• 24F., No.68, Sec.2, Wenhua 3rd Rd., Linkou Dist., New Taipei City
  • +886-963713897
  • sales@lambortech.com
  • ECM400-Thor
  • ECM400-Thor

ECM400-Thor

Groundbreaking AI Edge Platform, 2070 FP4 TFLOPS, 1×HDMI 2.0 Display
  • Powered by NVIDIA® Jetson Thor™ ( T5000 / T4000 ), up to 2070 / 1200 FP4 TFLOPS
  • 3-Axis Digital Accelerometer, 3-Axis Digital Gyroscope and 3-Axis Magnetometer
  • 1×M.2 M Key / 1×M.2 E Key / 1×M.2 B Key
  • Up to 5× USB3.2 Gen2 Type-C
  • 1×5GbE, 2×Custom High-speed Connectors for Expansion via Custom Cables
  • I2C/ SPI/ GPIO/ UART/ PWM, and CAN Bus (T5000 Only)

System

SOM NVIDIA Jetson Thor™ T5000 NVIDIA Jetson Thor™ T4000
CPU 14-Core ARM® Neoverse®-V3AE 64-bit CPU 12-Core ARM® Neoverse®-V3AE 64-bit CPU
GPU 2560-core NVIDIA Blackwell architecture GPU with 96 fifth-gen Tensor Cores 1536-core NVIDIA Blackwell architecture GPU with 64 fifth-gen Tensor Cores
AI Performance Up to 2070 FP4 TFLOPS Up to 1200 FP4 TFLOPS
Memory 128 GB 256-bit LPDDR5X
273 GB/s
64 GB 256-bit LPDDR5X
273 GB/s

Interface

Storage Supports External NVMe
Display Interface 1×HDMI2.0 (Up to 4K60)
Ethernet 1×RJ45 for 5GbE
DHCP Client
Expansion Slot Main Board
1×M.2 2230 E Key PCIe Gen4*1/USB2.0/SDIO Slot
1×M.2 2280 M Key PCIe Gen5×2 Slot
2×Custom High-speed connectors for external peripheral card (Up to PCIe Gen5×4 Each)
Daughter Board
1×M.2 2242/ 3042/ 3052 B Key USB3.2 Gen1 slot, nano SIM
USB 2× USB3.2 Gen2 Type-C (shared 10G)
1x micro USB (for console only)
MIPI 16×MIPI CSI-2 Lanes ( D-PHY 2.1 ( 40Gbps ), 4×4 | 3×4+2×2 | 2×4+4×2 | 1×4+5×2 | 6×2 MIPI Lanes, Support MIPI Camera, Capture Card )
CAN NVIDIA Jetson Thor™ T5000
2x CAN bus (header)
NVIDIA Jetson Thor™ T4000
N/A
Audio 1×3.5mm Line In or header
1×3.5mm Line Out or header
Peripheral Communication 6-axis IMU (3-Axis Accelerometer + 3-Axis Gyroscope)
Wafer or Pin Header
GPIO/ USB3/ USB2/ CAN*/ UART/ I2C/ SPI/ PWM/ FAN
Custom Header
CAN*/ I2C/ SPI/ GPIO/ UART/USB
Daughter Board
3-Axis magnetometer
3×USB3.2 Gen2 Type C (Shared 10G)
3×CAN bus (1 Shared with MB's CAN bus)*
2×RS485
RTK support + GNSS antenna connector (Optional)
* NVIDIA Jetson Thor™ T5000 only
Misc. Features Firmware Upgradable
AutoPower (Pin Header)

Development

OS Ubuntu: 20.04.2
Kernel 6.8.12-tegra or Higher
BSP Linux for Tegra(L4T) R38.2.0 or Higher
SDK NVIDIA JetPack™ 7.0.0

Environment

Power Supply DC input : 20~57V
Power Consumption TBA
Operating Temp. -20~45°C with Airflow
Storage Temp. -25~80°C
Certification CE, FCC

Mechanical

Main Board 100 x 130 mm

Video Feature

  NVIDIA Jetson Thor™ T5000 NVIDIA Jetson Thor™ T4000
Video Encode H.265 (UHP) :
6×4K60 | 12×4K30 | 26×1080p60 | 54×1080p30
H.264 (UHP) :
6×4K60 | 12×4K30 | 24×1080p60 | 48×1080p30
H.265 (UHP) :
3×4K60 | 6×4K30 | 13×1080p60 | 27×1080p30
H.264 (UHP) :
×4K60 | 6×4K30 | 12×1080p60 | 24×1080p30
Video Decode AV1 (Main Profile) :
2×8K30 | 6×4K60 | 12×4K30 | 26×1080p60 | 54×1080p30
H.265 (Main, Main10) :
4×8K30 | 10×4K60 | 20×4K30 | 40×1080p60 | 82×1080p30
H.264 (Baseline, Main, High) :
4×8K30 | 10×4K60 | 22×4K30 | 42×1080p60 | 84×1080p30
VP9 (Profile 0, Profile 2) :
2×8K30 | 8×4K60 | 16×4K30 | 32×1080p60 | 64×1080p30
AV1 (Main Profile) :
1×8K30 | 3×4K60 | 6×4K30 | 13×1080p60 | 17×1080p30
H.265 (Main, Main10) :
2×8K30 | 5×4K60 | 10×4K30 | 20×1080p60 | 41×1080p30
H.264 (Baseline, Main, High) :
2×8K30 | 5×4K60 | 11×4K30 | 21×1080p60 | 42×1080p30
VP9 (Profile 0, Profile 2) :
1×8K30 | 4×4K60 | 8×4K30 | 16×1080p60 | 32×1080p30

SDK

QCAP Capture
High Performance Renderer
Image Snapshot
Deinterlace, Alpha Blending Engine
Auto Signal Detection
2D/3D Video, Audio and VANC Streams Capture
Record
Encrypt / Sync / Clone / Recording
Time-Shifting / Rewind / Pre-Event / Recording
Multi-Streams ( 3D ) Recording
Animation Transition Effect
Video Cropping, Scaling and Alpha Blending Engine
Stream
2D/3D Universal Stream Client
2D/3D Multi-Streams Stream Server
RTSP, RTMP, HLS, SRT, TS, WebRTC. NDI-HX (*), Full NDI (*), Dante AV-H (*)
Animation Transition Effect
Video Cropping, Scaling and Alpha Blending Engine
*Separate License Required
QDEEP AI SDK Integrated Multiple Algorithms and Deep-Learning Models in Various Fields of Applications
Face Recognition
Objects Detection
Objects Segment
Optical Character Recognition
License Plate Recognition
Customizable Video AI Functions Upon Request