About
This document provides an overview of various embedded platforms/systems commonly used in the development of embedded applications in the industry. It covers a wide range of popular processors, microcontrollers, and development boards, highlighting their key features, performance, characteristics, and suitable use cases. By comparing different platforms such as Raspberry Pi, NVIDIA Jetson, ESP32, Arduino, and NXP i.MX Series, this document aims to assist in choosing the most appropriate solution based on their project requirements.
Embedded Platforms/Systems Overview
Raspberry Pi
A small low-cost computer the size of a credit card. A Raspberry Pi can be a desktop computer, robotics controller, server, game controller, etc. The Raspberry Pi is an excellent tool for many projects, from basic learning to advanced projects.
Pros
- Affordable
- Versatile operating system
- Very active community
- Easy to learn
- Flexible
- User Imagination
The Raspberry Pi stands out as a cost-effective solution. This makes it an excellent choice for hobbyists, educators, and professionals alike. Its versatility is its key strength, as it can be used for a wide range of applications. The Raspberry Pi community is also another major advantage. A wealth of resources, tutorials, and forums make troubleshooting straightforward.
Cons
- Limited power
- Limited processing power
- Limited RAM
- Limited Storage
Raspberry Pi comes with certain limitations. One of those is their limited power, which can restrict its ability to perform demanding tasks or multiple peripherals. Raspberry Pi also relies on external storage which offers slower speed and limited capacity. Although it is compatible with many open-source applications, some software may not be optimized or supported for the Raspberry Pi.
Applications
- Prototyping
- Automation
Models
- Raspberry Pi 5
- Raspberry Pi 500
- Raspberry Pi Pico
- Raspberry Pi Zero
Specifications
| Model | Raspberry Pi 5 | Raspberry Pi 500 | Raspberry Pi Pico | Raspberry Pi Zero |
|---|---|---|---|---|
| CPU | 2.4 GHz quad-core 64-bit ARM Cortex-A76 | 2.4 GHz quad-core 64-bit ARM Cortex-A76 | 133 MHz single-core ARM Cortex-M0+ | 1 GHz single-core ARM1176JZF-S (Zero 1), 1 GHz quad-core 64-bit ARM Cortex-A53 (Zero 2) |
| GPU | 800 MHz VideoCore VII | 800 MHz VideoCore VII | N/A | 250 MHz VideoCore IV |
| APU | N/A | N/A | N/A | N/A |
| RAM | 2/4/6/16GB LPDDR4x-4267 SDRAM | 2/4/6/16GB LPDDR4x-4267 SDRAM | 264kB SRAM | 512MB LPDDR2 SDRAM |
| Display | Up to Dual 4Kp60 micro HDMI with HDR support, 4Kp60 HEVC decoder | Up to Dual 4Kp60 micro HDMI with HDR support, 4Kp60 HEVC decoder | N/A | Up to 1080P mini HDMI |
| Ports | 2x USB 3.0, 2x USB 2.0, Gigabit Ethernet, PCIe Port | 2x USB 3.0, 2x USB 2.0, Gigabit Ethernet, PCIe Port | Micro-USB GPIO pins | Micro USB On-The-Go (OTG) |
| Storage options | eMMC, MicroSD Card | eMMC, MicroSD Card | Onboard 2MB flash memory | MicroSD Card |
| Network | Wi-Fi, BT 5.0, Gigabit Ethernet | Wi-Fi, BT 5.0, Gigabit Ethernet | N/A | Wi-Fi |
| Power | 5V/5A DC power via USB-C | 5V/5A DC power via USB-C | 5V via Micro-USB | 5V via Micro-USB |
| OS Support | Raspberry Pi OS, Ubuntu, Apertis, RISC OS, Ultramarine Linux, Alpine Linux, Other custom Linux distros (e.g. Yocto) | Raspberry Pi OS, Ubuntu, Apertis, RISC OS, Ultramarine Linux, Alpine Linux, Other custom Linux distros (e.g. Yocto) | MicroPython, CircuitPython | Raspberry Pi OS, RISC OS, Alpine Linux, Other custom Linux distros (e.g. Yocto) |
| Price Range | $50-$120 | $120-$150 | $4-$7 | $5-$15 |
| Documentation and Source Code | Documentation: Raspberry Pi Documentation,
Raspberry Pi Hardware Documentation Github: Raspberry Pi |
Documentation:Raspberry Pi Documentation,
Raspberry Pi Hardware Documentation Github: Raspberry Pi |
Documentation: Pico-series Microcontrollers - Raspberry Pi Documentation,
Github: Raspberry Pi Pico Examples |
Documentation: Raspberry Pi hardware - Raspberry Pi Documentation,
Github: dwelch67/raspberrypi-zero: Raspberry Pi Zero baremetal examples(for Zero 1), |
Why is Raspberry Pi Popular?
