Login
Login

Tegra#Tegra 4

{{short description|System on a chip by Nvidia}}

{{for|the genus of bush cricket|Tegra (katydid)}}

{{Use mdy dates|date=April 2014}}

File:NVIDIA T20 and T30 chips.jpg

File:NVIDIA@20nm@TegraX1@Erista@Shield TV@S Taiwan 1517A1 NPW020.M3W TM670D-A1 DSC00807 (32803128780).jpg

Tegra is a system on a chip (SoC) series developed by Nvidia for mobile devices such as smartphones, personal digital assistants, and mobile Internet devices. The Tegra integrates an ARM architecture central processing unit (CPU), graphics processing unit (GPU), northbridge, southbridge, and memory controller onto one package. Early Tegra SoCs are designed as efficient multimedia processors. The Tegra-line evolved to emphasize performance for gaming and machine learning applications without sacrificing power efficiency, before taking a drastic shift in direction towards platforms that provide vehicular automation with the applied "Nvidia Drive" brand name on reference boards and its semiconductors; and with the "Nvidia Jetson" brand name for boards adequate for AI applications within e.g. robots or drones, and for various smart high level automation purposes.

History

The Tegra APX 2500 was announced on February 12, 2008. The Tegra 6xx product line was revealed on June 2, 2008,{{cite web|url=http://www.techtree.com/India/News/Nvidia_Rolls_out_Tegra_Processors/551-89833-581.html|archive-url=https://web.archive.org/web/20080604024432/http://www.techtree.com/India/News/Nvidia_Rolls_out_Tegra_Processors/551-89833-581.html|url-status=dead|archive-date=June 4, 2008|title=Techtree.com India > News > Hardware > Nvidia Rolls out "Tegra" Chips|date=June 4, 2008}} and the APX 2600 was announced in February 2009. The APX chips were designed for smartphones, while the Tegra 600 and 650 chips were intended for smartbooks and mobile Internet devices (MID).{{cite web|url=http://www.nvidia.com/docs/IO/55043/NVIDIA_Tegra_FAQ_External.pdf|website=Nvidia.com|access-date=2008-06-04|title=NVIDIA Tegra FAQ|archive-date=March 20, 2012|archive-url=https://web.archive.org/web/20120320093928/http://www.nvidia.com/docs/IO/55043/NVIDIA_Tegra_FAQ_External.pdf|url-status=live}}

The first product to use the Tegra was Microsoft's Zune HD media player in September 2009, followed by the Samsung M1.{{cite web |url=http://tugatech.com.pt/t3108-nvidia-prepara-tegra-3-a-15-ghz |title=Nvidia prepara Tegra 3 a 1,5 GHz |publisher=TugaTech |date=2011-01-27 |access-date=2016-07-10 |archive-date=October 16, 2017 |archive-url=https://web.archive.org/web/20171016174636/https://tugatech.com.pt/t3108-nvidia-prepara-tegra-3-a-15-ghz |url-status=live }} Microsoft's Kin was the first cellular phone to use the Tegra;{{cite web |date=2010-04-13 |title=Microsoft's Kin are the first Tegra smartphones – PC World Australia |url=http://www.pcworld.idg.com.au/article/342826/microsoft_kin_first_tegra_smartphones/ |url-status=dead |archive-url=https://web.archive.org/web/20171016225935/https://www.pcworld.idg.com.au/article/342826/microsoft_kin_first_tegra_smartphones/ |archive-date=October 16, 2017 |access-date=2016-07-10 |website=Pcworld.idg.com.au}} however, the phone did not have an app store, so the Tegra's power did not provide much advantage. In September 2008, Nvidia and Opera Software announced that they would produce a version of the Opera 9.5 browser optimized for the Tegra on Windows Mobile and Windows CE.{{cite press release | url = http://www.opera.com/press/releases/2008/09/09/ | title = Nvidia and Opera team to accelerate the full Web on mobile devices | publisher = Opera Software | date = 2008-09-09 | access-date = 2009-01-09 | archive-date = March 30, 2012 | archive-url = https://web.archive.org/web/20120330170307/http://www.opera.com/press/releases/2008/09/09/ | url-status = live }}{{cite press release | url = http://www.nvidia.com/object/io_1220962341614.html | title = Nvidia And Opera Team To Accelerate The Full Web On Mobile Devices | publisher = NVIDIA | date = 2008-09-09 | access-date = 2009-04-17 | archive-date = December 24, 2011 | archive-url = https://web.archive.org/web/20111224110401/http://www.nvidia.com/object/io_1220962341614.html | url-status = live }} At Mobile World Congress 2009, Nvidia introduced its port of Google's Android to the Tegra.

On January 7, 2010, Nvidia officially announced and demonstrated its next generation Tegra system-on-a-chip, the Nvidia Tegra 250, at Consumer Electronics Show 2010.{{cite web|url=http://www.nvidia.com/object/io_1262837617533.html|title=New Nvidia Tegra Processor Powers The Tablet Revolution|access-date=2010-03-19|work=Nvidia|date=January 7, 2010|archive-date=December 24, 2018|archive-url=https://web.archive.org/web/20181224142539/https://www.nvidia.com/object/io_1262837617533.html|url-status=live}} Nvidia primarily supports Android on Tegra 2, but booting other ARM-supporting operating systems is possible on devices where the bootloader is accessible. Tegra 2 support for the Ubuntu Linux distribution was also announced on the Nvidia developer forum.{{cite press release | url = http://developer.nvidia.com/beta-forum#/discussion/46/what-operating-systems-does-tegra-support | title = What operating systems does Tegra support? | publisher = NVIDIA | date = 2011-08-17 | access-date = 2011-09-14 | archive-date = September 3, 2011 | archive-url = https://web.archive.org/web/20110903021625/http://developer.nvidia.com/beta-forum#/discussion/46/what-operating-systems-does-tegra-support | url-status = live }}

Nvidia announced the first quad-core SoC at the February 2011 Mobile World Congress event in Barcelona. Though the chip was codenamed Kal-El, it is now branded as Tegra 3. Early benchmark results show impressive gains over Tegra 2,{{cite web |url=http://www.brightsideofnews.com/news/2011/2/21/why-nvidiae28099s-tegra-3-is-faster-than-a-core-2-duo-t7200.aspx |title=Why nVidia's Tegra 3 is faster than a Core 2 Duo T7200 |publisher=Brightsideofnews.com |date=February 21, 2011 |access-date=2011-08-12 |archive-date=August 23, 2011 |archive-url=https://web.archive.org/web/20110823235058/http://www.brightsideofnews.com/news/2011/2/21/why-nvidiae28099s-tegra-3-is-faster-than-a-core-2-duo-t7200.aspx |url-status=live }}{{cite web |last=Hruska |first=Joel |url=http://hothardware.com/News/Nvidias-KalEl-Demonstration--Marred-By-Benchmark-Shenanigans/ |title=Nvidia's Kal-El Demonstration Marred By Benchmark Confusion |publisher=HotHardware |date=2011-02-22 |access-date=2016-07-15 |archive-date=February 26, 2012 |archive-url=https://web.archive.org/web/20120226140535/http://hothardware.com/News/Nvidias-KalEl-Demonstration--Marred-By-Benchmark-Shenanigans/ |url-status=dead }} and the chip was used in many of the tablets released in the second half of 2011.

In January 2012, Nvidia announced that Audi had selected the Tegra 3 processor for its In-Vehicle Infotainment systems and digital instruments display.{{cite web |url=http://eetimes.com/electronics-news/4234777/Audi-selects-Tegra3-processor-for-infotainment--dashboard |title=Audi selects Tegra processor for infotainment and dashboard |publisher=EE Times |date=2012-01-18 |access-date=2016-07-15 |archive-date=January 20, 2012 |archive-url=https://web.archive.org/web/20120120000330/http://www.eetimes.com/electronics-news/4234777/Audi-selects-Tegra3-processor-for-infotainment--dashboard |url-status=live }} The processor will be integrated into Audi's entire line of vehicles worldwide, beginning in 2013. The process is ISO 26262-certified.{{cite web|url=https://blogs.nvidia.com/blog/2016/07/15/automotive-grade/|title=What Is Automotive Grade? Here's What It Means|work=The Official NVIDIA Blog|date=July 15, 2016|access-date=2016-10-11|archive-date=October 11, 2016|archive-url=https://web.archive.org/web/20161011142810/https://blogs.nvidia.com/blog/2016/07/15/automotive-grade/|url-status=live}}

In summer of 2012 Tesla Motors began shipping the Model S all electric, high performance sedan, which contains two NVIDIA Tegra 3D Visual Computing Modules (VCM). One VCM powers the 17-inch touchscreen infotainment system, and one drives the 12.3-inch all digital instrument cluster."{{cite web | url=http://www.nvidia.com/object/automotive-infotainment-navigation.html | title=Tegra Automotive Infotainment and Navigation | publisher=NVIDIA | access-date=2013-01-04 | archive-date=January 23, 2013 | archive-url=https://web.archive.org/web/20130123033812/http://www.nvidia.com/object/automotive-infotainment-navigation.html | url-status=live }}

In March 2015, Nvidia announced the Tegra X1, the first SoC to have a graphics performance of 1 teraflop. At the announcement event, Nvidia showed off Epic Games' Unreal Engine 4 "Elemental" demo, running on a Tegra X1.

On October 20, 2016, Nvidia announced that the Nintendo Switch hybrid video game console will be powered by Tegra hardware.{{Cite news|url=https://blogs.nvidia.com/blog/2016/10/20/nintendo-switch/|title=NVIDIA Gaming Technology Powers Nintendo Switch {{!}} NVIDIA Blog|date=2016-10-20|newspaper=The Official NVIDIA Blog|language=en-US|access-date=2016-10-20|archive-date=January 26, 2017|archive-url=https://web.archive.org/web/20170126152747/https://blogs.nvidia.com/blog/2016/10/20/nintendo-switch/|url-status=live}} On March 15, 2017, TechInsights revealed the Nintendo Switch is powered by a custom Tegra X1 (model T210), with lower clockspeeds.{{cite web|url=http://www.techinsights.com/about-techinsights/overview/blog/nintendo-switch-teardown/|title=Nintendo Switch Teardown|last=techinsights.com|website=www.techinsights.com|access-date=2017-03-15|archive-date=March 13, 2017|archive-url=https://web.archive.org/web/20170313163420/http://www.techinsights.com/about-techinsights/overview/blog/nintendo-switch-teardown/|url-status=live}}

Models

= Tegra APX =

; Tegra APX 2500

; Tegra APX 2600

  • Enhanced NAND flash
  • Video codecs:{{cite web|url=http://www.nvidia.com/object/product_tegra_apx_us.html|access-date=2011-02-17|title=Nvidia Tegra APX Specifications|archive-date=January 27, 2011|archive-url=https://web.archive.org/web/20110127140206/http://www.nvidia.com/object/product_tegra_apx_us.html|url-status=live}}
  • 720p H.264 Baseline Profile encode or decode
  • 720p VC-1/WMV9 Advanced Profile decode
  • D-1 MPEG-4 Simple Profile encode or decode

= Tegra 6xx =

; Tegra 600

  • Targeted for GPS segment and automotive
  • Processor: ARM11 700 MHz MPCore
  • Memory: low-power DDR (DDR-333, 166 MHz)
  • SXGA, HDMI, USB, stereo jack
  • HD camera 720p

; Tegra 650

  • Targeted for GTX of handheld and notebook
  • Processor: ARM11 800 MHz MPCore
  • Low power DDR (DDR-400, 200 MHz)
  • Less than 1 watt envelope
  • HD image processing for advanced digital still camera and HD camcorder functions
  • Display supports 1080p at 24 frame/s, HDMI v1.3, WSXGA+ LCD and CRT, and NTSC/PAL TV output
  • Direct support for Wi-Fi, disk drives, keyboard, mouse, and other peripherals
  • A complete board support package (BSP) to enable fast time to market for Windows Mobile-based designs

= Tegra 2 =

File:Motorola Xoom - nvidia Tegra 2 T20-H-A2 on main board-0121.jpg

The second generation Tegra SoC has a dual-core ARM Cortex-A9 CPU, an ultra low power (ULP) GeForce GPU,{{cite web | url=http://www.anandtech.com/show/4144/lg-optimus-2x-nvidia-tegra-2-review-the-first-dual-core-smartphone/5 | title=LG Optimus 2X & Nvidia Tegra 2 Review: The First Dual-Core Smartphone | publisher=AnandTech | access-date=2011-08-12 | archive-date=April 26, 2014 | archive-url=https://web.archive.org/web/20140426235118/http://www.anandtech.com/show/4144/lg-optimus-2x-nvidia-tegra-2-review-the-first-dual-core-smartphone/5 | url-status=live }} a 32-bit memory controller with either LPDDR2-600 or DDR2-667 memory, a 32 KB/32 KB L1 cache per core and a shared 1 MB L2 cache.{{cite web | url=http://www.nvidia.com/object/tegra-2.html | title=NVidia Tegra 2 Product Information | publisher=NVidia | access-date=2011-09-05 | archive-date=May 4, 2012 | archive-url=https://web.archive.org/web/20120504014334/http://www.nvidia.com/object/tegra-2.html | url-status=live }} Tegra 2's Cortex A9 implementation does not include ARM's SIMD extension, NEON. There is a version of the Tegra 2 SoC supporting 3D displays; this SoC uses a higher clocked CPU and GPU.

The Tegra 2 video decoder is largely unchanged from the original Tegra and has limited support for HD formats.{{cite web | url=http://www.nvidia.com/object/tegra-superchip.html | title=NVidia Tegra 2 Product Information | publisher=NVidia | access-date=2015-11-01 | archive-date=May 8, 2012 | archive-url=https://web.archive.org/web/20120508013454/http://www.nvidia.com/object/tegra-superchip.html | url-status=live }} The lack of support for high-profile H.264 is particularly troublesome when using online video streaming services.

