All News | Boards | Chips | Devices | Software | LinuxDevices.com Archive | About | Contact | Subscribe
Follow LinuxGizmos:
Twitter Facebook Pinterest RSS feed
*   get email updates   *

AIO-3588L AI Mainboard: 8K Display Support, High RAM, and 48MP ISP

Dec 26, 2023 — by Giorgio Mendoza 424 views

The AIO-3588L 8K AI is a development board powered by the Rockchip RK3588 series, featuring an octa-core 64-bit ARM architecture, optional up to 32GB of RAM, and 8K video encoding and decoding capabilities. This SBC by T-Firefly targets applications such as cloud servers, edge computing and other industrial settings.

According to the product page, the AIO-3588L main board features the Rockchip RK3588 processor, which is designed for commercial-grade performance and utilizes an advanced 8nm manufacturing process. It’s worth noting that there may also be System-on-Module variants of this processor available, catering to industrial and automotive grade applications.

— ADVERTISEMENT —


  • RK3588 — Quad-core Cortex-A76 (up to 2.4 GHz), Quad-core Cortex-A55 cores (up to 1.8 GHz); Arm Mali-G610 MP4 GPU; 6 TOPS NPU


RK3588 block diagram
(click image to enlarge)

For graphics and AI processing, the AIO-3588L is equipped with an ARM Mali-G610 MP4 quad-core GPU, supporting OpenGL ES3.2, OpenCL 2.2, and Vulkan 1.1, with a peak performance of 450 GFLOPS. Additionally, it boasts a Neural Processing Unit capable of handling INT4/INT8/INT16 mixed operations and supports popular AI frameworks like TensorFlow, MXNet, PyTorch, and Caffe.

Another key feature of the processor is its Imaging Signal Processor which supports up to 48MP resolution with HDR and 3DNR, ensuring high-quality image processing. In terms of video, it offers hardware decoding for various formats, including 8K@60fps H.265/VP9/AVS2, 8K@30fps H.264 AVC/MVC, 4K@60fps AV1, and 1080P@60fps MPEG-2/-1/VC-1/VP8. It also supports hardware encoding for 8K@30fps H.265/H.264.


AIO-3588L 8K AI main board bottom view
(click image to enlarge)

The AIO-3588L offers flexible memory configurations with LPDDR4/LPDDR4X options, ranging from 4GB to 32GB of RAM. For storage, it supports eMMC storage ranging from 32GB to 256GB. Additionally, it provides storage expansion options with SATA 3.0 and PCIe 2.0 2242 SSD expansion, as well as a MicroSD Card slot located on the bottom side of the board.

This AI main board comes with a comprehensive set of connectivity options, including Gigabit Ethernet, Wi-Fi 6, and 5G/4G expansion via M.2 slot, making it suitable for various networking requirements. 

T-Firefly indicates that this device supports “multi-channel 8K video output and achieves up to five displays”. There is also a double-row pin header that provides various interfaces, including UART, SPI, ADC, DP, USB 2.0, I2C  among other interfaces for further expansion.


AIO-3588L 8K AI main board top view
(click images to enlarge)

Specifications listed for the AIO-3588L 8K AI main board include:

  • Memory/Storage:
    • Up to 32GB LPDDR4/LPDDR4x
    • Up to 256GB eMMC (optional)
    • 1x MicroSD card slot
  • Video Output:
    • 1x HDMI 2.1 
    • 1x HDMI 2.0
    • 2x MIPI DSI
    • 1x DP 1.4
  • Video Input:
    • 1x HDMI-IN
    • 2x MIPI CSI 
  • ISP:
    • 48MP ISP w/ HDR & 3DNR
  • Expansion:
    • SATA 3.0/PCIe 2.0 2242 SSD 
  • Connectivity:
    • 1x GbE port
    • Wi-Fi /Bluetooth module support
    • 4G LTE via Mini PCIe
  • I/O Interfaces:
    • 44x I/O Expansion header
  • USB:
    • 2x USB 3.0
    • 2x USB 2.0
    • 1x USB Type-C port
  • Other Features:
    • 1x Fan 4-pin connector
  • Operating Temperature:
    • -20°C to 60°C
  • OS:
    • Android, Linux
  • Power:
    • 12V DC
  • Mechanical:
    • 122.9 x 85mm

Further information

As of the release date, the AIO-3588L mainboard is offered in three distinct variants:

  • A 4GB RAM + 32GB eMMC variant, available at a price of $229.00.
  • An 8GB RAM + 64GB eMMC variant, priced at $249.00.
  • A 16GB RAM + 128GB eMMC variant, which is available for $309.00.
(advertise here)


Print Friendly, PDF & Email
PLEASE COMMENT BELOW

Please comment here...