Architecture Day 2020

Performance Index

ID	Date	Classification
615781	12/05/2024	Public

Demos

Intel® Xe-LP Enhanced Gaming on a Thin & Light Platform

XE-LP system 1: Preproduction TGL-U FFD; Memory: LPDDR4-4267, 32 GB; Driver: Intel UHD Graphics gfx-driver-ci-master-5252; Storage: Samsung 512GB SSD; OS: Win10-RS5 Build 18363.720; Application: Playerunknown’s Battlegrounds, Grid, Mount & Blade II: Bannerlord, Doom Eternal
XE-LP system 2: TGL-U RVP; platforms set to fixed CPU frequency and max GT Frequency, Memory: LPDDR4-4267, 16GB, 2ch 2r; Driver: Intel UHD Graphics gfx-driver-ci-master-4610; Storage: Intel SSD 660p; OS: Win10-RS5 Build 18363.720; Application: Battlefield V, Battlefield 1, Grid
Gen 11 Baseline system: ICL-U RVP; platforms set to fixed CPU frequency and max GT Frequency, Memory: LPDDR4-3733, 16GB, 2ch 2r; Driver: Intel UHD Graphics gfx-driver-ci-master-4610; Storage: Intel SSD 660p; OS: Win10-RS5 Build 18363.720; Application: Battlefield 1, Grid
Xe-LP Tiger Lake 15W Proxy: TGL-U RVP: pre-production platform set to lower fixed CPU & GT frequency to approximate 15W, Memory: LPDDR4-4267, 16GB, 2ch 2r; Driver: Intel UHD Graphics gfx-driver-ci-master-4610; Storage: Intel SSD 660p; OS: Win10-RS5 Build 18363.720; Application: Battlefield 1
Testing by Intel between 5/13/20 – 7/31/20

Gen over Gen AI Acceleration on Intel GPUs

System 1: WHL-U RVP; Memory: DDR4-2400, 16GB; Driver: Intel UHD Graphics gfx-driver-ci-master-5252; Storage: Intel SSD 760p; OS: Win10-RS5 Build 18363; Application: Gigapixel AI
System 2: ICL-U RVP; Memory: DDR4-2133, 16GB; Driver: Intel UHD Graphics gfx-driver-ci-master-5252; Storage: Intel SSD 760p; OS: Win10-RS5 Build 18363; Application: Gigapixel AI
System 3: TGL-U RVP: Memory: LPDDR4-4267, 16GB; Driver: Intel UHD Graphics gfx-driver-ci-master-5252; Storage: Intel SSD 760p; OS: Win10-RS5 Build 18363; Application: Gigapixel AI
Testing by Intel as of 6/30/20

Xe-HP

System: XE-HP 1T/2T/4T; Processor-1S ICX-HCC (connected to Whitley Host); Memory: DDR-3200, 32GB; Driver: 5.4.17 (Ubuntu); Storage: 1000GB SATA HDD; OS: Ubuntu 19.10; Applications: FFmpeg & clpeak
“Most FP32 peak performance in a single GPU package” based on FP32 peak performance, as measured by the Clpeak benchmark, on the 2T and 4T packages for Xe-HP architecture. These packages perform at over 21TF and over 42TF of peak FP32 performance, respectively, as compared to the A100 architecture (https://www.nvidia.com/en-us/data-center/a100/, last accessed August 7, 2020), and the AMD Radeon Instinct architecture (https://www.amd.com/en/graphics/servers-radeon-instinct-mi, last accessed August 7, 2020).
Testing by Intel as of 8/4/20

Fast Inferencing using Intel’s DL Boost with Bfloat16

System: 3rd Gen Intel® Xeon® Scalable Processor: 1-node, 4x 3rd Gen Intel® Xeon® Platinum 8380H processor (pre-production 28C, 250W) on Intel Reference Platform (Cooper City) with 384 GB (24 slots / 16GB / 3200) total memory, ucode 0x700001b, HT on, Turbo on, with Ubuntu 20.04 LTS, Linux 5.4.0-29-generic, Intel SSD 800GB OS Drive, ResNet-50 v1.5 Throughput, https://github.com/Intel-tensorflow/tensorflow -b bf16/base, commit#6ef2116e6a09, Modelzoo: https://github.com/IntelAI/models/ -b v1.6.1, Imagenet dataset, oneDNN 1.4, FP32, BF16, global BS=1024, 4 instances, 28-cores/instance, test by Intel on 06/01/2020.