Raspberry Pi is popular because it is a small, affordable, single-board computer that is easy to use. It is also versatile for a wide range of applications and the community is very active. With its robust hardware capabilities, including support for multiple operating systems, it has become a favorite among professionals and hobby enthusiasts.
NVIDIA Jetson
NVIDIA Jetson is the leading platform for robotics and embedded edge AI applications. A series of embedded computing boards designed specifically for AI and machine learning applications. NVIDIA Jetson offers several models to different needs, from low-power to high-performance applications.
Pros
- Powerful GPU and CPU
- Versatile applications
- Open to Developers
- Wide range of options
One of the key strengths of NVIDIA Jetson is its GPU, designed to accelerate deep learning, computer vision, and AI. This makes it an excellent choice for robotics, IoT, and autonomous systems. It is also supported by popular AI frameworks such as TensorFlow, PyTorch, and OpenCV. NVIDIA Jetson has rich software support from NVIDIA’s JetPack SDK, which simplifies development and deployment.
Cons
- Expensive
- Limited GPIO
- Can get hot
NVIDIA Jetson tends to be on the more expensive side than other single-board computers. Devices under NVIDIA Jetson can consume more power, which may not be ideal for battery-powered applications or energy-efficient applications. It requires a steep learning curve to set up and work with the hardware and software tools.
Applications
- Robotics
- Edge AI
Models
- Jetson AGX Orin series
- Jetson Orin NX series
- Jetson Orin Nano series
- Jetson AGX Xavier series
- Jetson Xavier NX series
- Jetson TX2 series
- Jetson Nano
Specifications
| Model | Jetson AGX Orin Series | Jetson Orin NX Series | Jetson Orin Nano Series | Jetson AGX Xavier Series | Jetson Xavier NX Series | Jetson TX2 Series | Jetson Nano |
|---|---|---|---|---|---|---|---|
| CPU | 2.0/2.2 GHz 12-core NVIDIA Arm Cortex A78AE v8.2 64-bit CPU 3MB L2 + 6MB L3 | 2.0 GHz 8-core NVIDIA Arm Cortex A78AE v8.2 64-bit CPU 2MB L2 + 4MB L3, 2.0 6-core NVIDIA Arm Cortex A78AE v8.2 64-bit CPU 2MB L2 + 4MB L3 | 1.7 GHz 6-core Arm Cortex-A78AE v8.2 64-bit CPU 1.5MB L2 + 4MB L3 | 2.0/2.2 GHz 8-core NVIDIA Carmel Armv8.2 64-bit CPU 8MB L2 + 4MB L3 | 1.9 GHz 6-core NVIDIA Carmel Arm v8.2 64-bit CPU 6MB L2 + 4MB L3 | 1.95/2.2 GHz Dual-Core NVIDIA Denver 2 64-Bit CPU, 1.92/2 GHz Quad-Core Arm Cortex-A57 MPCore processor | 1.43 GHz Quad-Core Arm Cortex-A57 MPCore processor |
| GPU | 200-275 TOPS, NVIDIA Ampere architecture GPU with up to 64 Tensor Cores | 117-157 TOPS, NVIDIA Ampere architecture GPU with up to 32 Tensor Cores | 34-67 TOPS, NVIDIA Ampere architecture GPU with up to 32 Tensor Cores | 30-32 TOPS, NVIDIA Volta architecture GPU with 64 Tensor Cores | 21 TOPS 1100 MHz 384-core NVIDIA Volta architecture GPU with 48 Tensor Cores | 1.26-1.33 TFLOPS, NVIDIA Pascal architecture GPU | 472 GFLOPS 921 MHz 128-core NVIDIA Maxwell architecture GPU |
