Introduction to AMD’s processor families (1) 1. Introduction to AMD’s processor families AMD’s early x86 processor history [1] AMD’s own processors Second sourced processors 1. Introduction to AMD’s processor families (2) Evolution of AMD’s early processors [2] 1. Introduction to AMD’s processor families (3) Historical remarks 1) Beyond x86 processors AMD also designed and marketed two embedded processor families; • the 2900 family of bipolar, 4-bit slice microprocessors (1975-?) used in a number of processors, such as particular DEC 11 family models, and • the 29000 family (29K family) of CMOS, 32-bit embedded microcontrollers (1987-95). In late 1995 AMD cancelled their 29K family development and transferred the related design team to the firm’s K5 effort, in order to focus on x86 processors [3]. 2) Initially, AMD designed the Am386/486 processors that were clones of Intel’s processors.
    [Show full text]
  • AMD Raven Ridge
    DELIVERING A NEW LEVEL OF VISUAL PERFORMANCE IN AN SOC AMD “RAVEN RIDGE” APU Dan Bouvier, Jim Gibney, Alex Branover, Sonu Arora Presented by: Dan Bouvier Corporate VP, Client Products Chief Architect AMD CONFIDENTIAL RAISING THE BAR FOR THE APU VISUAL EXPERIENCE Up to MOBILE APU GENERATIONAL 200% MORE CPU PERFORMANCE PERFORMANCE GAINS Up to 128% MORE GPU PERFORMANCE Up to 58% LESS POWER FIRST “Zen”-based APU CPU Performance GPU Performance Power HIGH-PERFORMANCE AMD Ryzen™ 7 2700U 7th Gen AMD A-Series APU On-die “Vega”-based graphics Scaled GPU Managed Improved Upgraded Increased LONG BATTERY LIFE and CPU up to power delivery memory display package Premium form factors reach target and thermal bandwidth experience performance frame rate dissipation efficiency density 2 | AMD Ryzen™ Processors with Radeon™ Vega Graphics - Hot Chips 30 | * See footnotes for details. “RAVEN RIDGE” APU AMD “ZEN” x86 CPU CORES CPU 0 “ZEN” CPU CPU 1 (4 CORE | 8 THREAD) USB 3.1 NVMe PCIe FULL PCIe GPP ----------- ----------- Discrete SYSTEM 4MB USB 2.0 SATA GFX CONNECTIVITY CPU 2 CPU 3 L3 Cache X64 DDR4 HIGH BANDWIDTH SOC FABRIC System Infinity Fabric & MEMORY Management SYSTEM Unit ACCELERATED Platform Multimedia Security MULTIMEDIA Processor Engines AMD GFX+ 1MB L2 EXPERIENCE X64 DDR4 (11 COMPUTE UNITS) Cache Video Audio Sensor INTEGRATED CU CU CU CU CU CU Display Codec ACP Fusion Controller Next SENSOR Next Hub FUSION HUB CU CU CU CU CU AMD “VEGA” GPU UPGRADED DISPLAY ENGINE 3 | AMD Ryzen™ Processors with Radeon™ Vega Graphics - Hot Chips 30 | SIGNIFICANT DENSITY INCREASE “Raven Ridge” die BGA Package: 25 x 35 x 1.38mm Technology: GLOBALFOUNDRIES 14nm – 11 layer metal Transistor count: 4.94B 59% 16% Die Size: 209.78mm2 more transistors smaller die than prior generation “Bristol Ridge” APU 4 | AMD Ryzen™ Processors with Radeon™ Vega Graphics - Hot Chips 30 | * See footnotes for details.
    [Show full text]
  • SMBIOS Specification
    1 2 Document Identifier: DSP0134 3 Date: 2019-10-31 4 Version: 3.4.0a 5 System Management BIOS (SMBIOS) Reference 6 Specification Information for Work-in-Progress version: IMPORTANT: This document is not a standard. It does not necessarily reflect the views of the DMTF or its members. Because this document is a Work in Progress, this document may still change, perhaps profoundly and without notice. This document is available for public review and comment until superseded. Provide any comments through the DMTF Feedback Portal: http://www.dmtf.org/standards/feedback 7 Supersedes: 3.3.0 8 Document Class: Normative 9 Document Status: Work in Progress 10 Document Language: en-US 11 System Management BIOS (SMBIOS) Reference Specification DSP0134 12 Copyright Notice 13 Copyright © 2000, 2002, 2004–2019 DMTF. All rights reserved. 14 DMTF is a not-for-profit association of industry members dedicated to promoting enterprise and systems 15 management and interoperability. Members and non-members may reproduce DMTF specifications and 16 documents, provided that correct attribution is given. As DMTF specifications may be revised from time to 17 time, the particular version and release date should always be noted. 18 Implementation of certain elements of this standard or proposed standard may be subject to third party 19 patent rights, including provisional patent rights (herein "patent rights"). DMTF makes no representations 20 to users of the standard as to the existence of such rights, and is not responsible to recognize, disclose, 21 or identify any or all such third party patent right, owners or claimants, nor for any incomplete or 22 inaccurate identification or disclosure of such rights, owners or claimants.