Common features:

  • CPU cache: L1: 32 KB instruction + 32 KB data, L2: 1 MB
  • 40 nm semiconductor technology

class="wikitable" style="text-align:center;"
rowspan=2 | Model
number

! colspan=3 | CPU

! colspan=3 | GPU

! colspan=4 | Memory

!| Adoption

Processor

! Cores

! Frequency

! Micro-
architecture

! Core config1

! Frequency

! Type

! Amount

! Bus
width

! Band-
width

! Availability

AP20H
(Ventana/
Unknown)

| rowspan="4" | Cortex-A9 || rowspan="4" | 2 || rowspan="2"| 1.0 GHz

| rowspan="4" | VLIW-based
VEC4 units{{cite web|title=The Tegra 4 GPU, NVIDIA Claims Better Performance Than iPad 4|url=http://www.anandtech.com/show/6666/the-tegra-4-gpu-nvidia-claims-better-performance-than-ipad-4|work=AnandTech|last=Shimpi|first=Anand Lal|author-link=Anand Lal Shimpi|access-date=2015-11-05|archive-date=January 21, 2019|archive-url=https://web.archive.org/web/20190121010843/https://www.anandtech.com/show/6666/the-tegra-4-gpu-nvidia-claims-better-performance-than-ipad-4|url-status=live}}|| rowspan="4" | 4:4:4:4{{cite web|url=https://www.techpowerup.com/gpu-specs/tegra-2-gpu.c3225|title = NVIDIA Tegra 2 GPU Specs| date=July 25, 2023 }} || 300 MHz

| rowspan="4"| LPDDR2
300 MHz
DDR2
333 MHz || rowspan="4"| ? || rowspan="4"| 32 bit
single-
channel || rowspan="4"| 2.4 GB/s
2.7 GB/s

| rowspan="2"| Q1 2010

T20
(Harmony/
Ventana)

| 333 MHz

AP25

| rowspan="2"| 1.2 GHz

| rowspan="2"| 400 MHz

| rowspan="2"| Q1 2011

T25

:1 Pixel shaders : Vertex shaders : Texture mapping units : Render output units

== Devices ==

File:Nvidia Tegra 2 T20 Die Shot.jpg

class="wikitable"
Model

! Devices

AP20H

| Motorola Atrix 4G, Motorola Droid X2, Motorola Photon, LG Optimus 2X / LG Optimus Dual P990 / Optimus 2x SU660 (?), Samsung Galaxy R, Samsung Captivate Glide, T-Mobile G2X P999, Acer Iconia Tab A200 and A500, LG Optimus Pad, Motorola Xoom,{{cite web |url=http://developer.motorola.com/products/xoom-mz601/ | title=Motorola Xoom Specifications Table | publisher=Motorola Mobility, Inc | date=February 16, 2011 | access-date=2011-02-16 | url-status=dead | archive-url=https://web.archive.org/web/20110220002957/http://developer.motorola.com/products/xoom-mz601/ | archive-date=February 20, 2011 | df=mdy-all }} Sony Tablet S, Dell Streak Pro,{{cite web |url=https://www.engadget.com/2011/05/19/dell-streak-pro-honeycomb-tablet-pictured-likely-to-be-with-us/|title=Dell Streak Pro Honeycomb tablet pictured, likely to be with us in June |last=Savov |first=Vlad |publisher=Engadget|date=2011-05-19 |access-date=2016-02-05|archive-date=October 24, 2017|archive-url=https://web.archive.org/web/20171024070704/https://www.engadget.com/2011/05/19/dell-streak-pro-honeycomb-tablet-pictured-likely-to-be-with-us/|url-status=live}} Toshiba Thrive{{cite web|title=Toshiba Thrive Review|url=http://www.tabletpcreview.com/default.asp?newsID=2476&review=toshiba+thrive+android+honeycomb+os+tablet|work=TabletPCReview|publisher=TechTarget, Inc.|access-date=November 21, 2013|date=August 3, 2011|archive-date=November 6, 2013|archive-url=https://web.archive.org/web/20131106161825/http://www.tabletpcreview.com/default.asp?newsID=2476&review=toshiba+thrive+android+honeycomb+os+tablet|url-status=live}} tablet, T-Mobile G-Slate

AP25

| Fusion Garage Grid 10{{Citation needed|date=September 2011}}

T20

| Avionic Design Tamonten Processor Board,{{cite web | url=http://www.avionic-design.de/uploads/pdf/tamonten_tegra_processor-COM_EN.pdf | title=Avionic Design Tegra 2 (T290) Tamonten Processor Module — Product Brief | publisher=Avionic Design | access-date=May 25, 2012 | archive-url=https://web.archive.org/web/20140521122311/http://www.avionic-design.de/uploads/pdf/tamonten_tegra_processor-COM_EN.pdf | archive-date=May 21, 2014 | url-status=dead }} Notion Ink Adam tablet, Olivetti OliPad 100, ViewSonic G Tablet, ASUS Eee Pad Transformer, Samsung Galaxy Tab 10.1, Toshiba AC100, CompuLab Trim-Slice nettop, Velocity Micro Cruz Tablet L510, Acer Iconia Tab A100

{{unk}}

| Tesla Motors Model S 2012~2017 and Model X 2015~2017 instrument cluster (IC)[https://arstechnica.com/gadgets/2014/05/nvidia-inside-hands-on-with-audi-lamborghini-and-tesla/ Nvidia inside: Hands on with Audi, Lamborghini, and Tesla] {{Webarchive|url=https://web.archive.org/web/20180315200138/https://arstechnica.com/gadgets/2014/05/nvidia-inside-hands-on-with-audi-lamborghini-and-tesla/ |date=March 15, 2018 }} by Megan Geuss in May 2014[https://teslatap.com/?s=processors Processors Analysis and Count] {{Webarchive|url=https://web.archive.org/web/20180315201613/https://teslatap.com/?s=processors |date=March 15, 2018 }} in May 2013

= Tegra 3 =

File:Nexus 7 (2012) - board - Nvidia T30L-P-A3-4854.jpg

NVIDIA's Tegra 3 (codenamed "Kal-El"){{cite web |url=http://www.androidcentral.com/nvidia-announces-tegra-3-pc-class-performance-comes-android-tablets |title=Nvidia announces the Tegra 3 – Kal-El brings PC class performance to Android |publisher=Android Central |date=2011-11-09 |access-date=2016-07-10 |archive-date=July 16, 2012 |archive-url=https://web.archive.org/web/20120716043745/http://www.androidcentral.com/nvidia-announces-tegra-3-pc-class-performance-comes-android-tablets |url-status=live }} is functionally a SoC with a quad-core ARM Cortex-A9 MPCore CPU, but includes a fifth "companion" core in what Nvidia refers to as a "variable SMP architecture".{{cite web |url=http://www.nvidia.com/object/tegra-3-processor.html |title=Tegra 3 Multi-Core Processors |publisher=NVIDIA |access-date=2016-07-15 |archive-date=April 28, 2012 |archive-url=https://web.archive.org/web/20120428132736/http://www.nvidia.com/object/tegra-3-processor.html |url-status=live }} While all cores are Cortex-A9s, the companion core is manufactured with a low-power silicon process. This core operates transparently to applications and is used to reduce power consumption when processing load is minimal. The main quad-core portion of the CPU powers off in these situations.

Tegra 3 is the first Tegra release to support ARM's SIMD extension, NEON.

The GPU in Tegra 3 is an evolution of the Tegra 2 GPU, with 4 additional pixel shader units and higher clock frequency. It can also output video up to 2560×1600 resolution and supports 1080p MPEG-4 AVC/h.264 40 Mbit/s High-Profile, VC1-AP, and simpler forms of MPEG-4 such as DivX and Xvid.{{cite web |url=http://hexus.net/mobile/news/tablets/32531-asus-transformer-prime-introduced-examined/ |title=ASUS Transformer Prime introduced and examined |date=November 9, 2011 |publisher=HEXUS.net |access-date=2011-11-11 |archive-date=November 11, 2011 |archive-url=https://web.archive.org/web/20111111210826/http://hexus.net/mobile/news/tablets/32531-asus-transformer-prime-introduced-examined/ |url-status=live }}

The Tegra 3 was released on November 9, 2011.{{cite web|url=http://pressroom.nvidia.com/easyir/customrel.do?easyirid=A0D622CE9F579F09&version=live&prid=819304&releasejsp=release_157&xhtml=true|archive-url=https://web.archive.org/web/20120111081853/http://pressroom.nvidia.com/easyir/customrel.do?easyirid=A0D622CE9F579F09&version=live&prid=819304&releasejsp=release_157&xhtml=true|url-status=dead|archive-date=January 11, 2012|title=NVIDIA Quad-Core Tegra 3 Chip Sets New Standards of Mobile Computing Performance, Energy Efficiency – NVIDIA Newsroom|date=January 11, 2012}}

Common features:

  • CPU cache: L1: 32 KB instruction + 32 KB data, L2: 1 MB
  • 40 nm LPG semiconductor technology by TSMC

class="wikitable" style="text-align:center;"
rowspan=2 | Model
number

! colspan=3 | CPU

! colspan=3 | GPU

! colspan=4 | Memory

!| Adoption

Processor

! Cores

! Frequency
(multi-/single-
core mode)

! Micro-
architecture

! Core
config1

! Frequency

! Type

! Amount

! Bus
width

! Band-
width

! Availability

T30L

| rowspan="4" | Cortex-A9 || rowspan="4" | 4+1 || 1.2 GHz /
up to
1.3 GHz

| rowspan="4" | VLIW-based
VEC4 units
|| rowspan="4" | 8:4:8:8
{{cite web|url=https://www.techpowerup.com/gpu-specs/tegra-3-gpu.c3226|title = NVIDIA Tegra 3 GPU Specs| date=July 25, 2023 }} || 416 MHz

| DDR3-1333 || ? || rowspan="4" | 32 bit
single-
channel || 5.3 GB/s {{cite web |url=http://www.anandtech.com/show/6036/asus-transformer-pad-infinity-tf700t-review/3 |title=A Faster Tegra 3, More Memory Bandwidth – ASUS Transformer Pad Infinity (TF700T) Review |website=Anandtech.com |access-date=2016-07-10 |archive-date=June 27, 2012 |archive-url=https://web.archive.org/web/20120627224447/http://www.anandtech.com/show/6036/asus-transformer-pad-infinity-tf700t-review/3 |url-status=live }}

| Q1 2012

T30

| rowspan="2" | 1.4 GHz /
up to
1.5 GHz

| rowspan="3" | 520 MHz

| rowspan="2" | LPDDR2-1066
DDR3-L-1500 || rowspan="2" | ? || rowspan="2" | 4.3 GB/s
6.0 GB/s {{cite web |url=http://www.nvidia.com/object/tegra-3-processor.html |title=Tegra 3 Multi-Core Processors |publisher=NVIDIA |access-date=2016-07-10 |archive-date=April 28, 2012 |archive-url=https://web.archive.org/web/20120428132736/http://www.nvidia.com/object/tegra-3-processor.html |url-status=live }}

| rowspan="2" | Q4 2011

AP33

T33

| 1.6 GHz /
up to
1.7 GHz

| DDR3-1600 || ? || 6.4 GB/s

| Q2 2012

:1 Pixel shaders : Vertex shaders : Texture mapping units : Render output units

== Devices ==

File:OUYA-Console-set-h.jpg uses a Tegra 3 T33-P-A3.]]

class="wikitable"
Model

! Devices

AP33

| LG Optimus 4X HD, HTC One X, XOLO Play T1000,{{cite web|url=http://www.xolo.in/xolo-play-features-specs|archive-url=https://web.archive.org/web/20130721034837/http://www.xolo.in/xolo-play-features-specs|url-status=dead|archive-date=July 21, 2013|title=XOLO – The Next Level|date=July 21, 2013}} Coolpad 8735

T30

| Asus Eee Pad Transformer Prime (TF201),{{cite web |url=http://asia.cnet.com/product/asus-eee-pad-transformer-prime-nvidia-tegra-3-processor-10-1-inch-display-45735891.htm |title=Asus Eee Pad Transformer Prime (Nvidia Tegra 3 Processor; 10.1-inch display) Review |date=December 30, 2011 |url-status=dead |archive-url=https://web.archive.org/web/20130402093316/http://asia.cnet.com/product/asus-eee-pad-transformer-prime-nvidia-tegra-3-processor-10-1-inch-display-45735891.htm |archive-date=April 2, 2013 |df=mdy-all }} IdeaTab K2 / LePad K2,{{cite web |url=http://www.glbenchmark.com/phonedetails.jsp?benchmark=glpro21&D=Lenovo+LePad+K2&testgroup=system |title=GFXBench – unified graphics benchmark based on DXBenchmark (DirectX) and GLBenchmark (OpenGL ES) |website=Glbenchmark.com |access-date=2016-07-15 |archive-date=January 22, 2012 |archive-url=https://web.archive.org/web/20120122105028/http://www.glbenchmark.com/phonedetails.jsp?benchmark=glpro21&D=Lenovo+LePad+K2&testgroup=system |url-status=live }} Acer Iconia Tab A510, Fuhu Inc. nabi 2 Tablet,{{cite web |last=Summerson |first=Cameron |url=http://www.androidpolice.com/2012/06/19/fuhu-nabi-2-review-a-quad-core-android-4-0-tablet-designed-just-for-your-kids-and-its-surprisingly-awesome/ |title=Fuhu Nabi 2 Review: A Quad-Core Android 4.0 Tablet Designed Just For Your Kids – And It's Surprisingly Awesome |website=Androidpolice.com |date=June 19, 2012 |access-date=2016-07-15 |archive-date=June 22, 2012 |archive-url=https://web.archive.org/web/20120622042940/http://www.androidpolice.com/2012/06/19/fuhu-nabi-2-review-a-quad-core-android-4-0-tablet-designed-just-for-your-kids-and-its-surprisingly-awesome/ |url-status=live }} Microsoft Surface RT,{{cite web |url=http://www.microsoft.com/en-us/news/Press/2012/Oct12/10-16announcementPR.aspx |title=Microsoft Announces New Surface Details {{pipe}} News Center |website=Microsoft.com |date=2012-10-16 |access-date=2016-07-15 |archive-date=July 12, 2014 |archive-url=https://web.archive.org/web/20140712031842/http://www.microsoft.com/en-us/news/press/2012/oct12/10-16announcementpr.aspx |url-status=live }} Lenovo IdeaPad Yoga 11,{{cite web |url=https://techcrunch.com/2012/10/09/lenovo-introduces-the-ideapad-yoga-11-and-13-the-first-tablet-laptop-ultrabook-hybrid/ |title=Lenovo Introduces The IdeaPad Yoga 11 and 13, The First Tablet & Laptop Ultrabook Hybrid |publisher=TechCrunch |date=2012-10-09 |access-date=2016-07-15 |archive-date=December 22, 2017 |archive-url=https://web.archive.org/web/20171222051842/https://techcrunch.com/2012/10/09/lenovo-introduces-the-ideapad-yoga-11-and-13-the-first-tablet-laptop-ultrabook-hybrid/ |url-status=live }}{{cite web |last=Jackson |first=Jerry |url=http://www.notebookreview.com/default.asp?newsID=6600&news=lenovo+ideapad+yoga |title=Lenovo Launches IdeaPad Yoga 11, Yoga 13 |website=Notebookreview.com |date=2012-10-09 |access-date=2016-07-15 |archive-date=October 18, 2012 |archive-url=https://web.archive.org/web/20121018184454/http://www.notebookreview.com/default.asp?newsID=6600&news=lenovo+ideapad+yoga |url-status=live }}

T30I

|Tesla Model S 2012~2017 and Model X 2015~2017 media control unit (MCU)[https://blog.lookout.com/hacking-a-tesla Hacking a Tesla Model S: What we found and what we learned] {{Webarchive|url=https://web.archive.org/web/20171220084432/https://blog.lookout.com/hacking-a-tesla|date=December 20, 2017}} by Kevin Mahaffey on August 7, 2015

T30L

| Asus Transformer Pad TF300T, Microsoft Surface, Nexus 7 (2012),{{cite web|title=Nexus 7 tablet hands-on|date=June 27, 2012 |url=https://www.engadget.com/2012/06/27/nexus-7-tablet-hands-on/|publisher=Engadget|access-date=June 27, 2012|archive-date=June 29, 2012|archive-url=https://web.archive.org/web/20120629052732/http://www.engadget.com/2012/06/27/nexus-7-tablet-hands-on/|url-status=live}} Sony Xperia Tablet S, Acer Iconia Tab A210, Toshiba AT300 (Excite 10),{{cite web|title=Toshiba Excite 10 Benchmark Test| date=May 29, 2012 |url=https://www.youtube.com/watch?v=8K8AseX6S1g|publisher=YouTube|access-date=November 25, 2012|archive-date=July 27, 2013|archive-url=https://web.archive.org/web/20130727052601/http://www.youtube.com/watch?v=8K8AseX6S1g|url-status=live}}{{unreliable source?|date=February 2021}} BLU Quattro 4.5,{{cite web|url=http://bluproducts.com/pro-detail/quattro45|archive-url=https://web.archive.org/web/20130420055230/http://bluproducts.com/pro-detail/quattro45|url-status=dead|archive-date=April 20, 2013|title=Blu Products: Quattro45|date=April 20, 2013}} Coolpad 9070

T33

| Asus Transformer Pad Infinity (TF700T), Fujitsu ARROWS X F-02E, HTC One X+, Ouya (T33-P-A3)

= Tegra 4 =

The Tegra 4 (codenamed "Wayne") was announced on January 6, 2013, and is a SoC with a quad-core CPU, but includes a fifth low-power Cortex A15 companion core which is invisible to the OS and performs background tasks to save power. This power-saving configuration is referred to as "variable SMP architecture" and operates like the similar configuration in Tegra 3.{{cite web |url=http://www.nvidia.com/object/tegra-4-processor.html |title=Tegra 4 Processors |publisher=NVIDIA |access-date=2016-07-15 |archive-date=January 27, 2013 |archive-url=https://web.archive.org/web/20130127013020/http://www.nvidia.com/object/tegra-4-processor.html |url-status=live }}

The GeForce GPU in Tegra 4 is again an evolution of its predecessors. However, numerous feature additions and efficiency improvements were implemented. The number of processing resources was dramatically increased, and clock rate increased as well. In 3D tests, the Tegra 4 GPU is typically several times faster than that of Tegra 3.{{cite web|last=Parrish |first=Kevin |url=http://www.tomshardware.com/reviews/nvidia-tegra-note-7-evga-tablet-review,3668-9.html |title=Results: GPU Benchmarks – EVGA Tegra Note 7 Review: Nvidia's Tegra 4 For $200 |website=Tomshardware.com |date=November 12, 2013 |access-date=2016-07-15}} Additionally, the Tegra 4 video processor has full support for hardware decoding and encoding of WebM video (up to 1080p 60 Mbit/s @ 60fps).{{cite web |url=http://www.nvidia.com/docs/IO/116757/Tegra_4_GPU_Whitepaper_FINALv2.pdf |title=NVIDIA Tegra Multi-processor Architecture |access-date=2013-07-10 |archive-date=March 20, 2013 |archive-url=https://web.archive.org/web/20130320024356/http://www.nvidia.com/docs/IO/116757/Tegra_4_GPU_Whitepaper_FINALv2.pdf |url-status=live }}

Along with Tegra 4, Nvidia also introduced i500, an optional software modem based on Nvidia's acquisition of Icera, which can be reprogrammed to support new network standards. It supports category 3 (100 Mbit/s) LTE but will later be updated to Category 4 (150 Mbit/s).