| APU | Integrated AI Accelerator (NVIDIA Tensor Cores) | Integrated AI Accelerator (NVIDIA Tensor Cores) | Integrated AI Accelerator (NVIDIA Tensor Cores) | Integrated AI Accelerator (NVIDIA Tensor Cores) | Integrated AI Accelerator (Tensor Cores) | Integrated AI Accelerator (NVIDIA Pascal GPU) | Integrated AI Accelerator (Maxwell GPU) |
| RAM | 32/64GB 256-bit LPDDR5 | 8/16GB 128-bit LPDDR5 | 4GB 64-bit LPDDR5, 8GB 128-bit LPDDR5 | 32/64GB 256-bit LPDDR4x | 8/16GB 128-bit LPDDR4x | 4/8GB 128-bit LPDDR4 | 4GB 64-bit LPDDR4 |
| Display | Up to 8K video output, dual 4K | Up to 4K video output | Up to 4K video output | Up to dual 4K video output | Up to 4K video output, Dual 4K displays | Up to 1080p video output | Up to 1080P video output |
| Ports | 3x USB 3.2 Gen2, 4x USB 2.0, 2x PCIe Gen 4 | 3x USB 3.2 Gen2, 3x USB 2.0, 1x PCIe Gen 4, Gigabit Ethernet | 3x USB 3.2 Gen2, 3x USB 2.0, 1x PCIe Gen 4, Gigabit Ethernet | 3x USB 3.2 Gen2, 4x USB 2.0, 1x PCIe Gen 3 | 1x USB3.2 Gen2, 3X USB 2.0 | 3x USB 3.0, 3x USB 2.0, 1x USB 3.0, 3x USB 2.0 | 1x USB 3.0, 3x USB 2.0 |
| Storage options | 32/64GB eMMC 5.1, M.2 NVMe | NVMe (supports external NVMe) | NVMe (supports external NVMe) | 32/64 GB eMMC 5.1 | 16GB eMMC 5.1 | 16/32GB eMMC 5.2 | 16GB eMMC 5.1 |
| Network | Wi-Fi 6, BT 5.0 | Wi-Fi, BT 5.0, Gigabit Ethernet | Wi-Fi, BT 5.0, Gigabit Ethernet | Wi-Fi, BT 5.0, Gigabit Ethernet | Wi-Fi, BT 5.0, Gigabit Ethernet | Wi-Fi, BT 4.2, Gigabit Ethernet | Gigabit Ethernet |
| Power | 15W-75W | 10W-40W | 10W-30W | 20W-40W | 10W-20W | 7.5W-20W | 5W-10W |
| OS Support | NVIDIA Jetpack SDK, Ubuntu, Other custom Linux distros (e.g. Yocto) | NVIDIA Jetpack SDK, Ubuntu, Other custom Linux distros (e.g. Yocto) | NVIDIA Jetpack SDK, Ubuntu, Other custom Linux distros (e.g. Yocto) | NVIDIA Jetpack SDK, Ubuntu, Other custom Linux distros (e.g. Yocto) | NVIDIA Jetpack SDK, Ubuntu, Other custom Linux distros (e.g. Yocto) | NVIDIA Jetpack SDK, Ubuntu, Other custom Linux distros (e.g. Yocto) | NVIDIA Jetpack SDK, Ubuntu, Other custom Linux distros (e.g. Yocto) |
| Price Range | $599-$1,100 | $199-$399 | $99-$199 | $699-$899 | $399-$599 | $300-$600 | $59-$99 |
| Documentation and Source Code | Documentation: Documentation: Jetson AGX Orin Developer Kit User Guide | Documentation: Documentation: Jetson Orin NX and Nano Series — Jetson Linux Developer Guide documentation | Documentation: Documentation: Jetson Orin NX and Nano Series — Jetson Linux Developer Guide documentation | Documentation: Documentation: Jetson AGX Xavier Series — Jetson Linux Developer Guide | Documentation: Jetson Xavier NX Series — Jetson Linux Developer Guide | Documentation: Jetson TX2 Developer Kit | Documentation: Jetson Nano Developer Kit, |
Why is NVIDIA Jetson Popular?
