Lunar Lake AI CPU for portable computing
28 June 2024
DSP, Micros & Memory
Intel’s latest Core Ultra processor is unlike any that came before it. The company is hoping that its next generation CPU, recently announced at Computex, will usher in a comeback in AI-infused Copilot+ computers that have lately been overshadowed by Qualcomm’s processors.
Intel is again targeting the low power aspect of its new platform in a bid to outdo Qualcomm’s recently launched Snapdragon X Elite range. All the of E-cores in Lunar Lake have been moved to a low-power architecture, which will result in power savings while simultaneously providing a performance boost.
In a move that has surprised industry, Lunar Lake has also done away with hyperthreading, instead opting for more single-threaded cores on the CPU to handle the workload. This will hopefully produce better performance for high-intensity tasks.
Following in the footsteps of Apple and their new M-series CPUs, Intel now have embedded DRAM onto the chip package. Initially, Lunar Lake will be available with 16 or 32 GB of DRAM. This will provide faster access to RAM theoretically producing better performance, but at the disadvantage of not being able to upgrade at a later stage.
These CPUs will be popping up in laptops starting in H2 2024.
For more information, visit www.intel.com
Further reading:
Ultra-low power with 32 MHz of performance
Future Electronics
DSP, Micros & Memory
The Renesas RA0E1 delivers up to 32 MHz of CPU performance using an Arm Cortex-M23 core with a wide supply voltage range from 1,6 to 5,5 V.
Read more...
Precision sensor interface SoC
Altron Arrow
DSP, Micros & Memory
The MAX40109 is a low-power, precision sensor interface SoC that includes a high-precision, programmable analogue frontend, and digital signal processing.
Read more...
Entry-level MCU with classical peripherals
EBV Electrolink
DSP, Micros & Memory
NXP’s MCX C04x microcontrollers feature an Arm Cortex-M0+ core up to 48 MHz and offer 32 KB Flash, 2 KB SRAM, and 8 KB boot ROM.
Read more...
High-performance PCIe Gen 5 SSD controller
RS South Africa
DSP, Micros & Memory
Microchip’s latest NVMe controller chip has a sequential read performance of 14 GB/s for use in AI-accelerated servers under demanding workloads.
Read more...
Low-power MCUs with high-resolution analogue peripherals
Future Electronics
DSP, Micros & Memory
Renesas has introduced the RA2A2 series of microcontrollers to give design engineers a new option when they require high-performance, high-resolution analogue peripherals combined with low power consumption.
Read more...
Reference design for mobile robotics
Avnet Silica
DSP, Micros & Memory
NavQPlus is a reference design for a mission computer, AKA companion computer, for use in mobile robotics using ROS2 and similar applications such as ground stations and smart cameras.
Read more...
Evaluation board for automated guided vehicles
Avnet Silica
DSP, Micros & Memory
NXP’s MR-CANHUBK344 is a general-purpose evaluation board targeted for mobile robotics applications such as autonomous mobile robots and automated guided vehicles.
Read more...
General-purpose evaluation board
Altron Arrow
DSP, Micros & Memory
Based on the 32-bit Arm Cortex-M7 S32K3 MCU in a 172 HDQFP package, the S32K3X4EVB-T172 offers dual cores configured in lockstep mode, ASIL D safety hardware, and HSE security engine.
Read more...
Arduino’s AKX00051 PLC starter kit
TRX Electronics
DSP, Micros & Memory
The AKX00051 PLC starter kit bridges the gap between theoretical knowledge and practical skills for industrial automation, and offers a hands-on experience applicable to conveyor belt management, automated anomaly detection, real-time monitoring, and other applications.
Read more...
4G LTE&GNSS Click board for tracking applications
Dizzy Enterprises
DSP, Micros & Memory
MIKROE has launched its 4G LTE&GNSS Click, a compact add-on board designed for advanced global tracking and telematics applications, enabling developers to build highly efficient and smaller solutions, without compromising performance.
Read more...
printer friendly version