Common features:

  • CPU cache: L1: 32 KB instruction + 32 KB data, L2: 2 MB
  • 28 nm HPL semiconductor technology

class="wikitable" style="text-align:center;"
rowspan=2 | Model
number

! CPU

! colspan=3 | GPU

! colspan=4 | Memory

!| Adoption

Processor
(Cores/Freq)

! Micro-
architecture

! Core config1

! Frequency

! Type

! Amount

! Bus
width

! Band-
width

! Availability

T114{{cite news |url=https://www.phoronix.com/scan.php?page=news_item&px=MTI1OTA |title=NVIDIA Publishes Their Next-Gen Tegra 4 Code |publisher=phoronix.com |first=Michael |last=Larabel |date=December 20, 2012 |access-date=August 2, 2013 |archive-date=May 14, 2013 |archive-url=https://web.archive.org/web/20130514083733/http://www.phoronix.com/scan.php?page=news_item&px=MTI1OTA |url-status=live }}

| 4+1 x 1.9 GHz
Cortex-A15

| VLIW-based
VEC4 units || 72
(48:24:4)

| 672 MHz{{cite web |last=Angelini |first=Chris |title=Nvidia's Tegra 4 GPU: Doubling Down On Efficiency |url=http://www.tomshardware.com/reviews/tegra-4-tegra-4i-gpu-architecture,3445.html |work=Tom's Hardware |date=February 24, 2013 |access-date=September 2, 2013 }}

| DDR3L or
LPDDR3 || ? || 32-bit
dual-
channel || up to
14.9 GB/s
(1866 MT/s
data rate){{cite web |url=http://www.nvidia.com/object/tegra-4-processor.html |title=Tegra 4 Processors |publisher=NVIDIA |access-date=2013-07-10 |archive-date=January 27, 2013 |archive-url=https://web.archive.org/web/20130127013020/http://www.nvidia.com/object/tegra-4-processor.html |url-status=live }}{{cite web |url=http://www.anandtech.com/show/6787/nvidia-tegra-4-architecture-deep-dive-plus-tegra-4i-phoenix-hands-on/5 |title=NVIDIA Tegra 4 Architecture Deep Dive, Plus Tegra 4i, Icera i500 & Phoenix Hands On |publisher=AnandTech |access-date=2013-07-10 |archive-date=February 27, 2013 |archive-url=https://web.archive.org/web/20130227020351/http://www.anandtech.com/show/6787/nvidia-tegra-4-architecture-deep-dive-plus-tegra-4i-phoenix-hands-on/5 |url-status=live }}

| Q2 2013{{cite web |url=http://www.anandtech.com/show/6746/tegra-4-shipment-date-still-q2-2013 |title=Tegra 4 Shipment Date: Still Q2 2013 |publisher=AnandTech |access-date=2013-07-10 |archive-date=February 17, 2013 |archive-url=https://web.archive.org/web/20130217051845/http://www.anandtech.com/show/6746/tegra-4-shipment-date-still-q2-2013 |url-status=live }}

:1 Pixel shaders : Vertex shaders : Pixel pipelines (pairs 1x TMU and 1x ROP)

== Devices ==

class="wikitable"
Model

! Devices

T114

| Nvidia Shield Portable, Tegra Note 7, Microsoft Surface 2, HP Slate 7 Extreme,{{cite web |url=http://h20564.www2.hp.com/hpsc/doc/public/display?docId=c04047110 |title=HP Slate 7 Extreme 4400CA Tablet Product Specifications |publisher=.hp.com |access-date=2016-09-22 |archive-date=September 23, 2016 |archive-url=https://web.archive.org/web/20160923100113/http://h20564.www2.hp.com/hpsc/doc/public/display?docId=c04047110 |url-status=live }} HP Slate 7 Beats Special Edition,{{cite web |url=http://support.hp.com/us-en/document/c04266509 |title=HP Slate7 Beats Special Edition 4501 Tablet Product Specifications |publisher=.hp.com |access-date=2016-09-22 |archive-date=September 23, 2016 |archive-url=https://web.archive.org/web/20160923114715/http://support.hp.com/us-en/document/c04266509 |url-status=live }} HP Slate 8 Pro,{{cite web |url=http://support.hp.com/us-en/document/c04005245 |title=HP Slate 8 Pro 7600us Tablet Product Specifications |publisher=hp.com |access-date=2016-09-22 |archive-date=September 23, 2016 |archive-url=https://web.archive.org/web/20160923190608/http://support.hp.com/us-en/document/c04005245 |url-status=live }} HP SlateBook x2,{{cite web |url=http://www8.hp.com/us/en/ads/x2/slatebook-x2.html |title=HP SlateBook x2 Overview – Android Tablet Notebook {{pipe}} HP Official Site |publisher=.hp.com |access-date=2013-07-10 |archive-date=July 12, 2013 |archive-url=https://web.archive.org/web/20130712192154/http://www8.hp.com/us/en/ads/x2/slatebook-x2.html |url-status=live }} HP SlateBook 14,{{cite web |url=http://h20564.www2.hp.com/hpsc/doc/public/display?docId=emr_na-c04336066 |title=HP SlateBook 14-p010nr Product Specifications |publisher=hp.com |access-date=2016-09-22 |archive-date=September 23, 2016 |archive-url=https://web.archive.org/web/20160923170749/http://h20564.www2.hp.com/hpsc/doc/public/display?docId=emr_na-c04336066 |url-status=live }} HP Slate 21,{{cite web |url=http://support.hp.com/in-en/document/c03897044 |title=HP Slate 21-s100 All-in-One Desktop PC – Product Specifications |publisher=hp.com |access-date=2016-09-22 |archive-date=September 23, 2016 |archive-url=https://web.archive.org/web/20160923101519/http://support.hp.com/in-en/document/c03897044 |url-status=live }} ZTE N988S, nabi Big Tab, Nuvola NP-1, Project Mojo, Asus Transformer Pad TF701T, Toshiba AT10-LE-A (Excite Pro), Vizio 10" tablet, Wexler.Terra 7, Wexler.Terra 10, Acer TA272HUL AIO, Xiaomi Mi 3 (TD-LTE version),{{cite web|url=http://cintiqcompanion.wacom.com/CintiqCompanionHybrid/en/|archive-url=https://web.archive.org/web/20130823043119/http://cintiqcompanion.wacom.com/CintiqCompanionHybrid/en/|url-status=dead|archive-date=August 23, 2013|title=Cintiq Companion Hybrid – Wacom|date=August 23, 2013}} Coolpad 8970L ({{lang|zh|大观}} 4),{{cite web |url=http://shop.coolpad.cn/goods/2047.htm |title=用户太多，系统繁忙 |website=Shop.coolpad.cn |access-date=2016-07-15 |archive-date=December 31, 2013 |archive-url=https://web.archive.org/web/20131231002206/http://shop.coolpad.cn/goods/2047.htm |url-status=live }} Audi Tablet,{{cite web |last=Shapiro |first=Danny |url=http://blogs.nvidia.com/blog/2015/03/11/audi-geneva/ |title=Audi Offers Taste of Tegra-Powered Future at Geneva Motor Show {{pipe}} NVIDIA Blog |website=Blogs.nvidia.com |access-date=2016-07-10 |archive-date=April 2, 2015 |archive-url=https://web.archive.org/web/20150402090541/http://blogs.nvidia.com/blog/2015/03/11/audi-geneva/ |url-status=live }} Le Pan TC1020 10.1",{{cite web |url=http://lepantab.com/v2/?portfolio=tc1020 |title=Le Pan – TC1020 |publisher=Lepantab.com |access-date=2016-09-22 |archive-date=September 23, 2016 |archive-url=https://web.archive.org/web/20160923104353/http://lepantab.com/v2/?portfolio=tc1020 |url-status=live }} Matrimax iPLAY 7,{{cite web |url=http://www.open-consoles-news.com/2015/06/test-matrimax-iplay.html |title=[Test] Matrimax iPlay |publisher=Open-consoles-news.com |access-date=2016-09-22 |archive-date=September 23, 2016 |archive-url=https://web.archive.org/web/20160923115712/http://www.open-consoles-news.com/2015/06/test-matrimax-iplay.html |url-status=live }} Kobo Arc 10HD{{cite web |url=https://www.cnet.com/products/kobo-arc-10-hd/specs/ |title=Kobo Arc 10 HD Specs |publisher=C-Net |access-date=2017-07-08 |archive-date=March 15, 2018 |archive-url=https://web.archive.org/web/20180315200014/https://www.cnet.com/products/kobo-arc-10-hd/specs/ |url-status=live }}

== Tegra 4i ==

The Tegra 4i (codenamed "Grey") was announced on February 19, 2013. With hardware support for the same audio and video formats, but using Cortex-A9 cores instead of Cortex-A15, the Tegra 4i is a low-power variant of the Tegra 4 and is designed for phones and tablets. Unlike its Tegra 4 counterpart, the Tegra 4i also integrates the Icera i500 LTE/HSPA+ baseband processor onto the same die.

Common features:

  • 28 nm HPM semiconductor technology
  • CPU cache: L1: 32 KB instruction + 32 KB data, L2: 1 MB

class="wikitable" style="text-align:center;"
rowspan=2 | Model
number

! CPU

! colspan=3 | GPU

! colspan=4 | Memory

!| Adoption

Processor
(Cores/Freq)

! Micro-
architecture

! Core config1

! Frequency

! Type

! Amount

! Bus
width

! Band-
width

! Availability

T148?{{cite web |last=Cunningham |first=Andrew |url=https://arstechnica.com/gadgets/2013/02/project-grey-becomes-tegra-4i-nvidias-latest-play-for-smartphones/ |title=Project Grey becomes Tegra 4i, Nvidia's latest play for smartphones |publisher=Ars Technica |date=February 19, 2013 |access-date=2013-07-10 |archive-date=December 2, 2017 |archive-url=https://web.archive.org/web/20171202063403/https://arstechnica.com/gadgets/2013/02/project-grey-becomes-tegra-4i-nvidias-latest-play-for-smartphones/ |url-status=live }}

| 4+1 x 2.0 GHz
Cortex-A9 "R4"

| VLIW-based
VEC4 units || 60{{cite web |last=Walrath |first=Josh |title=NVIDIA Details Tegra 4 and Tegra 4i Graphics|url=http://www.pcper.com/news/Graphics-Cards/NVIDIA-Details-Tegra-4-and-Tegra-4i-Graphics |work=PC Perspective |date=February 26, 2013 |access-date=September 2, 2013 |archive-date=December 23, 2014 |archive-url=https://web.archive.org/web/20141223084431/http://www.pcper.com/news/Graphics-Cards/NVIDIA-Details-Tegra-4-and-Tegra-4i-Graphics |url-status=live }}
(48:12:2)

| 660 MHz

| LPDDR3 || || 32-bit
single-
channel || 6.4–7.5 GB/s
(800–933 MHz)

| Q1 2014

:1 Pixel shaders : Vertex shaders : Pixel pipelines (pairs 1x TMU and 1x ROP)

=== Devices ===

class="wikitable"
Model

! Devices

T148?

| Blackphone, LG G2 mini LTE, Wiko Highway 4G,{{cite web|url=http://fr.wikomobile.com/m250-HIGHWAY-4G|archive-url=https://web.archive.org/web/20140917094609/http://fr.wikomobile.com/m250-HIGHWAY-4G|url-status=dead|archive-date=September 17, 2014|title=Wiko Mobile – HIGHWAY 4G|date=September 17, 2014}} Explay 4Game,{{cite web |url=http://www.nvidia.ru/object/explay-4game-smartphone-ru.html |title=Explay 4Game {{pipe}} Четырехъядерный смартфон на базе Tegra 4i {{pipe}} NVIDIA |website=Blogs.nvidia.com |access-date=2016-07-10 |archive-date=December 5, 2014 |archive-url=https://web.archive.org/web/20141205231828/http://www.nvidia.ru/object/explay-4game-smartphone-ru.html |url-status=live }} Wiko Wax{{cite web |last=Han |first=Mike |url=http://blogs.nvidia.com/blog/2014/02/24/lte-europe/ |title=NVIDIA LTE Modem Makes Landfall in Europe, with Launch of Wiko Tegra 4i LTE Smartphone {{pipe}} The Official NVIDIA Blog |website=Blogs.nvidia.com |date=2014-02-24 |access-date=2016-07-10 |archive-date=February 28, 2014 |archive-url=https://web.archive.org/web/20140228104125/https://blogs.nvidia.com/blog/2014/02/24/lte-europe/ |url-status=live }}{{cite web |url=http://www.devicespecifications.com/en/model/6e572cd4 |title=Wiko WAX |publisher=DeviceSpecifications |access-date=2014-05-21 |archive-date=May 21, 2014 |archive-url=https://web.archive.org/web/20140521202025/http://www.devicespecifications.com/en/model/6e572cd4 |url-status=live }} QMobile Noir LT-250{{cite web |url=http://www.hitmobile.pk/phones/qmobile/qmobile-noir-lt-250/ |title=QMobile Noir LT-250 |publisher=DeviceSpecifications |access-date=2014-02-10 |archive-date=February 10, 2015 |archive-url=https://web.archive.org/web/20150210170645/http://www.hitmobile.pk/phones/qmobile/qmobile-noir-lt-250/ |url-status=live }}

= {{Anchor|Jetson TK1}}Tegra K1 =

Nvidia's Tegra K1 (codenamed "Logan") features ARM Cortex-A15 cores in a 4+1 configuration similar to Tegra 4, or Nvidia's 64-bit Project Denver dual-core processor as well as a Kepler graphics processing unit with support for Direct3D 12, OpenGL ES 3.1, CUDA 6.5, OpenGL 4.4/OpenGL 4.5, and Vulkan.{{cite web |last=Park |first=Will |url=http://blogs.nvidia.com/blog/2014/05/15/nvidias-tegra-k1-powers-xiaomis-first-tablet/ |title=NVIDIA's Tegra K1 Powers Xiaomi's First Tablet {{pipe}} The Official NVIDIA Blog |website=Blogs.nvidia.com |date=2014-05-15 |access-date=2016-07-15 |archive-date=July 12, 2014 |archive-url=https://web.archive.org/web/20140712072810/http://blogs.nvidia.com/blog/2014/05/15/nvidias-tegra-k1-powers-xiaomis-first-tablet/ |url-status=live }}{{cite web|url=http://www.androidcentral.com/nvidia-shield-tablet-k1-android-601-update-brings-vulkan-graphics-api-support|title=NVIDIA Shield Tablet K1 gets Vulkan support with Android 6.0.1 update|access-date=May 3, 2016|archive-date=May 9, 2016|archive-url=https://web.archive.org/web/20160509225056/http://www.androidcentral.com/nvidia-shield-tablet-k1-android-601-update-brings-vulkan-graphics-api-support|url-status=live}} Nvidia claims that it outperforms both the Xbox 360 and the PS3, whilst consuming significantly less power.{{cite web|url=https://www.bbc.co.uk/news/technology-25618498|title=CES 2014: Nvidia Tegra K1 offers leap in graphics power|date=January 6, 2014|publisher=BBC|last=Kelion|first=Leo|access-date=January 11, 2014|archive-date=January 11, 2014|archive-url=https://web.archive.org/web/20140111062616/http://www.bbc.co.uk/news/technology-25618498|url-status=live}}