NVIDIA Jetson is popular because it provides high-performance AI computing in a compact, energy-efficient form factor. It is ideal for edge AI and robotics applications. From its entry-level to high-end modules, NVIDIA Jetson scales depending on the project's needs.
ESP32
Low-cost, low-power system-on-chip microcontrollers. A single 2.4 GHz WiFi and Bluetooth combo chip designed with TSMC low-power 40nm technology.
Pros
- Low-cost
- Low-power
- IoT (Internet of Things)
ESP32 is a powerful and versatile microcontroller because of its range of features, making it ideal for a variety of IoT applications. One of its biggest advantages is its low cost and low power consumption. It also has integrated Wi-Fi and Bluetooth, eliminating the need for external modules for wireless connectivity. ESP32 also has a strong developer community with support for ESP-IDF and is also compatible with Arduino IDE, making it great for beginners and experienced developers.
Cons
- Limited GPIO
- Low-power
- High learning curve
While ESP32’s performance is impressive, it may still fall short for resource-intensive tasks like high-end AI or complex video processing. Its lack of built-in storage can be limiting for applications that require large amounts of storage. Its GPIO pin count is also limiting for complex projects that require multiple connections.
Applications
- Simple smart home
- Low-power IoT sensors (ultrasonic, IR, etc.)
- Simple smart agriculture
- Speech & Image recognition (OpenCV)
Models
- ESP32-P series
- ESP32-S series
- ESP32-C series
- ESP32-H series
- ESP8266 series
Specifications
| Model | ESP32-P Series | ESP32-S Series | ESP32-C Series | ESP32-H Series | ESP8266 Series |
|---|---|---|---|---|---|
| CPU | Up to 240MHz Dual-core 32-bit Xtensa LX6 | Up to 240MHz Dual-core 32-bit Xtensa LX6 | Up to 160 MHz Single-core 32-bit Xtensa LX6 | Up to 160 MHz Single-core 32-bit Xtensa LX6 | Up to 160 MHz Single-core 32-bit Xtensa LX6 |
| GPU | N/A | N/A | N/A | N/A | N/A |
| APU | N/A | N/A | N/A | N/A | N/A |
| RAM | 320KB SRAM up to 512KB SRAM | 320KB SRAM up to 512KB SRAM | 128KB SRAM up to 192KB SRAM | 128KB SRAM up to 192KB SRAM | 128KB SRAM up to 192KB SRAM |
| Display | No direct display support. Can support external display | No direct display support. Can support external display | No direct display support. Can support external display | No direct display support. Can support external display | No direct display support. Can support external display |
| Ports | 1x SPI, 2x I2C, 2x UART, GPIO | 1x SPI, 2x I2C, 2x UART, GPIO | 1x SPI, 1x I2C, 1x UART, GPIO | 1x SPI, 1x I2C, 1x UART, GPIO | 1x SPI, 1x I2C, 1x UART, GPIO |
| Storage options | Up to 16MB external flash | Up to 16MB external flash | Up to 16MB external flash | Up to 16MB external flash | Up to 4MB external flash |
| Network | Wi-Fi, BT 4.2 | Wi-Fi, BT 4.2 | Wi-Fi, BT 4.2 | Wi-Fi, BT 4.2 | Wi-Fi |
| Power | 3.3V low power | 3.3V low power | 3.3V low power | 3.3V low power | 3.3V low power |