Intelligent Shading with Intel Xe Graphics

System: TGL-U RVP; platform set to fixed CPU frequency and max GT Frequency, Memory: LPDDR4-4267, 16GB, 2ch 2r; Driver: Intel UHD Graphics gfx-driver-ci-master-4610; Storage: Intel SSD 660p; OS: Win10-RS5 Build 18363.720; Application: Grid
Testing by Intel as of 7/14/20

Frequency Uplift with Willow Cove CPU Core

Baseline system: ICL-U RVP; Memory: LPDDR4-3733, 16GB, 2ch 2r; Storage: Intel SSD 760p; OS: Win10 19H1-18362.30; Application: WebXPRT 3 benchmark
Tiger Lake system: TGL-U RVP; Memory: LPDDR4-4267, 16GB, 2ch 2r; Storage: Intel SSD 660p; OS: Win10 20H1; Application: WebXPRT 3 benchmark
Testing by Intel as of 5/19/20

Other

Lakefield claims

Up to 24% higher power efficiency
- Based on WebXPRT*3 score per watt on Intel® Core™ i5-L16G7 processor (LFK) PL1=7W TDP, 5C5T, Memory: 2X4GB LPDDR4x - 4267MHz, Storage: Intel® 760p m.2 PCIe NVMe SSD, Display Resolution: 1920x1080 for Performance; 25x14 eDP 13.3”, OS: Windows 10 19H2 (18363.535), Graphics driver: 26.20.100.7916 as compared to Intel® Core™ i7-8500Y processor(AML-Y 22), PL1=5.0W TDP, 2C4T, Turbo up to 4.2GHz, Memory: 2x4GB LPDDR3-1866, Storage: Intel® 760p SSD, Display Resolution:1920x1080, Windows 10 Build RS3 17134.112, Graphics driver:whl.1006167-v2.
- Testing by Intel as of April 2020.
Up to 91% lower standby power
- Based on Intel® Core™ i5-L16G7 processor (LFK) PL1=7W TDP, 5C5T, Memory: 2X4GB LPDDR4x - 4267MHz, Storage: Intel® 760p m.2 PCIe NVMe SSD, Display Resolution: 1920x1080 for Performance; 25x14 eDP 13.3”, OS: Windows 10 19H2 (18363.535), Graphics driver: 26.20.100.7916 as compared to Intel® Core™ i7-8500Y processor(AML-Y 22), PL1=5.0W TDP, 2C4T, Turbo up to 4.2GHz, Memory: 2x4GB LPDDR3-1866, Storage: Intel® 760p SSD, Display Resolution:1920x1080, Windows 10 Build RS3 17134.112, Graphics driver:whl.1006167-v2.
- Testing by Intel as of April 2020.
Up to 33% higher web performance
- As measured by WebXPRT*3 score on Intel® Core™ i5-L16G7 processor (LFK) PL1=7W TDP, 5C5T, Memory: 2X4GB LPDDR4x - 4267MHz, Storage: Intel® 760p m.2 PCIe NVMe SSD, Display Resolution: 1920x1080 for Performance; 25x14 eDP 13.3”, OS: Windows 10 19H2 (18363.535), Graphics driver: 26.20.100.7916, with Sunny Cove on and off.
- Testing by Intel as of April 2020.
Up to 17% higher power efficiency
- As measured by WebXPRT*3 score per Watt on Intel® Core™ i5-L16G7 processor (LFK) PL1=7W TDP, 5C5T, Memory: 2X4GB LPDDR4x - 4267MHz, Storage: Intel® 760p m.2 PCIe NVMe SSD, Display Resolution: 1920x1080 for Performance; 25x14 eDP 13.3”, OS: Windows 10 19H2 (18363.535), Graphics driver: 26.20.100.7916, with Sunny Cove on and off.
- Testing by Intel as of April 2020.