Support Adaptive Scalable Texture Compression.{{cite web|url=http://on-demand.gputechconf.com/siggraph/2015/presentation/SIG1501-Piers-Daniell.pdf|title=Vulkan API|access-date=December 11, 2015|archive-date=December 22, 2015|archive-url=https://web.archive.org/web/20151222113116/http://on-demand.gputechconf.com/siggraph/2015/presentation/SIG1501-Piers-Daniell.pdf|url-status=live}}

In late April 2014, Nvidia shipped the "Jetson TK1" development board containing a Tegra K1 SoC and running Ubuntu Linux.{{cite web | url=https://www.phoronix.com/scan.php?page=news_item&px=MTY3NjA | title=NVIDIA's Tegra TK1 Jetson Board Is Now Shipping | date=29 April 2014 | last=Larabel | first=Michael | publisher=Phoronix | access-date=September 14, 2016 | archive-date=April 25, 2016 | archive-url=https://web.archive.org/web/20160425033757/http://www.phoronix.com/scan.php?page=news_item&px=MTY3NjA | url-status=live }}

  • Processor:
  • 32-bit variant quad-core ARM Cortex-A15 MPCore R3 + low power companion core
  • or 64-bit variant with dual-core Project Denver{{cite news|url=http://www.extremetech.com/computing/174023-tegra-k1-64-bit-denver-core-analysis-are-nvidias-x86-efforts-hidden-within|title=Tegra K1 64-bit Denver core analysis: Are Nvidia's x86 efforts hidden within?|last=Anthony|first=Sebastian|date=January 6, 2014|publisher=ExtremeTech|access-date=January 7, 2014|archive-date=January 7, 2014|archive-url=https://web.archive.org/web/20140107012023/http://www.extremetech.com/computing/174023-tegra-k1-64-bit-denver-core-analysis-are-nvidias-x86-efforts-hidden-within|url-status=live}} (variant once codenamed "Stark"[https://www.engadget.com/2011/10/21/nvidia-ceo-confirms-tegra-roadmap-building-all-now-kal-el-way/ NVIDIA CEO confirms Tegra roadmap, building all now: Kal-El, Wayne, Logan, Stark] {{Webarchive|url=https://web.archive.org/web/20170316205624/https://www.engadget.com/2011/10/21/nvidia-ceo-confirms-tegra-roadmap-building-all-now-kal-el-way/ |date=March 16, 2017 }}, October 21, 2011: Finally, he confirmed that the inner workings we've heard about in Project Denver will first be present in the Tegra line with the introduction of Stark(...))
  • GPU consisting of 192 ALUs using Kepler technology
  • 28 nm HPM process
  • Released in Q2 2014
  • Power consumption: 8 watts

class="wikitable" style="text-align:center;"
rowspan=2 | Model number

! CPU

! colspan=4 | GPU

! colspan=4 | Memory

!| Adoption

Processor
(Cores/Freq)

! Micro-
architecture

! Core
config1

! Frequency

! GFLOPS
(FP32)

! Type

! Amount

! Bus
width

! Band-
width

! Availability

T124{{cite web |url=http://www.nvidia.com/object/tegra-k1-processor.html |title=Tegra K1 Next-Gen Mobile Processor {{pipe}} NVIDIA Tegra |publisher=NVIDIA |access-date=2016-07-15 |archive-date=January 9, 2014 |archive-url=https://web.archive.org/web/20140109162108/http://www.nvidia.com/object/tegra-k1-processor.html |url-status=live }}{{cite web |title=NVIDIA Tegra K1 Preview & Architecture Analysis |url=http://www.anandtech.com/show/7622/nvidia-tegra-k1/3 |work=AnandTech |date=January 6, 2014 |access-date=2014-05-02 |last1=Klug |first1=Brian |last2=Shimpi |first2=Anand Lal |author-link2=Anand Lal Shimpi |page=3 |archive-date=April 19, 2014 |archive-url=https://web.archive.org/web/20140419205408/http://www.anandtech.com/show/7622/nvidia-tegra-k1/3 |url-status=live }}

| 4+1 x 2.3 GHz
Cortex-A15 R3
(32-bit)

| rowspan=2 | GK20A
(Kepler)

| rowspan=2 | 192:8:4

| rowspan=2 | 756–951 MHz

| rowspan=2 | 290–365

| rowspan=2 | DDR3L,
LPDDR3

| max 8 GB
(with 40-bit
address
extension2)

| 64 bit

| 17 GB/s

| Q2 2014

T132{{cite web |last=Stam |first=Nick |url=http://blogs.nvidia.com/blog/2014/08/11/tegra-k1-denver-64-bit-for-android/ |title=Mile High Milestone: Tegra K1 "Denver"? Will Be First 64-bit ARM Processor for Android {{pipe}} The Official NVIDIA Blog |website=Blogs.nvidia.com |access-date=2016-07-15 |archive-date=August 12, 2014 |archive-url=https://web.archive.org/web/20140812090907/http://blogs.nvidia.com/blog/2014/08/11/tegra-k1-denver-64-bit-for-android/ |url-status=live }}{{cite web|url=https://www.anandtech.com/show/8811/nvidia-tegra-x1-preview/2|title=NVIDIA Tegra X1 Preview & Architecture Analysis|last=Ho|first=Joshua (5 January 2015)|website=Anandtech|access-date=3 December 2018|archive-date=December 4, 2018|archive-url=https://web.archive.org/web/20181204102016/https://www.anandtech.com/show/8811/nvidia-tegra-x1-preview/2|url-status=live}}

| 2x 2.5GHz
Denver
(64-bit)

| max 8 GB

| ?

| ?

| Q3 2014

:1 Unified Shaders : Texture mapping units : Render output units

:2 ARM Large Physical Page Extension (LPAE) supports 1 TiB (240 bytes).

:The 8 GiB limitation is part-specific.

== Devices ==

class="wikitable"
Model

! Devices

T124

| Jetson TK1 development board,{{cite web|url=https://developer.nvidia.com/jetson-tk1|title=Jetson TK1 development board|access-date=May 1, 2014|archive-date=September 5, 2015|archive-url=https://web.archive.org/web/20150905181415/https://developer.nvidia.com/jetson-tk1|url-status=dead}} Nvidia Shield Tablet,{{cite web |url=http://www.geforce.com/whats-new/articles/shield-tablet-the-ultimate-tablet-for-gamers |title=SHIELD Tablet, The Ultimate Tablet For Gamers |publisher=GeForce |date=2014-07-22 |access-date=2016-07-15 |archive-date=July 25, 2014 |archive-url=https://web.archive.org/web/20140725005312/http://www.geforce.com/whats-new/articles/shield-tablet-the-ultimate-tablet-for-gamers |url-status=live }} Acer Chromebook 13,{{cite web |url=http://www.anandtech.com/show/8354/tegra-k1-lands-in-acers-newest-chromebook |title=Tegra K1 Lands in Acer's Newest Chromebook |publisher=Anandtech |date=2014-08-11 |access-date=August 11, 2014 |archive-date=July 20, 2018 |archive-url=https://web.archive.org/web/20180720194929/https://www.anandtech.com/show/8354/tegra-k1-lands-in-acers-newest-chromebook |url-status=live }} HP Chromebook 14 G3,{{cite web |url=https://support.hp.com/us-en/product/hp-chromebook-14-g3/7096564/document/c04481826 |title=HP Chromebook 14 G3 – Specifications |publisher=HP |date=2018-08-30 |access-date=August 30, 2018 |archive-date=August 30, 2018 |archive-url=https://web.archive.org/web/20180830210153/https://support.hp.com/us-en/product/hp-chromebook-14-g3/7096564/document/c04481826 |url-status=live }}
Xiaomi MiPad,{{cite web |url=https://www.techindeep.com/xiaomi-mi-pad-7.9 |title=Xiaomi MiPad 7.9|publisher=Techindeep|access-date=18 May 2018}} Snail Games OBox, UTStarcom MC8718, Google Project Tango tablet,{{cite web |url=https://www.google.com/atap/projecttango/ |title=Google |access-date=2016-07-15 |archive-date=March 16, 2014 |archive-url=https://web.archive.org/web/20140316165229/https://www.google.com/atap/projecttango/ |url-status=live }} Fuze Tomahawk F1,{{cite web |last1=Rothman |first1=Chelsea |title=Fuze Tomahawk F1: The Chinese Android XStation 4 |url=http://www.cgmagonline.com/2016/05/11/fuze-tomahawk-f1-the-chinese-android-xstation-4/ |website=Comics Gaming Magazine |access-date=1 June 2016 |archive-date=June 10, 2016 |archive-url=https://web.archive.org/web/20160610104517/http://www.cgmagonline.com/2016/05/11/fuze-tomahawk-f1-the-chinese-android-xstation-4/ |url-status=live}}
Apalis TK1 System on Module,{{cite web |url=https://www.toradex.com/computer-on-modules/apalis-arm-family/nvidia-tegra-k1 |title=NVIDIA Tegra K1 System/Computer on Module – Apalis TK1 SOM |website=Toradex.com |access-date=2016-07-15 |archive-date=March 4, 2016 |archive-url=https://web.archive.org/web/20160304121311/https://www.toradex.com/computer-on-modules/apalis-arm-family/nvidia-tegra-k1 |url-status=live }} JXD Singularity S192{{cite web |url=https://www.gsmarena.com/jxd_s192_retro_gaming_tablet_is_powered_by_nvidias_tegra_k1_chipset-blog-17904.php |title=JXD S192 "retro" gaming tablet is powered by Nvidia's Tegra K1 chipset|website=GSMArena.com |access-date=March 25, 2019 |archive-date=March 25, 2019 |archive-url=https://web.archive.org/web/20190325121842/https://www.gsmarena.com/jxd_s192_retro_gaming_tablet_is_powered_by_nvidias_tegra_k1_chipset-blog-17904.php |url-status=live}}

T132

| HTC Nexus 9{{cite web |url=https://www.google.com/nexus/9/ |title=Nexus 9 |access-date=2016-07-15 |archive-date=October 21, 2014 |archive-url=https://web.archive.org/web/20141021121458/https://www.google.com/nexus/9/ |url-status=live }}{{cite web |url=http://www.htc.com/us/tablets/nexus-9/ |title=Google Nexus 9 Specs and Reviews {{pipe}} HTC United States |website=Htc.com |access-date=2016-07-15 |archive-url=https://web.archive.org/web/20141102023131/http://www.htc.com/us/tablets/nexus-9/ |archive-date=November 2, 2014 |url-status=dead }}

In December 2015, the web page of wccftech.com published an article stating that Tesla is going to use a Tegra K1 based design derived from the template of the Nvidia Visual Computing Module (VCM) for driving the infotainment systems and providing visual driving aid in the respective vehicle models of that time.[https://wccftech.com/tesla-autopilot-story-in-depth-technology/3/#ixzz3uLlKYLr6 Exclusive: The Tesla AutoPilot – An In-Depth Look At The Technology Behind the Engineering Marvel] {{Webarchive|url=https://web.archive.org/web/20180316084835/https://wccftech.com/tesla-autopilot-story-in-depth-technology/3/#ixzz3uLlKYLr6 |date=March 16, 2018 }} by Usman Pirzada on Dec 3, 2015 This news has, as of now, found no similar successor or other clear confirmation later on in any other place on such a combination of a multimedia with an auto pilot system for these vehicle models.

= {{Anchor|Jetson TX1}}Tegra X1 =

File:NVIDIA@20nm@TegraX1@Erista@Shield TV@S Taiwan 1517A1 NPW020.M3W TM670D-A1 DSC00807 (32803128780) (cropped).jpg]]

File:NVIDIA@20nm@TegraX1@Erista@Shield TV@S Taiwan 1517A1 NPW020.M3W TM670D-A1 Stack-DSC00878-DSC00919 - ZS-retouched (33028871972).jpg

Released in 2015, Nvidia's Tegra X1 (codenamed "Erista") features two CPU clusters, one with four ARM Cortex-A57 cores and the other with four ARM Cortex-A53 cores, as well as a Maxwell-based graphics processing unit.{{cite web |url=http://www.nvidia.com/object/tegra-x1-processor.html |title=Tegra X1 Super Chip {{pipe}} NVIDIA Tegra |publisher=NVIDIA |access-date=2016-07-10 |archive-date=January 5, 2015 |archive-url=https://web.archive.org/web/20150105071435/http://www.nvidia.com/object/tegra-x1-processor.html |url-status=live }}{{cite web |url=http://www.anandtech.com/show/8811/nvidia-tegra-x1-preview |title=NVIDIA Tegra X1 Preview & Architecture Analysis |website=Anandtech.com |access-date=2016-07-10 |archive-date=January 5, 2015 |archive-url=https://web.archive.org/web/20150105113525/http://www.anandtech.com/show/8811/nvidia-tegra-x1-preview |url-status=live }}

It supports Adaptive Scalable Texture Compression. Only one cluster of cores can be active at once, with the cluster switch being handled by software on the BPMP-L. Devices utilizing the Tegra X1 have only been seen to utilize the cluster with the more powerful ARM Cortex-A57 cores. The other cluster with four ARM Cortex-A53 cores cannot be accessed without first powering down the Cortex-A57 cores (both clusters must be in the CC6 off state).Tegra_X1_TRM_DP07225001_v1.0.pdf Nvidia has removed the ARM Cortex-A53 cores from later versions of technical documentation, implying that they have been removed from the die.{{cite web|url=https://devtalk.nvidia.com/default/topic/904289/jetson-tx1/does-anyone-get-8-cpus-listed-/post/4759807/#4759807|title=Tegra X1 advertised as four core to developers|date=December 19, 2015|publisher=NVIDIA|access-date=2017-04-04|archive-date=October 25, 2019|archive-url=https://web.archive.org/web/20191025060621/https://devtalk.nvidia.com/default/topic/904289/jetson-tx1/does-anyone-get-8-cpus-listed-/post/4759807/#4759807|url-status=live}}{{cite web |url=http://www.anandtech.com/show/9972/the-google-pixel-c-review/2 |title=Tegra X1's A53 cores are disabled on the Pixel C |publisher=Anandtech |access-date=2017-04-04 |archive-date=April 4, 2017 |archive-url=https://web.archive.org/web/20170404221055/http://www.anandtech.com/show/9972/the-google-pixel-c-review/2 |url-status=live }} The Tegra X1 was found to be vulnerable to a Fault Injection (FI) voltage glitching attack, which allowed for arbitrary code execution and homebrew software on the devices it was implemented in.{{cite book |last1=Bittner |first1=Otto |last2=Krachenfels |first2=Thilo |first3=Andreas |last3=Galauner |first4=Jean-Pierre |last4=Seifert |title=2021 Workshop on Fault Detection and Tolerance in Cryptography (FDTC) |chapter=The Forgotten Threat of Voltage Glitching: A Case Study on Nvidia Tegra X2 SoCs |date=16 August 2021 |pages=86–97 |doi=10.1109/FDTC53659.2021.00021 |arxiv=2108.06131v2 |isbn=978-1-6654-3673-1 |s2cid=237048483 }}

A revision (codenamed "Mariko") with greater power efficiency, known officially as Tegra X1+ was released in 2019,{{cite web|url=https://www.guru3d.com/articles_pages/nvidia_shield_android_tv_2019_review,2.html|title=NVIDIA Shield Android TV 2019 review|website=Guru3D.com|language=en-us|access-date=2020-03-25|archive-date=October 31, 2020|archive-url=https://web.archive.org/web/20201031223249/https://www.guru3d.com/articles_pages/nvidia_shield_android_tv_2019_review,2.html|url-status=live}} fixing the Fusée Gelée exploit. It's also known as T214 and T210B01.