    [Show full text]
  • CPU Benchmarks - List of Benchmarked Cpus
    11.09.2020 PassMark - CPU Benchmarks - List of Benchmarked CPUs CPU Benchmarks CPU Benchmarks Over 1,000,000 CPUs Benchmarked CPU List Below is an alphabetical list of all CPU types that appear in the charts. Clicking on a specific processor name will take you to the chart it appears in and will highlight it for you. Results for Single CPU Systems and Multiple CPU Systems are listed separately. Find CPU Single CPU Systems Multi CPU Systems CPUS High End Single CPU Systems High Mid Range Last updated on the 11th of September 2020 Low Mid Range Low End Column CPU Mark Rank CPU Value Price Best Value CPU Name (higher is (lower is (higher is (USD) (On Market) better) better) better) Best Value XY AMD 3015e 2,678 1285 NA NA Scatter Best Value AMD 3020e 2,721 1272 NA NA (All time) AMD A4 Micro-6400T APU 1,004 2126 NA NA New Desktop AMD A4 PRO-3340B 1,519 1790 NA NA New Laptop AMD A4 PRO-7300B APU 1,421 1839 NA NA Single Thread AMD A4 PRO-7350B 1,024 2108 NA NA Systems with AMD A4-1200 APU 445 2572 NA NA Multiple CPUs Overclocked AMD A4-1250 APU 432 2583 NA NA Power AMD A4-3300 APU 994 2131 76.50 $12.99 Performance CPU Mark by Socket AMD A4-3300M APU 665 2394 22.19 $29.99* Type Cross-Platform CPU AMD A4-3305M APU 807 2273 38.78 $20.81 Performance AMD A4-3310MX APU 844 2239 NA NA CPU Mega List AMD A4-3320M APU 877 2212 23.77 $36.90 Search Model AMD A4-3330MX APU 816 2265 NA NA 0 1,067 2078 Compare AMD A4-3400 APU 53.35 $20.00 https://www.cpubenchmark.net/cpu_list.php AMD A4 3420 APU 8 01 $12 9 * 1/87 11.09.2020 PassMark - CPU Benchmarks - List
    [Show full text]
  • Best Practice Guide Modern Processors
    Best Practice Guide Modern Processors Ole Widar Saastad, University of Oslo, Norway Kristina Kapanova, NCSA, Bulgaria Stoyan Markov, NCSA, Bulgaria Cristian Morales, BSC, Spain Anastasiia Shamakina, HLRS, Germany Nick Johnson, EPCC, United Kingdom Ezhilmathi Krishnasamy, University of Luxembourg, Luxembourg Sebastien Varrette, University of Luxembourg, Luxembourg Hayk Shoukourian (Editor), LRZ, Germany Updated 5-5-2021 1 Best Practice Guide Modern Processors Table of Contents 1. Introduction .............................................................................................................................. 4 2. ARM Processors ....................................................................................................................... 6 2.1. Architecture ................................................................................................................... 6 2.1.1. Kunpeng 920 ....................................................................................................... 6 2.1.2. ThunderX2 .......................................................................................................... 7 2.1.3. NUMA architecture .............................................................................................. 9 2.2. Programming Environment ............................................................................................... 9 2.2.1. Compilers ........................................................................................................... 9 2.2.2. Vendor performance libraries
    [Show full text]
  • Take a Way: Exploring the Security Implications of AMD's Cache Way
    Take A Way: Exploring the Security Implications of AMD’s Cache Way Predictors Moritz Lipp Vedad Hadžić Michael Schwarz Graz University of Technology Graz University of Technology Graz University of Technology Arthur Perais Clémentine Maurice Daniel Gruss Unaffiliated Univ Rennes, CNRS, IRISA Graz University of Technology ABSTRACT 1 INTRODUCTION To optimize the energy consumption and performance of their With caches, out-of-order execution, speculative execution, or si- CPUs, AMD introduced a way predictor for the L1-data (L1D) cache multaneous multithreading (SMT), modern processors are equipped to predict in which cache way a certain address is located. Conse- with numerous features optimizing the system’s throughput and quently, only this way is accessed, significantly reducing the power power consumption. Despite their performance benefits, these op- consumption of the processor. timizations are often not designed with a central focus on security In this paper, we are the first to exploit the cache way predic- properties. Hence, microarchitectural attacks have exploited these tor. We reverse-engineered AMD’s L1D cache way predictor in optimizations to undermine the system’s security. microarchitectures from 2011 to 2019, resulting in two new attack Cache attacks on cryptographic algorithms were the first mi- techniques. With Collide+Probe, an attacker can monitor a vic- croarchitectural attacks [12, 42, 59]. Osvik et al. [58] showed that tim’s memory accesses without knowledge of physical addresses an attacker can observe the cache state at the granularity of a cache or shared memory when time-sharing a logical core. With Load+ set using Prime+Probe. Yarom et al. [82] proposed Flush+Reload, Reload, we exploit the way predictor to obtain highly-accurate a technique that can observe victim activity at a cache-line granu- memory-access traces of victims on the same physical core.