| OS Support | ESP-IDF, FreeRTOS, Arduino IDE | ESP-IDF, FreeRTOS, Arduino IDE | ESP-IDF, FreeRTOS, Arduino IDE | ESP-IDF, FreeRTOS, Arduino IDE | FreeRTOS, Arduino IDE |
| Price Range | $1-$3 | $2-$6 | $1.50-$5 | $1.50-$5 | $1-$3 |
| Documentation and Source Code | Documentation: Documentation: ESP-IDF Programming Guide - ESP32 - ESP-IDF Programming Guide v5.4 documentation,
Github: Espressif Systems, espressif/esp-idf: Espressif IoT Development Framework. Official development framework for Espressif SoCs. |
Documentation: Documentation: ESP-IDF Programming Guide - ESP32 - ESP-IDF Programming Guide v5.4 documentation,
Github: Espressif Systems, espressif/esp-idf: Espressif IoT Development Framework. Official development framework for Espressif SoCs. |
Documentation: Documentation: ESP-IDF Programming Guide - ESP32 - ESP-IDF Programming Guide v5.4 documentation,
Github: Espressif Systems, espressif/esp-idf: Espressif IoT Development Framework. Official development framework for Espressif SoCs. |
Documentation: Documentation: ESP-IDF Programming Guide - ESP32 - ESP-IDF Programming Guide v5.4 documentation,
Github: Espressif Systems, espressif/esp-idf: Espressif IoT Development Framework. Official development framework for Espressif SoCs. |
Documentation: Documentation: ESP-IDF Programming Guide - ESP32 - ESP-IDF Programming Guide v5.4 documentation,
Github: Espressif Systems, espressif/esp-idf: Espressif IoT Development Framework. Official development framework for Espressif SoCs. |
Why is ESP32 popular?
ESP32 is popular because it is cost-effective, has integrated Wi-Fi and Bluetooth, benefits from active community support, and has a well-developed ecosystem. ESP32s are heavily used in industries, especially in IoT. ESP32 offers robust connectivity and performance for a wide range of applications.
Texas Instruments Sitara Processors
Developed by Texas Instruments, featuring ARM9, ARM Cortex-A8, ARM Cortex-A9, ARM Cortex A-15, and ARM Cortex-A53 cores. Supported by the Beagle community as well as TI's open-source development community.
Pros
- Power efficient
- Cost-effective
- Scalability
- Real-time capability
Texas Instruments Sitara Processors offer a wide range of powerful features for embedded and industrial applications. Sitara Processors provide low-power and higher-performance models, making them suitable for battery-operated devices and high-demand industrial systems. They also offer a developer-friendly environment with tools like Code Composer Studio and RTOS support.
Cons
- Limited raw performance
- Complexity
- Limited high-end applications
Sitara processors have challenges related to complexity and cost. Simple applications may offer more performance than necessary, leading to increased power consumption and system cost. The software ecosystem is more specialized compared to platforms like Raspberry Pi or ESP32, making development challenging for those unfamiliar with Code Composer Studio and RTOS. Long-term support for certain models may also be a concern as some projects rely on specific processor families.