  • CPU: ARMv8 ARM Cortex-A57 quad-core (64-bit) + (unused?) ARM Cortex-A53 quad-core (64-bit)
  • GPU: Maxwell-based 256 core GPU (Jetson Nano: only 128 cores)
  • MPEG-4 HEVC VP8 encoding/decoding & VP9 decoding support{{cite web |last=Crider |first=Michael |url=http://www.androidpolice.com/2015/01/04/nvidia-announces-the-new-tegra-x1-mobile-chipset-with-256-core-maxwell-gpu/ |title=NVIDIA Announces The New Tegra X1 Mobile Chipset With 256-Core Maxwell GPU |website=Androidpolice.com |date=January 5, 2015 |access-date=2016-07-10 |archive-date=January 5, 2015 |archive-url=https://web.archive.org/web/20150105104046/http://www.androidpolice.com/2015/01/04/nvidia-announces-the-new-tegra-x1-mobile-chipset-with-256-core-maxwell-gpu/ |url-status=live }} (Jetson Nano: encoders are H.265, H.264/Stereo, VP8, JPEG; decoders are H.265, H.264/Stereo, VP8, VP9, VC-1, MPEG-2, JPEG)
  • TSMC 20 nm process for the Tegra X1
  • TSMC 16 nm process for the Tegra X1+.
  • TDP:
  • T210: 15 W,{{cite web |url=https://devblogs.nvidia.com/parallelforall/nvidia-jetson-tx1-supercomputer-on-module-drives-next-wave-of-autonomous-machines/ |title=NVIDIA Jetson TX1 Supercomputer-on-Module Drives Next Wave of Autonomous Machines {{pipe}} Parallel Forall |website=Devblogs.nvidia.com |date=2015-11-11 |access-date=2016-07-15 |archive-date=May 3, 2016 |archive-url=https://web.archive.org/web/20160503041616/https://devblogs.nvidia.com/parallelforall/nvidia-jetson-tx1-supercomputer-on-module-drives-next-wave-of-autonomous-machines/ |url-status=live }} with average power consumption less than 10 W
  • Jetson Nano: 10 W (mode 0);{{cite web|url=https://info.nvidia.com/rs/156-OFN-742/images/Jetson_Nano_Webinar.pdf|title=Slide set from Jetson Nano webinar|access-date=May 3, 2019|archive-date=May 3, 2019|archive-url=https://web.archive.org/web/20190503091405/https://info.nvidia.com/rs/156-OFN-742/images/Jetson_Nano_Webinar.pdf|url-status=live}} mode 1: 5W (only 2 CPU cores @ 918 MHz, GPU @ 640 MHz)

class="wikitable" style="text-align:center;"
rowspan="2" | Model
number

! rowspan="2" | SoC / Variant

! rowspan="2" | Process

! CPU

! colspan="4" | GPU

! colspan="4" | Memory

!| Adoption

Processor
(Cores/Freq1)

! Micro-
architecture

! Frequency
(Core config2)

! GFLOPS
(FP32)

! GFLOPS
(FP16)

! Type

! Amount3

! Bus
width

! Band-
width4

! Availability

rowspan="2" | T210

| ODNX02-A2
TM670D-A1
TM670M-A2
TM671D-A2

| rowspan="2" | TSMC
20 nm

| 4x 2.2 GHz[https://nv-tegra.nvidia.com/r/gitweb?p=linux-4.9.git;a=blob;f=drivers/clk/tegra/clk-tegra124-dfll-fcpu.c;h=1d63c90cff0067d6798b86dad810e75aca047811;hb=refs/heads/l4t/l4t-r32.7.1-4.9#l98][https://nv-tegra.nvidia.com/r/gitweb?p=linux-4.9.git;a=blob;f=drivers/clk/tegra/clk-tegra124-dfll-fcpu.c;h=1d63c90cff0067d6798b86dad810e75aca047811;hb=refs/heads/l4t/l4t-r32.7.1-4.9#l513 Tegra T210 dfll table]
Cortex-A57 +
4x 1.3 GHz
Cortex-A53

| rowspan="2" | GM20B
(Maxwell)
{{rp|14}}

| 1000 MHz
(256:16:16)
{{cite web |url=http://developer.nvidia.com/embedded/dlc/tegra-x1-technical-reference-manual |title=Tegra X1 (SoC) Technical Reference Manual |website=developer.nvidia.com |language=en-US |access-date=2018-02-20 |edition=v1.2p }} {{registration required}}{{rp|753}}

| 512

| 1024

| LPDDR3
LPDDR4

| 8 GB

| rowspan="3" | 64 bit

| rowspan="2" | 25.6
GB/s

| Q2 2015

TM660M-A2

| 4x 1.4 GHz
Cortex-A57 +
4x 1.0 GHz
Cortex-A53

| 921 MHz
(128:16:16)
{{rp|773}}

| 236

| 472

| LPDDR3?
LPDDR4

| 4 GB

| March 2019

T214 /
T210b01

| ODNX10-A1
TM675M-A1

| TSMC 16 nm

| 4x 2.1 GHz[https://nv-tegra.nvidia.com/r/gitweb?p=linux-4.9.git;a=blob;f=drivers/clk/tegra/clk-tegra124-dfll-fcpu.c;h=1d63c90cff0067d6798b86dad810e75aca047811;hb=refs/heads/l4t/l4t-r32.7.1-4.9#l513 Tegra T210b01 dfll table]
Cortex-A57

| GM21B (Maxwell)
[https://cdn.discordapp.com/attachments/604689935483797524/1021885750897283132/unknown.png Strings found in libnvrm_gpu.so and in glxinfo when driver is loaded in linux]

| 1267 MHz
(256:16:16)
{{cite web |url=https://www.eurogamer.net/articles/digitalfoundry-2019-switch-new-tegra-x1-silicon-comes-into-focus |title=Switch's next Tegra X1 looks set to deliver more performance and longer battery life |last=Leadbetter |first=Richard |date=2019-06-27 |website=Eurogamer |language=en |access-date=2019-07-19 |archive-date=July 25, 2019 |archive-url=https://web.archive.org/web/20190725112546/https://www.eurogamer.net/articles/digitalfoundry-2019-switch-new-tegra-x1-silicon-comes-into-focus |url-status=live }}

| 649

| 1298

| LPDDR4
LPDDR4X

| 8 GB

| 34.1
GB/s

| Q2 2019

:1 CPU frequency may be clocked differently than the maximum validated by Nvidia at the OEM's discretion

:2 Unified Shaders : Texture mapping units : Render output units

:3 Maximum validated amount of memory, implementation is board specific

:4 Maximum validated memory bandwidth, implementation is board specific

== Devices ==

File:Nintendo-Switch-wJoyCons-BlRd-Standing-FL.jpg video game console.]]

class="wikitable"
Model

! SoC / Variant

! Devices

rowspan="6" | T210

| ODNX02-A2

| Nintendo Switch (2017, HAC-001) {{cite web|url=http://dystify.com/Overview/contents/Pages/Page_124923644.html|archive-url=https://web.archive.org/web/20170213231705/http://dystify.com/Overview/contents/Pages/Page_124923644.html|url-status=dead|archive-date=2017-02-13|title=3.3 Hardware Specifications|website=dystify.com|language=en-US|access-date=2017-02-27}}

TM670D-A1

| Nvidia Shield Android TV (2015)

TM670M-A2

| Nvidia Shield Android TV (2017)

TM660M-A2

| Jetson Nano 4 GB, Jetson Nano 2 GB

TM671D-A2

| Google Pixel C

{{Unknown}}

| Nvidia Jetson TX1 development board,{{cite web |date=2015-03-18 |title=Embedded Systems Development Solutions from NVIDIA Jetson |url=http://www.nvidia.com/object/embedded-systems.html |url-status=live |archive-url=https://web.archive.org/web/20160625121000/http://www.nvidia.com/object/embedded-systems.html |archive-date=June 25, 2016 |access-date=2016-07-10 |publisher=NVIDIA}} Nvidia Drive CX & PX

rowspan="2" | T210b01

| ODNX10-A1

| Nintendo Switch (2019, HAC-001(-01)), Nintendo Switch Lite (HDH-001),
Nintendo Switch: OLED Model (HEG-001)

TM675M-A1

| Nvidia Shield Android TV (2019)

= Tegra X2 =

Nvidia's Tegra X2{{cite web |title=DATA SHEET - NVIDIA Jetson TX2 System-on-Module.pdf |url=https://www.assured-systems.com/uploads/media/products/axiomtek/eboxs/jetson%20tx2/data%20sheet%20-%20nvidia%20jetson%20tx2%20system-on-module.pdf}}[https://devblogs.nvidia.com/jetson-tx2-delivers-twice-intelligence-edge/ NVIDIA Jetson TX2 Delivers Twice the Intelligence to the Edge] {{Webarchive|url=https://web.archive.org/web/20180227141512/https://devblogs.nvidia.com/jetson-tx2-delivers-twice-intelligence-edge/ |date=February 27, 2018 }} by Dustin Franklin on March 7, 2017 at Nvidia Developer Blogs (codenamed "Parker") features Nvidia's own custom general-purpose ARMv8-compatible core Denver 2 as well as code-named Pascal graphics processing core with GPGPU support.https://developer.nvidia.com/embedded/dlc/jetson-tx2-module-data-sheet {{registration required}} The chips are made using FinFET process technology using TSMC's 16 nm FinFET+ manufacturing process.{{cite web |url=http://www.anandtech.com/show/9902/nvidia-discloses-2016-tegra |title=NVIDIA Discloses Next-Generation Tegra SoC; Parker Inbound? |website=Anandtech.com |date=2016-01-05 |access-date=2016-07-10 |archive-date=June 29, 2016 |archive-url=https://web.archive.org/web/20160629140253/http://www.anandtech.com/show/9902/nvidia-discloses-2016-tegra |url-status=live }}{{cite web|url=https://www.anandtech.com/show/10596/hot-chips-2016-nvidia-discloses-tegra-parker-details|title=Hot Chips 2016: NVIDIA Discloses Tegra Parker Details|first=Joshua|last=Ho|website=www.anandtech.com|access-date=March 25, 2019|archive-date=March 25, 2019|archive-url=https://web.archive.org/web/20190325121844/https://www.anandtech.com/show/10596/hot-chips-2016-nvidia-discloses-tegra-parker-details|url-status=live}}{{cite news|last1=Ho|first1=Joshua|title=Hot Chips 2016: NVIDIA Discloses Tegra Parker Details|url=http://www.anandtech.com/show/10596/hot-chips-2016-nvidia-discloses-tegra-parker-details|access-date=25 August 2016|publisher=Anandtech|date=25 August 2016|archive-date=December 16, 2017|archive-url=https://web.archive.org/web/20171216100655/https://www.anandtech.com/show/10596/hot-chips-2016-nvidia-discloses-tegra-parker-details|url-status=live}}

  • CPU: Nvidia Denver2 ARMv8 (64-bit) dual-core + ARMv8 ARM Cortex-A57 quad-core (64-bit)
  • RAM: up to 8 GB LPDDR4{{cite web|url=https://www.nvidia.com/en-us/autonomous-machines/embedded-systems/jetson-tx2/|title=NVIDIA Jetson TX2: High Performance AI at the Edge|website=NVIDIA|access-date=April 9, 2019|archive-date=April 7, 2019|archive-url=https://web.archive.org/web/20190407201830/https://www.nvidia.com/en-us/autonomous-machines/embedded-systems/jetson-tx2/|url-status=live}}
  • GPU: Pascal-based, 256 CUDA cores; type: GP10B{{cite web | url=https://www.phoronix.com/scan.php?page=news_item&px=NVIDIA-Nouveau-GV11B-Volta-Xav | title=NVIDIA Bringing up Open-Source Volta GPU Support for Their Xavier SoC }}

  • TSMC 16 nm, FinFET process
  • TDP: 7.5–15 W[http://www.anandtech.com/show/11185/nvidia-announces-jetson-tx2-parker NVIDIA Announces Jetson TX2: Parker Comes To NVIDIA's Embedded System Kit] {{Webarchive|url=https://web.archive.org/web/20180108225553/http://www.anandtech.com/show/11185/nvidia-announces-jetson-tx2-parker |date=January 8, 2018 }}, March 7, 2017

class="wikitable" style="text-align:center;"
rowspan=2 | Model
number

! rowspan=2 | SoC
Variant

! CPU

! colspan=4 | GPU

! colspan=4 | Memory

! Adoption

Processor
(Cores / Freq)

! Micro-
architecture

! Frequency
(Core config1)

! GFLOPS
(FP32)

! GFLOPS
(FP16)

! Type

! Amount

! Bus
width

! Band-
width

! Availability

T186

| Tegra X2
(Parker)

| 2x 1.4–2.0 GHz
Denver2 +
4x 1.2–2.0 GHz
Cortex-A57

| GP10B
(Pascal)
[http://www.phoronix.com/scan.php?page=news_item&px=Tegra-X2-Nouveau-Support NVIDIA Rolls Out Tegra X2 GPU Support In Nouveau] {{Webarchive |url=https://web.archive.org/web/20170809090937/http://www.phoronix.com/scan.php?page=news_item&px=Tegra-X2-Nouveau-Support |date=August 9, 2017 }} by Michael Larabel at phoronix.com on March 29, 2017

| 854–1465 MHz
256:16:16
(2){{cite web |url=https://www.techpowerup.com/gpu-specs/jetson-tx2-gpu.c3231 |title=NVIDIA Jetson TX2 GPU Specs {{pipe}} TechPowerUp GPU Database |publisher=Techpowerup.com |date=August 22, 2022 |accessdate=2022-08-22}}

| 437–
750

| 874–
1500

| LPDDR4

| 8 GB

| 128 bit

| 59.7 GB/s

|

:1 Unified Shaders : Texture mapping units : Render output units (SM count)

== Devices ==

class="wikitable"
Model

! Devices

T186

| Nvidia Drive PX2 (variants),
ZF ProAI 1.1{{cite web |url=https://blogs.nvidia.com/blog/2017/01/04/zf-ces/|title=ZF Launches ProAI, DRIVE PX 2 Self-Driving System for Cars, Trucks, Factories – NVIDIA Blog |first=Danny |last=Shapiro |date=January 4, 2017 |website=The Official NVIDIA Blog |access-date=December 13, 2017|archive-date=December 14, 2017 |archive-url=https://web.archive.org/web/20171214014429/https://blogs.nvidia.com/blog/2017/01/04/zf-ces/ |url-status=live }}

T186

| Nvidia Jetson TX2

{{unk}}

| Mercedes-Benz MBUX (infotainment system)[https://blogs.nvidia.com/blog/2018/01/09/mercedes-ces-2018/ NVIDIA Powers Mercedes-Benz MBUX, Its Next-Gen AI Cockpit] {{Webarchive|url=https://web.archive.org/web/20180316023124/https://blogs.nvidia.com/blog/2018/01/09/mercedes-ces-2018/ |date=March 16, 2018 }} by Danny Shapiro on January 9, 2018 via Nvidia company blogs

{{unk}}

| 1 unit along with 1 GPU semiconductor is part of the ECU for "Tesla vision"
functionality in all Tesla vehicles since October 2016[https://electrek.co/2017/05/22/tesla-nvidia-supercomputer-self-driving-autopilot/ Look inside Tesla's onboard Nvidia supercomputer for self-driving] {{Webarchive|url=https://web.archive.org/web/20180328041335/https://electrek.co/2017/05/22/tesla-nvidia-supercomputer-self-driving-autopilot/ |date=March 28, 2018 }} by Fred Lambert on May 22, 2017[https://www.extremetech.com/extreme/256171-tesla-working-amd-self-driving-car Tesla Working With AMD on Self-Driving Car Processor] {{Webarchive |url=https://web.archive.org/web/20180315200046/https://www.extremetech.com/extreme/256171-tesla-working-amd-self-driving-car |date=March 15, 2018 }} by Joel Hruska on September 21, 2017

T186

| Magic Leap One{{Cite news |url=https://venturebeat.com/2018/07/11/magic-leap-one-will-ship-this-summer-with-nvidia-tegra-x2-processor/ |title=Magic Leap One will ship this summer with Nvidia Tegra X2 processor |date=2018-07-11 |work=VentureBeat |access-date=2018-07-11 |language=en-US |archive-date=July 12, 2018 |archive-url=https://web.archive.org/web/20180712031441/https://venturebeat.com/2018/07/11/magic-leap-one-will-ship-this-summer-with-nvidia-tegra-x2-processor/ |url-status=live }}[https://www.ifixit.com/Teardown/Magic+Leap+One+Teardown/112245 Magic Leap One teardown] {{Webarchive|url=https://web.archive.org/web/20180824112954/https://www.ifixit.com/Teardown/Magic+Leap+One+Teardown/112245 |date=August 24, 2018 }} at ifixit.com (mixed environment glasses)