    [Show full text]
  • EPYC: Designed for Effective Performance
    EPYC: Designed for Effective Performance By Linley Gwennap Principal Analyst June 2017 www.linleygroup.com EPYC: Designed for Effective Performance By Linley Gwennap, Principal Analyst, The Linley Group Measuring server-processor performance using clock speed (GHz) or even the traditional SPEC_int test can be misleading. AMD’s new EPYC processor is designed to deliver strong performance across a wide range of server applications, meeting the needs of modern data centers and enterprises. These design capabilities include advanced branch prediction, data prefetching, coherent interconnect, and integrated high-bandwidth DRAM and I/O interfaces. AMD sponsored the creation of this white paper, but the opinions and analysis are those of the author. Trademark names are used in an editorial fashion and are the property of their respective owners. Although many PC users can settle for “good enough” performance, data-center opera- tors are always seeking more. Web searches demand more performance as the Internet continues to expand. Newer applications such as voice recognition (for services such as Alexa and Siri) and analyzing big data also require tremendous performance. Neural networks are gaining in popularity for everything from image recognition to self-driving cars, but training these networks can tie up hundreds of servers for days at a time. Processor designers must meet these greater performance demands while staying within acceptable electrical-power ratings. Server processors are often characterized by core count and clock speed (GHz), but these characteristics provide only a rough approximation of application performance. As important as speed is, the amount of work that a processor can accomplish with each tick of the clock, a parameter known as instructions per cycle (IPC), is equally important.
    [Show full text]
  • AMD Zen Rohin, Vijay, Brandon Outline
    AMD Zen Rohin, Vijay, Brandon Outline 1. History and Overview 2. Datapath Structure 3. Memory Hierarchy 4. Zen 2 Improvements History and Overview AMD History ● IBM production too large, forced Intel to license their designs to 3rd parties ● AMD fills the gap, produces clones for 15ish years - legal battles ensued ● K5 first in-house x86 chip in 1996 ● Added more features like out of order, L2 caches, etc ● Current CPUs are Zen* tomshardware.com/picturestory/71 3-amd-cpu-history.html Zen Brand ● Performance desktop and mobile computing ○ Athlon ○ Ryzen 3, Ryzen 5, Ryzen 7, Ryzen 9 ○ Ryzen Threadripper ● Server ○ EPYC https://en.wikichip.org/wiki/amd/microarchitectures/zen Zen History ● Aimed to replace two of AMD’s older chips ○ Excavator: high performance architecture ○ Puma: low power architecture https://en.wikichip.org/wiki/amd/microarchitectures/zen#Block_Diagram Zen Architecture ● Quad-core ● Fetch 4 instructions/cycle ● Op cache 2k instructions ● 168 physical integer registers ● 72 out of order loads ● Large shared L3 cache ● 2 threads per core https://www.slideshare.net/AMD/amd-epyc-microp rocessor-architecture Datapath Structure Fetch ● Decoupled branch predictor ○ Runs ahead of fetches ○ Successful predictions help latency and memory parallelism ○ Mispredictions incur power penalty ● 3 layer TLB ○ L0: 8 entries ○ L1: 64 entries ○ L2: 512 entries https://www.anandtech.com/show/10591/amd-zen-microarchiture-p art-2-extracting-instructionlevel-parallelism/3 Branch Predictor ● Perceptron: simple neural network ● Table of perceptrons, each a vector of weights ● Branch address used to access perceptron table ● Dot product between weight vector and branch history vector Perceptron Branch Predictor ● ~10% improve prediction rates over gshare predictor - (2, 2) correlating predictor ● Can utilize longer branch histories ○ Hardware requirements scale linearly whereas they scale exponentially for other predictors D.
    [Show full text]