So far, if you wanted to evaluate Renesas RA8P1 Cortex-M85 MCU clocked at 1 GHz, you had to spend close to $200 to get the EK-RA8P1 evaluation kit, but the RT-Thread RA8P1 Titan board allows you to do that for about $50.
Mostly designed for RT-Thread real-time OS development and evaluation, the RA8P1 Titan board comes with 32MB HyperRAM, 64MB HyperFlash, display and camera interfaces, two Gigabit Ethernet ports with Time Sensitive Networking support, WiFi 4 and Bluetooth 4.1, a USB 2.0 Type-C port, CAN Bus, Serial, and RS485 interfaces, a 40-pin Raspberry Pi GPIO header, a built-in debugger, and flexible power options with 5V via USB-C or 9V-24V via an XT60 connector.
RA8P1 Titan board specifications:
- SoC – Renesas RA8P1 (R7KA8P1)
- MCU cores
- Arm Cortex-M85 clocked at 1 GHz with Helium MVE (M-Profile Vector Extension) with 32KB I/D caches; 7300+ CoreMarks
- Arm Cortex-M33 clocked at 250 MHz with 32KB I/D caches
- GPU – D/AVE 2D drawing engine
- NPU – Arm Ethos-U55 NPU delivering 256 GOPS at 500 MHz
- Memory – 2 MB SRAM, 1MB MRAM
- Manufacturing Process – TSMC 22nm Ultra-Low Leakage (ULL) technology
- MCU cores
- Memory – 32MB HyperRAM (Winbond W95D8BNBYA5I)
- Storage
- 64MB HyperFlash (Winbond W32T51NWT)
- MicroSD card slot (4-bit SDIO, SDR25 support)
- Display I/F
- 22-pin MIPI DSI FPC connector
- 22-pin RGB565 LFC connector for LCD
- Camera I/F
- 22-pin CEU (Camera Engine Unit) FPC connector
- 22-pin MIPI CSI-2 FPC connector
- Networking
- 2x Gigabit Ethernet RJ45 ports with TSN support
- WiFi 4 (802.11b/g/n) and Bluetooth 4.1 module (CYW43438)
- USB – USB 2.0 FS/HS Type-C port
- Serial
- 2x CAN Bus interfaces
- 2x RS485 connectors
- 3x serial port connectors
- Sensors
- 6-axis accelerometer/gyroscope (BMI088 IMU)
- Magnetic compass (IST8310)
- Expansion – 40-pin GPIO header for Raspberry Pi HAT
- Debugging – DAP-Link debugger via USB-C port?
- Misc
- Boot/User and Reset buttons
- 3x LEDs
- Buzzer
- Power Supply
- 5V/1A via USB Type-C port
- 9-24V via 2-pin XT60 connector (Yellow)
- Dimensions – TBD
The hardware documentation provided on Medium is not super detailed and even wrong in some cases, so I did my best reconstructing it above using that Medium post, the photos, the PDF schematics, and a video on AliExpress.
Software looks better with basic instructions to get started with RT-Thread Studio and a long list of code samples from blinking an LED to using Gigabit Ethernet or the built-in NPU for AI face detection available on GitHub. You’ll also find the PDF schematics and Renesas RA8P1 datasheet and user manual there.
We’re also told developers can run embedded neural network models such as Yolo-Fastest, develop GUI applications with LVGL, and replies on RPMsg-Lite dual-core communication solution. They provided a few more details about the latter
RPMsg-Lite is a lightweight implementation of Remote Processor Messaging , primarily used for data exchange and message passing between different cores in multi-core processor systems. Based on the OpenAMP framework, it uses shared memory and virtual channels (Virtio Channels) to enable efficient and reliable inter-core communication. In this project, it runs on the Core0 (Cortex-M85) of the RA8P1 Titan Board as the master controller of the dual-core system, and its functions are as follows:
- Responsible for system initialization, peripheral resource coordination, and Core1 startup/management.
- Establishes a message channel (virtqueue/vring) with Core1 in shared memory through RPMsg-Lite, used for command dispatch, data exchange, event notification, etc.
- Main Features: multi-core Communication, RT-Thread support, shared Mmmory and cache safety, clear boot sequence (Core0 is responsible for loading/starting Core1 and performing Link-Up/handshake), and trimmable protocol layer
That makes it one of the most cost-effective high-performance MCU boards around, especially if you needs that 1 GHz frequency… The YAML file in the BSP mentions the board sells for 375 CNY ($52.6) in China. On AliExpress, the regular price is around $55, but it’s now priced at $50.75, at least until November 20, 2025. Adding a camera module brings the price up to $54.61.
Jean-Luc started CNX Software in 2010 as a part-time endeavor, before quitting his job as a software engineering manager, and starting to write daily news, and reviews full time later in 2011.
Support CNX Software! Donate via cryptocurrencies, become a Patron on Patreon, or purchase goods on Amazon or Aliexpress. We also use affiliate links in articles to earn commissions if you make a purchase after clicking on those links.
