Nvidia Jetson AGX Xavier Developer Kit vs Jetson Nano Benchmark

Introduction

NVIDIA Jetson AGX Xavier is an embedded system-on-module (SoM) from the NVIDIA AGX Systems family. The Jetson AGX Xavier module makes AI-powered autonomous machines possible, running in as little as 10W and delivering up to 32 TOPs. It is a computer available now for embedded designers, researchers, and DIY makers, delivering the power of modern AI in a compact, easy-to-use platform with full software programmability. It is a small, powerful computer that lets you run multiple neural networks in parallel for applications like image classification, object detection, segmentation, and speech processing. 

NVIDIA announced the Jetson Nano Developer Kit at the 2019 NVIDIA GPU Technology Conference (GTC), a $99 [USD] computer available now for embedded designers, researchers, and DIY makers, delivering the power of modern AI in a compact, easy-to-use platform with full software programmability. NVIDIA^® Jetson Nano^™ Developer Kit is a small, powerful computer that lets you run multiple neural networks in parallel for applications like image classification, object detection, segmentation, and speech processing.

Platform	CPU	GPU	Memory	Storage	MRP
Jetson AGX Xavier	8x ARM v8.2 @ 2.26GHz	512x Volta GPU @ 64 Tensor Cores	32GB LPDDR4 (137GB/s)	32GB eMMC	$699
Jetson Nano	4x ARM Cortex A57 @ 1.43 GHz	128x Maxwell @ 921 MHz (472 GFLOPS)	4GB LPDDR4 (25.6 GB/s)	Micro SD	$99

Jetson Nano vs AGX Xavier Specifications

Features	Jetson Nano 4GB	Jetson Nano 2GB	Jetson AGX Xavier
CPU	ARM Cortex-A57 (quad-core) @ 1.43GHz	ARM Cortex-A57 (quad-core) @ 1.43GHz	NVIDIA Carmel ARMv8.2 (octal-core) @ 2.26GHz(4x2MB L2 + 4MB L3)
GPU	128-core NVIDIA Maxwell @ 921MHz	128-core NVIDIA Maxwell @ 921MHz	512-core Volta @ 1377 MHz + 64 Tensor Cores
DL	NVIDIA GPU support (CUDA, cuDNN, TensorRT)	NVIDIA GPU support (CUDA, cuDNN, TensorRT)	Dual NVIDIA Deep Learning Accelerators
Memory	4GB 64-bit LPDDR4 @ 1600MHz | 25.6 GB/s	2GB 64-bit LPDDR4 @ 1600MHz | 25.6 GB/s	32GB 256-bit LPDDR4x @ 2133MHz | 137GB/s
Storage	MicroSD card	MicroSD card	32GB eMMC 5.1
Vision	NVIDIA GPU support (CUDA, VisionWorks, OpenCV)	NVIDIA GPU support (CUDA, VisionWorks, OpenCV)	7-way VLIW Vision Accelerator
Encoder	(1x) 4Kp30, (2x) 1080p60, (4x) 1080p30	4Kp30 | 4x 1080p30 | 9x 720p30 (H.264/H.265)	(4x) 4Kp60, (8x) 4Kp30, (32x) 1080p30
Decoder	4Kp60, (2x) 4Kp30, (4x) 1080p60, (8x) 1080p30	4Kp60 | 2x 4Kp30 | 8x 1080p30 | 18x 720p30 (H.264/H.265)	(2x) 8Kp30, (6x) 4Kp60, (12x) 4Kp30
Camera	12 lanes MIPI CSI-2 | 1.5 Gbps per lane	1x MIPI CSI-2 connector	16 lanes MIPI CSI-2 | 6.8125Gbps per lane
Display	2x HDMI 2.0 / DP 1.2 / eDP 1.2 | 2x MIPI DSI	HDMI	(3x) eDP 1.4 / DP 1.2 / HDMI 2.0 @ 4Kp60
Wireless	M.2 Key-E site on carrier	802.11ac wireless	M.2 Key-E site on carrier
Ethernet	10/100/1000 BASE-T Ethernet	Gigabit Ethernet	10/100/1000 BASE-T Ethernet
USB	(4x) USB 3.0 + Micro-USB 2.0	1x USB 3.0 A | 2x USB 2.0 A | USB 2.0 Micro-B	(3x) USB 3.1 + (4x) USB 2.0
PCIe	PCIe Gen 2 x1/x2/x4		PCIe Gen 4 x16 | 1×8 + 1×4 + 1×2 + 2×1
CAN	NA	NA	Dual CAN bus controller
Misc IO	UART, SPI, I2C, I2S, GPIOs	UART, SPI, I2C, I2S, GPIOs	UART, SPI, I2C, I2S, GPIOs
Socket	260-pin edge connector, 45x70mm	260-pin edge connector, 45x70mm	699-pin board-to-board connector, 100x87mm
Thermals	-25°C to 80°C	-25°C to 80°C	-25°C to 80°C
Power	5/10W	5V / 5-10W	10/15/30W