{{unk}}

| Skydio 2 (drone)[https://www.cnet.com/news/skydios-second-gen-drone-a-1000-self-flying-action-cam-sells-out-for-2019/ Skydio's second-gen drone, a $1,000 self-flying action cam, sells out for 2019] {{Webarchive |url=https://web.archive.org/web/20200412103404/https://www.cnet.com/news/skydios-second-gen-drone-a-1000-self-flying-action-cam-sells-out-for-2019/ |date=April 12, 2020 }} by Stephen Shankland on October 2, 2019

= Xavier =

The Xavier Tegra SoC, named after the comic book character Professor X, was announced on 28 September 2016, and by March 2019, it had been released.{{cite web|url=https://devblogs.nvidia.com/nvidia-jetson-agx-xavier-32-teraops-ai-robotics/?ncid=so-fac-mdjngxxrmllhml-69163|title=NVIDIA Jetson AGX Xavier Delivers 32 TeraOps for New Era of AI in Robotics|first=Dustin|last=Franklin|date=December 12, 2018|website=devblogs.nvidia.com|access-date=March 30, 2019|archive-date=March 30, 2019|archive-url=https://web.archive.org/web/20190330120601/https://devblogs.nvidia.com/nvidia-jetson-agx-xavier-32-teraops-ai-robotics/?ncid=so-fac-mdjngxxrmllhml-69163|url-status=live}} It contains 7 billion transistors and 8 custom ARMv8 cores, a Volta GPU with 512 CUDA cores, an open sourced TPU (Tensor Processing Unit) called DLA (Deep Learning Accelerator).{{cite web|url=https://www.anandtech.com/show/11360/the-nvidia-gpu-tech-conference-2017-keynote-live-blog|title=The NVIDIA GPU Tech Conference 2017 Keynote Live Blog|first=Ryan|last=Smith|website=www.anandtech.com|access-date=March 25, 2019|archive-date=March 25, 2019|archive-url=https://web.archive.org/web/20190325121842/https://www.anandtech.com/show/11360/the-nvidia-gpu-tech-conference-2017-keynote-live-blog|url-status=live}}{{cite web|url=https://blogs.nvidia.com/blog/2017/05/24/ai-revolution-eating-software/|title=The AI Revolution Is Eating Software: NVIDIA Is Powering It {{pipe}} NVIDIA Blog|first=Jensen|last=Huang|date=May 24, 2017|website=The Official NVIDIA Blog|access-date=August 22, 2017|archive-date=August 22, 2017|archive-url=https://web.archive.org/web/20170822221216/https://blogs.nvidia.com/blog/2017/05/24/ai-revolution-eating-software/|url-status=live}} It is able to encode and decode 8K Ultra HD (7680×4320). Users can configure operating modes at 10 W, 15 W, and 30 W TDP as needed and the die size is 350 mm2.{{cite web|url=http://www.anandtech.com/show/10714/nvidia-teases-xavier-a-highperformance-arm-soc|title=NVIDIA Teases Xavier, a High-Performance ARM SoC for Drive PX & AI|last=Smith|first=Ryan|access-date=2016-09-28|archive-date=September 29, 2016|archive-url=https://web.archive.org/web/20160929143349/http://www.anandtech.com/show/10714/nvidia-teases-xavier-a-highperformance-arm-soc|url-status=live}}{{cite web|url=https://blogs.nvidia.com/blog/2016/09/28/xavier/|title=Introducing NVIDIA Xavier – NVIDIA Blog|first=Danny|last=Shapiro|date=September 28, 2016|website=The Official NVIDIA Blog|access-date=September 28, 2016|archive-date=October 2, 2016|archive-url=https://web.archive.org/web/20161002082808/https://blogs.nvidia.com/blog/2016/09/28/xavier/|url-status=live}}{{cite news|last1=Cutress|first1=Ian|last2=Tallis|first2=Billy|title=CES 2017: Nvidia Keynote Liveblog|url=http://www.anandtech.com/show/10999/ces-2017-nvidia-keynote-live-blog|access-date=9 January 2017|publisher=Anandtech.com|date=4 January 2016|archive-date=January 10, 2017|archive-url=https://web.archive.org/web/20170110090128/http://www.anandtech.com/show/10999/ces-2017-nvidia-keynote-live-blog|url-status=live}} Nvidia confirmed the fabrication process to be 12 nm FinFET at CES 2018.{{cite news|last1=Baldwin|first1=Roberto|title=NVIDIA unveils its powerful Xavier SOC for self-driving cars|url=https://www.engadget.com/2018/01/07/nvidia-xavier-soc-self-driving-cars/|access-date=8 January 2018|publisher=Engadget|date=8 January 2018|archive-date=January 8, 2018|archive-url=https://web.archive.org/web/20180108190349/https://www.engadget.com/2018/01/07/nvidia-xavier-soc-self-driving-cars/|url-status=live}}

  • CPU: Nvidia custom Carmel ARMv8.2-A (64-bit), 8 cores 10-wide superscalar[https://wccftech.com/nvidia-drive-xavier-soc-detailed/ NVIDIA Drive Xavier SOC Detailed] {{Webarchive|url=https://web.archive.org/web/20180224191519/https://wccftech.com/nvidia-drive-xavier-soc-detailed/ |date=February 24, 2018 }} by Hassan Mujtaba on Jan 8, 2018 via WccfTech
  • GPU: Volta-based, 512 CUDA cores with 1.4 TFLOPS;{{cite web|url=https://www.fudzilla.com/news/45328-nvidia-xavier-sampling-in-q1-18|title=Nvidia Xavier sampling in Q1 18|first=Fuad|last=Abazovic|website=www.fudzilla.com|access-date=February 6, 2018|archive-date=February 7, 2018|archive-url=https://web.archive.org/web/20180207122422/https://www.fudzilla.com/news/45328-nvidia-xavier-sampling-in-q1-18|url-status=live}} type: GV11B{{cite web | url=https://docs.nvidia.com/jetson/l4t/index.html#page/Tegra%20Linux%20Driver%20Package%20Development%20Guide/power_management_jetson_xavier.html | title=Welcome — Jetson LinuxDeveloper Guide 34.1 documentation }}
  • TSMC 12 nm, FinFET process
  • 20 TOPS DL and 160 SPECint @ 20 W; 30 TOPS DL @ 30 W (TOPS DL = Deep Learning Tera-Ops)
  • 20 TOPS DL via the GPU based tensor cores
  • 10 TOPS DL (INT8) via the DLA unit that shall achieve 5 TFLOPS (FP16)
  • 1.6 TOPS in the PVA unit (Programmable Vision Accelerator,{{cite web|url=https://www.freepatentsonline.com/y2016/0321074.html|title=Programmable Vision Accelerator|access-date=March 3, 2021|archive-date=February 27, 2021|archive-url=https://web.archive.org/web/20210227131227/http://www.freepatentsonline.com/y2016/0321074.html|url-status=live}} for StereoDisparity/OpticalFlow/ImageProcessing)
  • 1.5 GPix/s in the ISP unit (Image Signal Processor, with native full-range HDR and tile processing support)
  • Video processor for 1.2 GPix/s encoding and 1.8 GPix/s decode including 8k video support
  • MIPI-CSI-3 with 16 lanes{{cite web|url=https://www.electronicdesign.com/communications/understanding-mipi-alliance-interface-specifications|title=Understanding MIPI Alliance Interface Specifications|date=April 1, 2014|website=Electronic Design|access-date=March 25, 2019|archive-date=March 25, 2019|archive-url=https://web.archive.org/web/20190325121842/https://www.electronicdesign.com/communications/understanding-mipi-alliance-interface-specifications|url-status=live}}{{cite web|url=https://wccftech.com/nvidia-drive-xavier-soc-detailed/|title=NVIDIA Xavier SOC Is The Most Biggest and Complex SOC To Date|first=Hassan|last=Mujtaba|date=January 8, 2018|access-date=February 7, 2018|archive-date=February 24, 2018|archive-url=https://web.archive.org/web/20180224191519/https://wccftech.com/nvidia-drive-xavier-soc-detailed/|url-status=live}}
  • 1 Gbit/s Ethernet
  • 10 Gbit/s Ethernet

class="wikitable" style="text-align:center;"
rowspan="2" | Module
(Model)

! rowspan="2" | SoC Variant

! CPU

! colspan="4" | GPU

! Deep
Learning

! colspan="4" | Memory

! Adoption

! rowspan="2" | TDP
(W)

Processor
(Cores/Freq)

! Micro-
architecture

! Frequency
(Core config1)

! TFLOPS
(FP32)

! TFLOPS
(FP16)

! TOPS
(INT8)

! Type

! Amount

! Bus
width

! Band-
width

! Availability

Xavier AGX

| 64 GB

| rowspan="2" | Carmel
(12 MB cache)
8x 2.2 GHz

| rowspan="3" | GV11B
(Volta)

| rowspan="2" | 1377 MHz
512:64
(8, 4, 1)

| rowspan="2" | 1.41

| rowspan="2" | 2.82

| rowspan="2" | 32

| rowspan="2" | LPDDR4X

| 64 GB

| rowspan="2" | 256-bit

| rowspan="2" | 136.5 GB/s

|

| rowspan="2" | 10-30

Xavier AGX

| 32 GB

| 32 GB

|

Xavier AGX

| Industrial

| Carmel
(12 MB cache)
8x 2.0 GHz

| 1221 MHz
512:64
(8, 4, 1)

| 1.24

| 2.48

| 30

| LPDDR4X

| 32 GB

| 256-bit

| 136.5 GB/s

|

| 20-40

Xavier NX

| 16 GB

| rowspan="2" | Carmel
(10 MB cache)
6x 1.9 GHz

| rowspan="2" | Volta

| rowspan="2" | 1100 MHz
384:48
(6, 3, 1)

| rowspan="2" | 0.84

| rowspan="2" | 1.69

| rowspan="2" | 21

| rowspan="2" | LPDDR4X

| 16 GB

| rowspan="2" | 128-bit

| rowspan="2" | 59.7 GB/s

|

| rowspan="2" | 10-20

Xavier NX

| 8 GB

| 8 GB

|

:1 CUDA cores : Tensor cores (SMs, TPCs, GPCs)

== Devices ==

class="wikitable"
Model

! SoC Variant

! Devices

rowspan="9" | T194

| {{Unknown}}

| Nvidia Drive Xavier (Drive PX-series){{cite web |last=Schilling |first=Andreas |title=Auf Pegasus folgt Orin: Drive-PX-Plattform mit Turing- oder Ampere-Architektur |url=https://www.hardwareluxx.de/index.php/news/hardware/prozessoren/46029-auf-pegasus-folgt-orin-drive-px-plattform-mit-turing-oder-ampere-architektur.html |url-status=live |archive-url=https://web.archive.org/web/20180527201350/https://www.hardwareluxx.de/index.php/news/hardware/prozessoren/46029-auf-pegasus-folgt-orin-drive-px-plattform-mit-turing-oder-ampere-architektur.html |archive-date=May 27, 2018 |access-date=May 26, 2018 |website=Hardwareluxx|date=March 27, 2018 }}
(formerly named Xavier AI Car Supercomputer)

{{Unknown}}

| Nvidia Drive Pegasus (Drive PX-series)

{{Unknown}}

| Nvidia Drive AGX Xavier Developer Kit{{cite web|url=https://blogs.nvidia.com/blog/2018/09/12/nvidia-drive-agx-developer-kit-autonomous-driving/|title=Introducing NVIDIA DRIVE AGX Xavier Developer Kit – NVIDIA Blog|first=Shri|last=Sundaram|date=September 12, 2018|website=The Official NVIDIA Blog|access-date=December 11, 2018|archive-date=December 24, 2018|archive-url=https://web.archive.org/web/20181224142505/https://blogs.nvidia.com/blog/2018/09/12/nvidia-drive-agx-developer-kit-autonomous-driving/|url-status=live}}

{{Unknown}}

| Nvidia Jetson AGX Xavier Developer Kit{{cite web|url=https://developer.nvidia.com/embedded/buy/jetson-agx-xavier-devkit|title=Jetson AGX Xavier Developer Kit|date=July 9, 2018|website=NVIDIA Developer|access-date=March 25, 2019|archive-date=March 25, 2019|archive-url=https://web.archive.org/web/20190325121851/https://developer.nvidia.com/embedded/buy/jetson-agx-xavier-devkit|url-status=live}}

{{Unknown}}

| Nvidia Jetson Xavier

TE860M-A2

| Nvidia Jetson Xavier NX{{cite web|url=https://www.nvidia.com/en-us/autonomous-machines/embedded-systems/jetson-xavier-nx/|title=Jetson Xavier NX Developer Kit|date=November 6, 2019|website=NVIDIA Developer|access-date=November 6, 2019|archive-date=November 6, 2019|archive-url=https://web.archive.org/web/20191106185950/https://www.nvidia.com/en-us/autonomous-machines/embedded-systems/jetson-xavier-nx/|url-status=live}}

{{Unknown}}

| Nvidia Clara AGX{{cite web|url=https://blogs.nvidia.com/blog/2018/09/12/nvidia-clara-platform/|title=NVIDIA Clara Platform to Usher in Next Generation of Medical Instruments – NVIDIA Blog|first=Kimberly|last=Powell|date=September 12, 2018|website=The Official NVIDIA Blog|access-date=December 11, 2018|archive-date=December 15, 2018|archive-url=https://web.archive.org/web/20181215173558/https://blogs.nvidia.com/blog/2018/09/12/nvidia-clara-platform/|url-status=live}} "Clara AGX is based on NVIDIA Xavier and NVIDIA Turing GPUs."{{cite web|url=https://www.phoronix.com/scan.php?page=news_item&px=NVIDIA-GTC-Japan-2018|title=NVIDIA Rolls Out Tesla T4 GPUs, DRIVE AGX Xavier & Clara Platform – Phoronix|website=www.phoronix.com|access-date=December 11, 2018|archive-date=December 15, 2018|archive-url=https://web.archive.org/web/20181215221854/https://www.phoronix.com/scan.php?page=news_item&px=NVIDIA-GTC-Japan-2018|url-status=live}}{{unreliable source?|date=February 2021}}

{{Unknown}}

| Bosch and Nvidia designed Self Driving System{{cite news|last1=Shilov|first1=Anton|title=Bosch and Nvidia Team Up for Xavier based Self-Driving Systems for Mass Market Cars|url=http://www.anandtech.com/show/11205/bosch-and-nvidia-team-up-for-xavierbased-selfdriving-systems-for-mass-market-cars|access-date=22 June 2017|publisher=Anandtech.com|date=18 March 2017|archive-date=June 5, 2017|archive-url=https://web.archive.org/web/20170605215658/http://www.anandtech.com/show/11205/bosch-and-nvidia-team-up-for-xavierbased-selfdriving-systems-for-mass-market-cars|url-status=live}}

{{Unknown}}

| ZF ProAI{{cite web|url=https://vision.zf.com/site/magazine/en/articles_3392.html|title=Dream safety: 'Dream Car' learns to drive autonomously|website=vision.zf.com}}{{cite web|url=https://techwireasia.com/2018/01/baidu-nvidia-and-zf-team-to-drive-autonomous-vehicles-in-china/|title=Baidu, NVIDIA and ZF team to drive autonomous vehicles in China|date=January 8, 2018|website=Tech Wire Asia|access-date=March 25, 2019|archive-date=March 25, 2019|archive-url=https://web.archive.org/web/20190325121843/https://techwireasia.com/2018/01/baidu-nvidia-and-zf-team-to-drive-autonomous-vehicles-in-china/|url-status=live}}

On the Linux Kernel Mailing List, a Tegra194 based development board with type ID "P2972-0000" got reported:

:The board consists of the P2888 compute module and the P2822 baseboard.[http://lkml.iu.edu/hypermail/linux/kernel/1802.1/06008.html Linux Kernel Mailing List: (PATCH v3 7/7) arm64: tegra: Add device tree for the Tegra194 P2972-0000 board] {{Webarchive|url=https://web.archive.org/web/20180315200105/http://lkml.iu.edu/hypermail/linux/kernel/1802.1/06008.html |date=March 15, 2018 }} by Mikko Perttunen on Feb 15 2018

= Orin =

Nvidia announced the next-gen SoC codename Orin on March 27, 2018, at GPU Technology Conference 2018.{{cite web |url=https://www.anandtech.com/show/12598/nvidia-arm-soc-roadmap-updated-after-xavier-comes-orin |title=NVIDIA ARM SoC Roadmap Updated: After Xavier Comes Orin |first=Ryan |last=Smith |website=www.anandtech.com |access-date=April 18, 2018 |archive-date=April 19, 2018 |archive-url=https://web.archive.org/web/20180419053125/https://www.anandtech.com/show/12598/nvidia-arm-soc-roadmap-updated-after-xavier-comes-orin|url-status=live }}

:It contains 17 billion transistors and 12 ARM Hercules cores and is capable of 200 INT8 TOPs @ 65W.{{cite web |url=https://www.anandtech.com/show/15245/nvidia-details-drive-agx-orin-a-herculean-arm-automotive-soc-for-2022 |title=NVIDIA Details DRIVE AGX Orin: A Herculean Arm Automotive SoC For 2022 |last=Smith |first=Ryan |website=www.anandtech.com |access-date=2019-12-21 |archive-date=December 19, 2019 |archive-url=https://web.archive.org/web/20191219132408/https://www.anandtech.com/show/15245/nvidia-details-drive-agx-orin-a-herculean-arm-automotive-soc-for-2022 |url-status=live }}

:The Drive AGX Orin board system family was announced on December 18, 2019, at GTC China 2019.

