Publications & Demos
Unique reference design to deploy a variety of applications ranging from signal monitoring and video analytics to massively parallel data mining.
The CogniBlox board features four CM1K Chip or a total of 4096 neurons, a Field Programmable Gate Array (FPGA), memory and a variety of connectors to interface with sensors and other devices. Finally CogniBlox is stackable so the neural network can be expanded seamlessly by increment of 4096 neurons with no impact on the program interfacing with the base board.
|CogniBlox w/Express SDK||$1,500|
|Full Featured SDK||$2,500|
- 4 CM1K chips
- Lattice XP2 FPGA with 40,000 logic elements
- 2 Mbytes (2M x16bits) Everspin MRAM, 35 ns access time
- 2 “spinal” spring-loaded 18-pin connectors for vertical stackability of up to 20 CogniBlox modules.
- 4 “cardinal” 8-pin push-in connectors for communication with sensors and other devices including another stack of CogniBlox.
- Typical interface can be LVDS, I2C, SPI or else. Switch configuration allows for 3 differential pairs and 2 single-ended wires, or eight single-ended wires
- USB connector to USB high speed 480 Mbit FTDI chip
- 24v power supply, Hot swap regulator
|Frequently Asked Questions|
|Additional tutorials and demonstrations.|