DFRobot Launches the LattePanda Sigma, a Compact But Beastly 12-Core 16-Thread Single-Board Computer
Absolutely miles ahead of the LattePanda 3 Delta in terms of performance, the LattePanda Sigma promises a new era in SBC compute.
DFRobot has announced the launch of the LattePanda Sigma, a 12-core Intel-based single-board computer (SBC) which absolutely blows its previous devices out of the water in terms of performance — putting it in a whole new class.
The LattePanda Sigma board is a follow-up to the LattePanda 3 Delta, unveiled in November 2021 as the company's fastest design yet — thanks to an 11th-generation quad-core Intel Celeron N5105 processor at its heart. The Sigma, by contrast, jumps away from the low-power Celeron range to a 13th-generation Intel Core i5-1340P — a 12-core 16-thread part running at boost speeds up to 4.6GHz on its four "Performance Cores" and 3.4GHz on its eight "Efficient Cores." Combined with integrated Intel Iris Xe graphics running at up to 1.45GHz, the Sigma is in an entirely different performance class.
Elsewhere on the board is 16GB of LPDDR5 dual-channel memory with 102.4GB/s bandwidth, two 2.5Gb/s Ethernet ports, two USB 2.0 Type-A ports, two USB 3.2 Gen. 2 Type-A ports, and two Thunderbolt 4 Type-C ports, HDMI 2.1, DisplayPort 1.4a via USB Type-C, and Embedded DisplayPort 1.4b supporting a total of four 4k displays, two M.2 M-key slot, one four-lane PCI Express Gen. 3.0 and one four-lane PCIe Gen. 4.0, an M.2 B-key slot with SATA, USB 2.0/3.0, PCIe Gen. 3.0, and connectivity to a micro-SIM slot, and an M.2 E-key slot with PCIe Gen. 3.0, USB 2.0, and Intel's CNVio connectivity, plus an RS232/RS485 header and SATA port for additional storage.
Despite all this, the dimensions of the board are impressively slim: 146×102mm (around 5.75×4.02"), admittedly larger than the 125×78mm (around 4.92×3.07") of the LattePanda 3 Delta but not exactly a behemoth. There have, however, been some losses. The board comes without Wi-Fi and Bluetooth connectivity, requiring the user to add their own using one of the M.2 slots, and there's no on-board eMMC storage.
Still present, thankfully, is the Arduino-compatible Microchip ATmega32U4-MU coprocessor — a microcontroller which can operate independently from the machine's main processor, running real-time workloads and interfacing with external hardware from its own general-purpose input/output (GPIO) header. As a result, there's no surprise to see DFRobot recommending the Sigma for home automation tasks — as well as gaming, media server duties, and even on-device machine learning workloads.
The LattePanda Sigma is now available to order on the DFRobot store at $579, including hefty heatsink and fan assembly but with no case; a bundle with 500GB SSD and Wi-Fi 6E module is priced at $648.