Nvidia has sent papers to the press documenting that the known (from Xavier series) clock and voltage scaling on the semiconductors

:and by pairing multiple such chips a wider range of application can be realized with the thus resulting board concepts.{{cite web |url=https://www.heise.de/newsticker/meldung/Autonomes-Fahren-Nvidia-bringt-Kombiprozessor-Orin-mit-Next-Gen-GPU-4619375.html |title=Nvidia Orin: Next-Gen-Prozessor für autonome Fahrzeuge mit hoher Rechenleistung |first=heise |last=online |website=heise online |date=December 18, 2019 |access-date=January 26, 2021 |archive-date=January 31, 2021 |archive-url=https://web.archive.org/web/20210131162759/https://www.heise.de/newsticker/meldung/Autonomes-Fahren-Nvidia-bringt-Kombiprozessor-Orin-mit-Next-Gen-GPU-4619375.html |url-status=live }}

:In early 2021, Nvidia announced the Chinese vehicle company NIO will be using an Orin-based chip in their cars.{{cite web |url=https://blogs.nvidia.com/blog/2021/01/09/nio-selects-nvidia-intelligent-electric-vehicles/ |title=Chinese Automaker NIO Selects NVIDIA for Electric Vehicles {{pipe}} NVIDIA Blog |first=Danny |last=Shapiro |date=January 9, 2021 |website=The Official NVIDIA Blog |access-date=January 26, 2021 |archive-date=January 26, 2021 |archive-url=https://web.archive.org/web/20210126091412/https://blogs.nvidia.com/blog/2021/01/09/nio-selects-nvidia-intelligent-electric-vehicles/|url-status=live }}

The so far published specifications for Orin are:

  • CPU: 12× Arm Cortex-A78AE (Hercules) ARMv8.2-A (64-bit){{cite web |url=https://www.theregister.com/2020/09/29/arm_cortex_a78ae/ |title=Arm hasn't given up on self-driving car brains – its new Cortex-A78AE is going into Nvidia's Orin chip, for a start |first=Chris |last=Williams |website=www.theregister.com |access-date=2020-09-29 |archive-date=October 1, 2020 |archive-url=https://web.archive.org/web/20201001202035/https://www.theregister.com/2020/09/29/arm_cortex_a78ae/ |url-status=live }}{{cite web |url=https://www.arm.com/products/silicon-ip-cpu/cortex-a/cortex-a78ae|title=Cortex-A78AE – Arm |first=Arm |last=Ltd |website=Arm {{pipe}} The Architecture for the Digital World |access-date=October 3, 2020 |archive-date=October 5, 2020 |archive-url=https://web.archive.org/web/20201005151701/https://www.arm.com/products/silicon-ip-cpu/cortex-a/cortex-a78ae |url-status=live }}
  • GPU: Ampere-based, 2048https://blogs.nvidia.com/blog/2021/01/09/nio-selects-nvidia-intelligent-electric-vehicles/ {{Webarchive|url=https://web.archive.org/web/20210126091412/https://blogs.nvidia.com/blog/2021/01/09/nio-selects-nvidia-intelligent-electric-vehicles/ |date=January 26, 2021 }} 8192 cores / 4 SoCs = 2048 cores / SoC CUDA cores and 64 tensor cores1; "with up to 131 Sparse TOPs
    of INT8 Tensor compute, and up to 5.32 FP32 TFLOPs of CUDA compute."{{cite web |title=NVIDIA Jetson AGX Orin Technical Brief.pdf |url=https://www.nvidia.com/content/dam/en-zz/Solutions/gtcf21/jetson-orin/nvidia-jetson-agx-orin-technical-brief.pdf}}
  • 5.3 CUDA TFLOPs (FP32){{cite web |title=NVIDIA Orin Brings Arm and Ampere to the Edge at Hot Chips 34 |date=August 23, 2022 |url=https://www.servethehome.com/nvidia-orin-brings-arm-and-ampere-to-the-edge-at-hot-chips-34/}}
  • 10.6 CUDA TFLOPs (FP16)
  • Samsung 8 nm process
  • 275 TOPS (INT8) DL
  • 170 TOPS DL (INT8) via the GPU
  • 105 TOPS DL (INT8) via the 2x NVDLA 2.0 units (DLA, Deep Learning Accelerator)
  • 85 TOPS DL (FP16)
  • 5 TOPS in the PVA v2.0 unit (Programmable Vision Accelerator for Feature Tracking)
  • 1.85 GPix/s in the ISP unit (Image Signal Processor, with native full-range HDR and tile processing support)
  • Video processor for ? GPix/s encoding and ? GPix/s decode
  • 4× 10 Gbit/s Ethernet, 1× 1 Gbit/s Ethernet

:1 Orin uses the double-rate tensor cores in the A100, not the standard tensor cores in consumer Ampere GPUs.

Nvidia announced the latest member of the family, "Orin Nano" in September 2022 at the GPU Technology Conference 2022.{{cite web |url=https://nvidianews.nvidia.com/news/nvidia-jetson-orin-nano-sets-new-standard-for-entry-level-edge-ai-and-robotics-with-80x-performance-leap|title=NVIDIA Jetson Orin Nano Sets New Standard for Entry-Level Edge AI and Robotics With 80x Performance Leap |author=Nvidia |website=nvidianews.nvidia.com|access-date=September 23, 2022 |archive-date= September 23, 2022 |archive-url=https://web.archive.org/web/20220923183044/https://nvidianews.nvidia.com/news/nvidia-jetson-orin-nano-sets-new-standard-for-entry-level-edge-ai-and-robotics-with-80x-performance-leap |url-status=live }}

The Orin product line now features SoC and SoM (System-On-Module) based on the core Orin design and scaled for different uses from 60W all the way down to 5W. While less is known about the exact SoC's that are being manufactured, Nvidia has publicly shared detailed technical specifications about the entire Jetson Orin SoM product line. These module specifications illustrate how Orin scales providing insight into future devices that contain an Orin derived SoC.

class="wikitable" style="text-align:center;"
rowspan="2" | Module
(Model)

! rowspan="2" | SoC Variant

! CPU

! colspan="4" | GPU

! Deep
Learning

! colspan="4" | Memory

!| Adoption

! rowspan="2" | TDP
(W)

Processor
(Cores × Freq)

! Micro-
architecture

! Frequency
(Core config1)

! TFLOPS
(FP32)

! TFLOPS
(FP16)

! TOPS
(INT8)

! Type

! Amount

! Bus
width

! Band-
width

! Availability

Orin AGX
64 GB {{cite web |title=kernel/git/next/linux-next.git - The linux-next integration testing tree |url=https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/?id=b0e0423cfabc1eb407baee52cabbd9df2830feb0 |access-date=2020-09-22 |website=git.kernel.org }}{{cite web |url=https://www.phoronix.com/scan.php?page=news_item&px=Linux-5.10-ARM-Platform |title=Linux 5.10 Has Initial Support For NVIDIA Orin, DeviceTree For Purism's Librem 5 - Phoronix |website=www.phoronix.com |access-date=January 26, 2021 |archive-date=January 31, 2021 |archive-url=https://web.archive.org/web/20210131012357/https://www.phoronix.com/scan.php?page=news_item&px=Linux-5.10-ARM-Platform |url-status=live }}

|

| Cortex-A78AE
(9 MB cache)
12× 2.2 GHz

| rowspan="6" | Ampere

| 1300 MHz
2048:64:8
(16, 8, 2)

| 5.32

| 10.649

| 275

| rowspan="6" | LPDDR5

| 64 GB

| rowspan="2" | 256-bit

| rowspan="2" | 204.8 GB/s

| Sample 2021,
Kit Q1 2022,
Prod Dec 2022

| 15-60

Orin AGX
32 GB {{cite web|url=https://www.nvidia.com/en-us/autonomous-machines/embedded-systems/jetson-orin/#orion-prod-module-dev-kit-specs|title=Jetson Orin for Next-Gen Robotics NVIDIA |author=Nvidia |website=nvidia |access-date=September 23, 2022|archive-date=September 23, 2022 |archive-url=https://web.archive.org/web/20220923184905/https://www.nvidia.com/en-us/autonomous-machines/embedded-systems/jetson-orin/ |url-status=live }}

|

| Cortex-A78AE
(6 MB cache)
8× 2.2 GHz

| 930 MHz
1792:56:7
(14, 7, 2)

| 3.365

| 6.73

| 200

| 32 GB

| Oct 2022

| 15-40

Orin NX
16 GB

| TE980-M {{Cite web |date=September 28, 2022 |title=Jetson Orin NX Series - Thermal Design Guide |url-access=registration |url=https://developer.download.nvidia.com/assets/embedded/secure/jetson/orin_nx/docs/Jetson_Orin_NX_Series_Thermal_Design_Guide_TDG-11127-001_v1.0.pdf |access-date=September 29, 2022 }}{{Dead link|date=September 2023}}

| Cortex-A78AE
(6 MB cache)
8× 2.0 GHz

| 918 MHz
1024:32:4
(8, 4, 1)

| 1.88

| 3.76

| 100

| 16 GB

| rowspan="3" | 128-bit

| rowspan="2" | 102.4 GB/s

| Dec 2022

| 10-25

Orin NX
8 GB

| TE980-M

| Cortex-A78AE
(5.5 MB cache)
6× 2.0 GHz

| 765 MHz
1024:32:4
(8, 4, 1)

| 1.57

| 3.13

| 70

| rowspan="2" | 8 GB

| rowspan="3" | Jan 2023

| 10-20

Orin Nano
8 GB

|

| rowspan="2" | Cortex-A78AE
(5.5 MB cache)
6× 1.5 GHz

| 625 MHz
1024:32:4
(8, 4, 1)

| 1.28

| 2.56

| 40

| 68 GB/s

| 7-15

Orin Nano
4 GB

|

| 625 MHz
512:16:2
(4, 2, 1)

| 0.64

| 1.28

| 20

| 4 GB

| 64-bit

| 34 GB/s

| 5-10

:1 CUDA cores : Tensor cores : RT cores (SMs, TPCs, GPCs)

== Devices ==

class="wikitable"
Model

! Devices

! Comments

T234{{cite web |url=https://www.phoronix.com/scan.php?page=news_item&px=NVIDIA-Orin-Tegra234-Audio |title=Linux 5.18 Adding Audio Support for NVIDIA's Orin SoC }}

| Nvidia Jetson AGX Orin{{cite web|url=https://www.nvidia.com/en-us/autonomous-machines/embedded-systems/jetson-agx-orin/|title = NVIDIA Jetson AGX Orin}}

|comes in 32 GB and 64 GB RAM configurations, available as standalone module or devkit;
intended for industrial robotics and/or embedded HPC applications

{{unk}}

| Nvidia Jetson Orin NX{{cite web|url=https://www.nvidia.com/en-us/autonomous-machines/embedded-systems/jetson-orin-nx/|title=Embedded Robotics Modules - Jetson Orin NX|author=|website=nvidia|access-date=March 8, 2022|archive-date=March 8, 2022|archive-url=https://web.archive.org/web/20220308224158/https://www.nvidia.com/en-us/autonomous-machines/embedded-systems/jetson-orin-nx/|url-status=live}}

| mid-power SODIMM-form factor Orin-series module, available only as standalone module;
pin-compatible with Xavier NX carrier

{{unk}}

| Nvidia Jetson Orin Nano{{cite web |title=Jetson Orin for Next-Gen Robotics |url=https://www.nvidia.com/en-us/autonomous-machines/embedded-systems/jetson-orin/ |website=nvidia.com |publisher=NVIDIA Corporation |access-date=8 May 2023}}

| low-power, cost-effective SODIMM-form factor Orin-series module, available
as standalone module or devkit; intended for entry-level usage

{{unk}}

| Nvidia DRIVE AGX Orin

| used in automotive ADAS applications. 1×Orin 12×A78AE 32GB LPDDR5 @100W, 167+87 INT8 TOPS

{{unk}}

| Nio Adam{{cite web |url=https://blogs.nvidia.com/blog/2021/12/20/nio-et5-designed-autonomous-era-drive-orin/ |title=NIO ET5 Designed for Autonomous Era with DRIVE Orin|date=December 20, 2021 }}{{cite web |url=https://blogs.nvidia.com/blog/2021/01/09/nio-selects-nvidia-intelligent-electric-vehicles/ |title=Chinese Automaker NIO Selects NVIDIA for Electric Vehicles |date=January 9, 2021 }}

| built from 4× Nvidia Drive Orin, totals to 48 CPU cores and 8,192 CUDA cores;
for use in vehicles ET7 in March 2022 and ET5 in September 2022

= Grace =

The Grace CPU is an NVIDIA-developed ARM Neoverse V2 (Demeter) CPU platform, targeted at large-scale AI and HPC applications, available within several NVIDIA products. The NVIDIA OVX platform combines the Grace Superchip (two Grace dies on one board) with desktop NVIDIA GPUs in a server form-factor, while the NVIDIA HGX platform is available with either the Grace Superchip or the Grace Hopper Superchip.{{Cite web |title=Introducing Grace |url=https://www.nvidia.com/en-us/data-center/grace-cpu/ |access-date=2023-05-08 |website=NVIDIA |language=en-us}}

The latter is an HPC platform in of itself, combining a Grace CPU with a Hopper-based GPU, announced by NVIDIA on March 22, 2022.{{Cite web |title=NVIDIA Introduces Grace CPU Superchip |url=http://nvidianews.nvidia.com/news/nvidia-introduces-grace-cpu-superchip |access-date=2023-05-08 |website=NVIDIA Newsroom |language=en-us}}

Kernel patchsets indicate that a single Grace CPU is also known as T241, placing it under the Tegra SoC branding, despite the chip itself not including a GPU (a referenced T241 patchset cites impact to "NVIDIA server platforms that use more than two T241 chips...interconnected," pointing to the Grace Superchip design).{{Cite web |title=LKML: Marc Zyngier: Re: [PATCH] irqchip/gicv3: Workaround for NVIDIA erratum T241-FABRIC-4 |url=https://lkml.org/lkml/2023/3/7/162 |access-date=2023-05-08 |website=lkml.org}}

class="wikitable" style="text-align:center;"
rowspan="2" | Model
number

! colspan="3" |CPU

! colspan="4" | Memory

!| Adoption

Processor
(Cores/Frequency)

! Cache

! TFLOPS
(FP64)