Applications
- IoT gateways
- Smart thermostats
- Industrial automation
- HMI
Models
- AM335x
- AM35x
- AM37x
- AM437x
- AM57x
- AM62x
- AM65x
Specifications
| Model | AM335x | AM35x | AM37x | AM437x | AM57x | AM62x | AM65x |
|---|---|---|---|---|---|---|---|
| CPU | ARM Cortex-A8, up to 1GHz | ARM Cortex-A8, up to 600MHz | ARM Cortex-A8, up to 1GHz | ARM Cortex-A9, up to 1GHz | Dual ARM Cortex-A15, up to 1.5GHz | Quad ARM Cortex-A53, up to 1.4GHz | Dual ARM Cortex-A53, up to 1.1GHz, Real-Time ARM Cortex-R5F |
| GPU | PowerVR SGX530 | N/A | PowerVR SGX530 | PowerVR SGX544 | PowerVR SGX544MP2 | ARM Mali-G52 MP2 | N/A |
| APU | PRU-ICSS | N/A | N/A | PRU-ICSS | Embedded Vision Engine (EVE), DSP C66x | N/A | PRU-ICSSG, Embedded Vision Engine (EVE) |
| RAM | Up to 1GB DDR2/3L | Up to 256MB DDR2 | Up to 512MB DDR2/3L | Up to 1GB DDR3/3L | Up to 4GB DDR3/3L, DDR4 | Up to 2GB DDR4 | Up to 8GB DDR4 |
| Display | 24-bit LCD Controller | 24-bit LCD Controller | 24-bit LCD Controller | 24-bit LCD Controller | 24-bit LCD Controller, HDMI | 24-bit LCD Controller | 24-bit LCD Controller, DP/HDMI |
| Ports | Up to 2x USB 2.0, Up to 2x Gigabit Ethernet, Up to 2x CAN ports, Up to 6x UARTS, Up to 3x I2C, GPIO | USB 2.0, Gigabit Ethernet, UART, I2C, GPIO | USB 2.0, Gigabit Ethernet, UART, I2C, GPIO | Up to 2x USB 2.0, Up to 2x Gigabit Ethernet, Up to 2x CAN ports, Up to 6x UARTS, Up to 3x I2C, GPIO | Up to 2x USB 2.0, Up to 2x USB 3.0, Up to 2x PCIe, Gigabit Ethernet, Up to 2x CAN ports, Up to 3x I2C, GPIO | Up to 2x USB 2.0, Up to 2x USB 3.0, Up to 2x PCIe, Gigabit Ethernet, Up to 2x CAN ports, Up to 3x I2C, GPIO | Up to 2x USB 2.0, Up to 2x USB 3.0, Up to 4x PCIe, Up to 2x Gigabit Ethernet, Up to 2x CAN ports, Up to 3x I2C, GPIO |
| Storage options | eMMC, SD/SDIO, NAND | NAND, SD | NAND, SD | eMMC, SD/SDIO, NAND | eMMC, SD/SDIO, NAND, SATA | eMMC, SD/SDIO | eMMC, SD/SDIO, NVMe |
| Network | Ethernet | Ethernet | Ethernet | Ethernet | Gigabit Ethernet | Gigabit Ethernet | Gigabit Ethernet |
| Power | 1.8V/3.3V | 1.8V/3.3V | 1.8V/3.3V | 1.8V/3.3V | 1.8V/3.3V | 1.8V/3.3V | 1.8V/3.3V |
| OS Support | Linaro, Linux (Yocto), RTOS, Android, Windows Embedded Compact | Linux, Windows CE | Linaro, Linux (Yocto), RTOS, Android, Windows Embedded Compact | Linaro, Linux (Yocto), RTOS, Android, Windows Embedded Compact | Linux (Yocto), RTOS, Android | Linux (Yocto), Android | Linux (Yocto), QNX, Android |
| Price Range | $5-$20 | $10-$25 | $15-$30 | $10-$25 | $50-$150 | $25-$60 | $70-$200 |
| Documentation and Source Code | Documentation: AM335x Sitara™ Processors datasheet (Rev. L), | Documentation: AM35x ARM Microprocessor Technical Reference Manual (Rev. C), | Documentation: AM3703 Sitara ARM Microprocessors datasheet (Rev. F), | Documentation: AM437x Sitara™ Processors datasheet (Rev. E), | Documentation: sprt689a.pdf, | Documentation: AM62x Sitara™ Processors datasheet (Rev. B), | Documentation: AM65x/DRA80xM Processors Technical Reference Manual (Rev. E), |
Why are Texas Instruments Sitara Processors popular?
Texas Instruments Sitara Processors are popular due to their reliability, cost-effectiveness, and scalable processing power. They excel in integrating key peripherals such as Ethernet, CAN, and industrial interfaces.