Price Comparison Jetson Nano vs AGX Xavier

Port Comparison

Jetson Nano 4GB	Jetson Nano 2GB	AGX Xavier
MicroSD Card Slot	MicroSD Card Slot	MicroSD Card Slot
40-pin GPIO Expansion Header	40-pin GPIO Expansion Header	40-pin GPIO Expansion Header
Micro USB B	Micro USB B	USB C
Gigabit Ethernet Port	Gigabit Ethernet Port	Gigabit Ethernet Port
4x USB 3.0 A	1x USB 3.0 A | 2x USB 2.0 A	(3x) USB 3.1 + (4x) USB 2.0
HDMI | Display Port	HDMI	HDMI 2.0
DC Barrel Jack Power Input	NA	DC Barrel Jack Power Input
2x MIPI CSI	1x MIPI CSI	16x MIPI CSI-2

GPIO AGX Xavier

Connector Label	Pin	Pin	Connector Label
3.3 VDC	1	2	5.0 VDC
I2C_GP5_DAT	3	4	5.0 VDC
I2C_GP5_CLK	5	6	GND
MCLK05	7	8	UART1_TX
GND	9	10	UART1_RX
UART1_RTS	11	12	I2S2_CLK
PWM01	13	14	GND
GPIO27_PWM2	15	16	GPIO8
3.3 VDC	17	18	GPIO35_PWM3
SPI1_MOSI	19	20	GND
SPI1_MISO	21	22	GPIO17
SPI1_SCLK	23	24	SPI1_CS0
GND	25	26	SPI1_CS1
I2C_GP2_DAT	27	28	I2C_GP2_CLK
CAN0_DIN	29	30	GND
CAN0_DOUT	31	32	GPIO9_CAN1
CAN1_DOUT	33	34	GND
I2S_FS	35	36	UART1_CTS
CAN1_DIN	37	38	I2S_SDIN
GND	39	40	I2S_SDOUT
Connector Label	Pin	Pin	Connector Label

GPIO Jetson Nano

Connector Label	Pin	Pin	Connector Label
3.3 VDC	1	2	5.0 VDC
I2C_2_SDA	3	4	5.0 VDC
I2C_2_SCL	5	6	GND
AUDIO_MCLK	7	8	UART_2_TX
GND	9	10	UART_2_RX
UART_2_RTS	11	12	I2S_4_SCLK
SPI_2_SCK	13	14	GND
LCD_TE	15	16	SPI_2_CS1
3.3 VDC Power	17	18	SPI_2_CS0
SPI_1_MOSI	19	20	GND
SPI_1_MISO	21	22	SPI_2_MISO
SPI_1_SCK	23	24	SPI_1_CS0
GND	25	26	SPI_1_CS1
I2C_1_SDA	27	28	I2C_1_SCL
CAM_AF_EN	29	30	GND
GPIO_PZ0	31	32	LCD_BL_PWM
GPIO_PE6	33	34	GND
I2S_4_LRCK	35	36	UART_2_CTS
SPI_2_MOSI	37	38	I2S_4_SDIN
GND	39	40	I2S_4_SDOUT
Connector Label	Pin	Pin	Connector Label

Phoronix Test Suite :

So I started with some of the standard tests of the Phoronix Test Suite. You can also run the same test by executing the below commands.

sudo apt-get install -y php-cli php-xml

# Download PTS  https://www.phoronix-test-suite.com/ and  Install

phoronix-test-suite benchmark 1809111-RA-ARMLINUX005

# Accept for Dependency Installation and Wait a few hours, Test Result is available at : 

~/.phoronix-test-suite/test-results/

Test	Jetson Nano 4GB	AGX Xavier	Notes
Tinymembench (memcpy)	3501	6103.9	MB/s, More Is Better
TTSIOD 3D Renderer	41	73.29	FPS, More Is Better
7-Zip Compression	3501	10219	MIPS, More Is Better
C-Ray	932	328.39	Seconds, Lower Is Better
Primesieve	468	158.96	Seconds, Lower Is Better
AOBench	187	161.61	Seconds, Lower Is Better
FLAC Audio Encoding	104.01	90	Seconds, Lower Is Better
LAME MP3 Encoding	144.21	60.99	Seconds, Lower Is Better
Perl (Pod2html)	0.7114	0.5707	Seconds, Lower Is Better
Redis (GET)	568431	680922	Seconds, Lower Is Better
PyBench	7080	6359	Milliseconds, Lower Is Better

• 

•