! Type

! Amount

! Bus
width

! Band-
width

! Availability

T241{{cite web | url=https://www.spinics.net/lists/linux-gpio/msg66949.html | title=[PATCH 0/2] gpio: tegra186: Add support for Tegra241 — Linux GPIO }}

| Grace - 72x
ARM Neoverse
V2 cores
(ARMv9){{Cite web |date=2023-01-20 |title=NVIDIA Grace CPU Superchip Architecture In Depth |url=https://developer.nvidia.com/blog/nvidia-grace-cpu-superchip-architecture-in-depth/ |access-date=2023-05-08 |website=NVIDIA Technical Blog |language=en-US}}

| L1: 64 KB I-cache
+ 64 KB D-cache
per core
L2: 1 MB per core
L3: 117 MB shared

| 3.551

| LPDDR5X
ECC

| Up to
480 GB1

| ?

| 500 GB/s

| H2 2023{{Cite web |author1=Paul Alcorn |date=2023-03-22 |title=Nvidia CEO Comments on Grace CPU Delay, Teases Sampling Silicon |url=https://www.tomshardware.com/news/nvidia-ceo-jensen-huang-grace-delay |access-date=2023-05-08 |website=Tom's Hardware |language=en}}

1Figures cut in half from full Grace Superchip specification

= Atlan =

Nvidia announced the next-gen SoC codename Atlan on April 12, 2021,{{cite web |url=https://blogs.nvidia.com/blog/2021/04/12/nvidia-drive-atlan-autonomous-vehicle-platform/ |title = NVIDIA Unveils DRIVE Atlan Autonomous Vehicle Platform |date = April 12, 2021}} at GPU Technology Conference 2021.{{cite web |url=https://nvidianews.nvidia.com/news/nvidia-unveils-nvidia-drive-atlan-an-ai-data-center-on-wheels-fornext-gen-autonomous-vehicles |title = NVIDIA Unveils NVIDIA DRIVE Atlan, an AI Data Center on Wheels for Next-Gen Autonomous Vehicles}}

Nvidia announced the cancellation of Atlan on September 20, 2022, and their next SoC will be Thor.{{cite web |url=https://nvidianews.nvidia.com/news/nvidia-unveils-drive-thor-centralized-car-computer-unifying-cluster-infotainment-automated-driving-and-parking-in-a-single-cost-saving-system |title=NVIDIA Unveils DRIVE Thor — Centralized Car Computer Unifying Cluster, Infotainment, Automated Driving, and Parking in a Single, Cost-Saving System}}

Functional units known so far are:

  • Grace Next CPU
  • Ada Lovelace GPU{{cite web |title=NVIDIA Drops DRIVE Atlan SoC, Introduces 2 PFLOPS DRIVE Thor for 2025 Autos |url=https://www.anandtech.com/show/17582/nvidia-drops-drive-atlan-soc-introduces-2-pflops-drive-thor-for-2025-autos |website=Anandtech |access-date=6 January 2023}}
  • Bluefield DPU (Data Processing Unit)
  • other Accelerators
  • Security Engine
  • Functional Safety Island
  • On-Chip-Memory
  • External Memory Interface(s)
  • High-Speed-IO Interfaces

class="wikitable" style="text-align:center;"
rowspan="2" | Model
number

! CPU

! colspan="5" | GPU

! Deep
Learning

! colspan="4" | Memory

!| Adoption

Processor
(Cores/Freq)

! Micro-
architecture

! Core
config1

! Freq

! GFLOPS
(FP32)

! GFLOPS
(FP16)

! TOPS
(INT8)

! Type

! Amount

! Bus
width

! Band-
width

! Availability

T254?

| Grace-Next
(?/?)

| Ada
Lovelace

| ?

| ?

| ?

| ?

| >1000

| ?

| ?

| ?

| ?

| Cancelled

= Thor =

Nvidia announced the next-gen SoC codename Thor on September 20, 2022, at GPU Technology Conference 2022, replacing the cancelled Atlan.

A patchset adding support for Tegra264 to mainline Linux was submitted May 5, 2023, likely indicating initial support for Thor.{{Cite web |title='[PATCH 1/5] dt-bindings: mailbox: tegra: Document Tegra264 HSP' - MARC |url=https://marc.info/?l=linux-tegra&m=168354850122564&w=2 |access-date=2023-05-08 |website=marc.info}}

The ARM Neoverse V3AE (Poseidon-AE) CPU is built to deliver maximum performance for automotive applications, central compute and machine learning (ML) workloads.{{Cite web |title=Neoverse V3AE - CPU - Arm • www.arm.com › products › silicon-ip-cpu › neoverse |url=https://www.arm.com/products/silicon-ip-cpu/neoverse/neoverse-v3ae |website=arm.com}}

class="wikitable" style="text-align:center;"
rowspan="2" | Model
number

! CPU

! colspan="5" |GPU

! Deep
Learning

! colspan="4" | Memory

!| Adoption

Processor
(Cores/Freq)

! Micro-
architecture

! Core
config1

! Freq.

! TFLOPS
(FP32)

! TFLOPS
(FP16)

! TOPS
(FP8)

! Type

! Amount

! Bus
width

! Band-
width

! Availability

T264?

| 14x
Neoverse
V3AE{{cite web |title=NVIDIA DRIVE Thor Strikes AI Performance Balance, Uniting AV and Cockpit on a Single Computer|date=September 20, 2022 |url=https://blogs.nvidia.com/blog/2022/09/20/drive-thor/}}
(2.6 GHz)

| Blackwell

| 2560:96:?
(20, 10, 3)

| 1575 MHz

| 8.064

| 500

| 1035

| LPDDR5X

| 128 GB

| 256-bit

| 273 GB/s

| 2025

== Devices ==

  • Nvidia DRIVE Thor{{cite web | url=https://nvidianews.nvidia.com/news/nvidia-drive-powers-next-generation-transportation | title=NVIDIA DRIVE Powers Next Generation of Transportation — from Cars and Trucks to Robotaxis and Autonomous Delivery Vehicles }}
  • Jetson AGX Thor{{cite web | url=https://nvidianews.nvidia.com/news/foundation-model-isaac-robotics-platform | title=NVIDIA Announces Project GR00T Foundation Model for Humanoid Robots and Major Isaac Robotics Platform Update }}

Comparison

class="wikitable" style="font-size: 90%; text-align: center;"

|+

! colspan="2" |Generation

!Tegra 2
(Ventana)

!Tegra 3
(Kal-El)

!Tegra 4
(Wayne)

!Tegra 4i
(Grey)

! colspan="2" |Tegra K1
(Logan)

!Tegra X1
(Erista)

!Tegra{{nbs}}X1+
(Mariko)

!Tegra X2
(Parker)

!Tegra
Xavier

!Tegra
Orin

!Tegra
Thor

rowspan="6" |CPU

!Models

| T25 || T30/T33 || T114 || T148? || T124 || T132 || T210 || T214 || T186 || T194 || T234 || T264?

Cores

|2x
Cortex-A9

|4+1
Cortex-A9

|4+1
Cortex-A15

|4+1
Cortex-A9

|4+1
Cortex-A15

|2x
Denver

| colspan="2" |4x Cortex-A53
(disabled)
+ 4x Cortex-A57

|2x Denver2
+ 4x Cortex-A57

|8x
Carmel

|12x
Cortex-A78AE

|14x
Neoverse
V3AE

Instruction
set

| colspan="5" style="text-align:center;" |ARMv7-A (32-bit)

| colspan="4" style="text-align:center;" |ARMv8-A (64-bit)

| colspan="2" style="text-align:center;" |ARMv8.2-A (64-bit)

| style="text-align:center;" |ARMv9.2-A
(64-bit)

L1 cache
(I/D)

| colspan="5" |32/32 KB

|128/64 KB

| colspan="2" |32/32 KB +
64/32 KB

|128/64 KB +
48/32 KB

|128/64 KB

| colspan="2" |64/64 KB

L2 cache

| colspan="2" |1 MB

| colspan="4" |2 MB

| colspan="2" |128 KB + 2 MB

|2 MB + 2 MB

|8 MB

|3 MB

|14 MB

L3 cache

| colspan="9" |N/A

|4 MB

|6 MB

|16 MB

rowspan="4" |GPU

!Architecture

| colspan="4" |Vec4

| colspan="2" |Kepler

| colspan="2" |Maxwell

|Pascal

|Volta

|Ampere

|Blackwell

CUDA cores

|4+4*

|8+4*

|48+24*

|48+12*

| colspan="2" |192

| colspan="3" |256

|512

|2048

|2560

Tensor cores

| colspan="9" |N/A

| colspan="2" |64

|96

RT cores

| colspan="10" |N/A

|8

|?

rowspan="3" |RAM

!Protocol

|DDR2/
LPDDR2

|DDR3/
LPDDR2

| colspan="4" |DDR3/
LPDDR3

|LPDDR3/
LPDDR4

| colspan="3" |LPDDR4/
LPDDR4X

|LPDDR5

|LPDDR5X

Max. size

|1 GB

|2 GB

| colspan="2" |4 GB

| colspan="5" |8 GB

| colspan="2" |64 GB

|128 GB

Bandwidth

|2.7 GB/s

| colspan="2" |6.4 GB/s

|7.5 GB/s

| colspan="2" |14.9 GB/s

|25.6 GB/s

|34.1 GB/s

|59.7 GB/s

|136.5 GB/s

|204.8 GB/s

|273.0 GB/s

colspan="2" |Process

| colspan="2" |40 nm

| colspan="4" |28 nm

|20 nm

| colspan="2" |16 nm

|12 nm

|8 nm

|4 nm

* VLIW-based Vec4: Pixel shaders + Vertex shaders. Since Kepler, Unified shaders are used.

Software support

= FreeBSD =

FreeBSD supports a number of different Tegra models and generations, ranging from Tegra K1,{{cite web|url=https://kernelnomicon.org/?p=628|title=FreeBSD on Jetson TK1 {{pipe}} FreeBSD developer's notebook|website=kernelnomicon.org|access-date=December 26, 2020|archive-date=September 28, 2020|archive-url=https://web.archive.org/web/20200928061429/https://kernelnomicon.org/?p=628|url-status=live}} to Tegra 210.{{cite web|url=https://cgit.freebsd.org/src/commit/?id=b9cbd68d1cbbb21eade18182a797d5fa7d0dc11|title=src - FreeBSD source tree|website=cgit.freebsd.org}}

= Linux =

Nvidia distributes proprietary device drivers for Tegra through OEMs and as part of its "Linux for Tegra" (formerly "L4T") development kit, also Nvidia provides JetPack SDK with "Linux for Tegra" and other tools with it. The newer and more powerful devices of the Tegra family are now supported by Nvidia's own Vibrante Linux distribution. Vibrante comes with a larger set of Linux tools plus several Nvidia provided libraries for acceleration in the area of data processing and especially image processing for driving safety and automated driving up to the level of deep learning and neuronal networks that make e.g. heavy use of the CUDA capable accelerator blocks, and via OpenCV can make use of the NEON vector extensions of the ARM cores.

{{as of|April 2012}}, due to different "business needs" from that of their GeForce line of graphics cards, Nvidia and one of their Embedded Partners, Avionic Design GmbH from Germany, are also working on submitting open-source drivers for Tegra upstream to the mainline Linux kernel.{{cite mailing list

|url=http://lists.freedesktop.org/archives/dri-devel/2012-April/021833.html

|title=[RFC 0/4] Add NVIDIA Tegra DRM support

|date=April 20, 2012

|access-date=2012-08-21

|mailing-list=dri-devel

|last=Mayo

|first=Jon

|archive-date=December 25, 2014

|archive-url=https://web.archive.org/web/20141225171354/http://lists.freedesktop.org/archives/dri-devel/2012-April/021833.html

|url-status=live

}}{{cite web

|url=https://www.phoronix.com/scan.php?page=news_item&px=MTA4NjA

|title=A NVIDIA Tegra 2 DRM/KMS Driver Tips Up

|last=Larabel

|first=Michael

|date=April 11, 2012

|publisher=Phoronix Media

|access-date=2012-08-21

|archive-date=October 7, 2016

|archive-url=https://web.archive.org/web/20161007001029/https://www.phoronix.com/scan.php?page=news_item&px=MTA4NjA

|url-status=live

}} Nvidia co-founder & CEO laid out the Tegra processor roadmap using Ubuntu Unity in GPU Technology Conference 2013.{{cite web |url=https://www.youtube.com/watch?v=8kIQWWJs_po&t=9m35s |title=GTC 2013: NVIDIA's Tegra Roadmap (6 of 11) |date=March 19, 2013 |publisher=YouTube |access-date=2013-07-10 |archive-date=December 24, 2018 |archive-url=https://web.archive.org/web/20181224142710/https://www.youtube.com/watch?v=8kIQWWJs_po&t=9m35s |url-status=live }}{{unreliable source?|date=February 2021}}

By end of 2018 it is evident that Nvidia employees have contributed substantial code parts to make the T186 and T194 models run for HDMI display and audio with the upcoming official Linux kernel 4.21 in about Q1 2019. The affected software modules are the open source Nouveau and the closed source Nvidia graphics drivers along with the Nvidia proprietary CUDA interface.{{cite web|url=https://www.phoronix.com/scan.php?page=news_item&px=Tegra-X2-Xavier-HDMI-Audio|title=NVIDIA Tegra X2 & Xavier Get HDMI Audio With Linux 4.21 – Phoronix|website=www.phoronix.com|access-date=December 11, 2018|archive-date=December 23, 2018|archive-url=https://web.archive.org/web/20181223042635/https://www.phoronix.com/scan.php?page=news_item&px=Tegra-X2-Xavier-HDMI-Audio|url-status=live}}{{unreliable source?|date=February 2021}}

As of May, 2022, NVIDIA has open-sourced their GPU kernel modules for both Jetson and desktop platforms, allowing all but proprietary userspace libraries to be open-source on Tegra platforms with official NVIDIA drivers starting with T234 (Orin).{{Cite web |date=2022-05-19 |title=NVIDIA Releases Open-Source GPU Kernel Modules |url=https://developer.nvidia.com/blog/nvidia-releases-open-source-gpu-kernel-modules/ |access-date=2023-05-08 |website=NVIDIA Technical Blog |language=en-US}}

= QNX =

The Drive PX2 board was announced with QNX RTOS support at the April 2016 GPU Technology Conference.{{cite web|url=https://vrworld.com/2016/04/05/nvidia-drive-px2-next-gen-tegra-pascal-gpu/|title=DRIVE PX 2 Shows Next-Gen Nvidia Tegra, Pascal Processors|date=April 5, 2016|access-date=March 8, 2017|archive-date=March 8, 2017|archive-url=https://web.archive.org/web/20170308221220/https://vrworld.com/2016/04/05/nvidia-drive-px2-next-gen-tegra-pascal-gpu/|url-status=live}}

Similar platforms

SoCs and platforms with comparable specifications (e.g. audio/video input, output and processing capability, connectivity, programmability, entertainment/embedded/automotive capabilities & certifications, power consumption) are:

{{columns-list|colwidth=22em|

}}

See also

References

{{Reflist|30em}}

External links

  • [http://www.nvidia.com/object/tegra.html Official website]
  • [http://www.nvidia.com/object/apx_2500.html Nvidia's Tegra APX website]
  • [http://www.nvidia.com/docs/IO/55043/NVIDIA_Tegra_FAQ_External060408.pdf Nvidia's Tegra FAQ]
  • [http://international.download.nvidia.com/pdf/tegra/Tegra-X1-whitepaper-v1.0.pdf Tegra X1 Whitepaper]
  • [http://www.nvidia.com/content/PDF/tegra_white_papers/tegra-K1-whitepaper.pdf Tegra K1 Whitepaper]
  • [http://www.nvidia.com/docs/IO//116757/NVIDIA_Quad_a15_whitepaper_FINALv2.pdf Tegra 4 CPU Whitepaper]
  • [http://www.nvidia.com/docs/IO//116757/Tegra_4_GPU_Whitepaper_FINALv2.pdf Tegra 4 GPU Whitepaper]
  • [http://www.nvidia.com/content/PDF/tegra_white_papers/tegra-whitepaper-0911a.pdf Tegra 3 Whitepaper]
  • [http://www.nvidia.com/content/PDF/tegra_white_papers/Benefits-of-Multi-core-CPUs-in-Mobile-Devices_Ver1.2.pdf Tegra 2 Whitepaper]

{{Nvidia}}

{{ARM-based chips}}

Category:ARM-based systems on chips

Category:Mobile computers

Category:Nvidia hardware

Category:System on a chip