Agilex performance and power claims

Based on internal builds of Intel® Quartus® Prime Pro Design Software on a wide variety of internal benchmarks. The computer systems used for the evaluations were Intel® Xeon® CPU E5-2690 v4 @ 2.60GHz class machines running Red Hat Enterprise Linux Server 6.10 (Santiago) operating system. The performance results represent average improvements across a wide variety of internal benchmarks, and results may vary for each testcase.
Testing by Intel as of January 2020.

Scale out DLRM training performance claims

Based on Intel® Xeon® Platinum 8280 CPU; Bios: SE5C620.86B.02.01.0009.092820190230; Memory: 12x 18ASF2G72PDZ-2G6D1; Network: 2x Intel Corporation Omni-Path HFI Silicon 100 Series; OS: Cent OS 7.6; Bootline: BOOT_IMAGE=/vmlinuz-3.10.0-957.el7.x86_64 rhgb quiet intel_pstate=disable nmi_watchdog=0 intel_idle.max_cstate=1 nohz_full=2-111 LANG=en_US.UTF-8; OPA stack: 10.9.2.0-7; Pytorch: 1.4.1. 26 sockets refers to 13 dual-socket nodes and 1 socket refers to 1 node of Intel Xeon 8280. 1 socket optimization was achieved with Intel extension of PyTorch.
Testing by Intel as of March - April 2020.

Notices & Disclaimers

Performance varies by use, configuration and other factors.

Intel optimizations, for Intel compilers or other products, may not optimize to the same degree for non-Intel products.

Performance results are based on testing as of dates shown in configurations and may not reflect all publicly available updates. See above for configuration details. No product or component can be absolutely secure.

Results that are based on pre-production systems and components as well as results that have been estimated or simulated using an Intel Reference Platform (an internal example new system), internal Intel analysis or architecture simulation or modeling are provided to you for informational purposes only. Results may vary based on future changes to any systems, components, specifications, or configurations. Intel technologies may require enabled hardware, software or service activation.

Intel contributes to the development of benchmarks by participating in, sponsoring, and/or contributing technical support to various benchmarking groups, including the BenchmarkXPRT Development Community administered by Principled Technologies.

Statements in this presentation that refer to future plans and expectations are forward-looking statements that involve a number of risks and uncertainties. Words such as "anticipates," "expects," "intends," "goals," "plans," "believes," "seeks," "estimates," "continues," "may," "will," "would," "should," "could," and variations of such words and similar expressions are intended to identify such forward-looking statements. Statements that refer to or are based on estimates, forecasts, projections, uncertain events or assumptions, including statements relating to future products and technology and the expected availability and benefits of such products and technology, market opportunity, and anticipated trends in our businesses or the markets relevant to them, also identify forward-looking statements. Such statements are based on management's current expectations and involve many risks and uncertainties that could cause actual results to differ materially from those expressed or implied in these forward-looking statements. Important factors that could cause actual results to differ materially from the company's expectations are set forth in Intel's earnings release dated July 23, 2020, which is included as an exhibit to Intel's Form 8-K furnished to the SEC on such date, and Intel's SEC filings, including the company's most recent reports on Forms 10-K and 10-Q. Copies of Intel's Form 10-K, 10-Q and 8-K reports may be obtained by visiting our Investor Relations website at www.intc.com or the SEC's website at www.sec.gov. Intel does not undertake, and expressly disclaims any duty, to update any statement made in this presentation, whether as a result of new information, new developments or otherwise, except to the extent that disclosure may be required by law.

Code names are used by Intel to identify products, technologies, or services that are in development and not publicly available. These are not "commercial" names and not intended to function as trademarks.

Intel does not control or audit third-party data. You should consult other sources to evaluate accuracy.