ECS Unveils LIVA Q1L Mini PC with Dual Gigabit Ethernet, LIVA Q1D with HDMI & DisplayPort

LIVA Q1D

ECS introduces one of the smallest Intel mini PCs with LIVA Q Apollo Lake mini PC in 2017, followed by LIVA Q2 Gemini Lake micro PC the following year both just 70x70x31~33mm in size. Ian reviewed LIVA Q2 Micro PC with Windows 10 and Ubuntu 18.10 last year, and he was impressed with the performance considering the form factor, with the mini PC being perfectly suitable for browsing the web, home or office use, or as a digital signage solution. ECS has now updated its original LIVA Q model with the same processor options, but with LIVA Q1L offering two Gigabit Ethernet ports and one HDMI output, while LIVA Q1D comes with two video outputs (HDMI and DisplayPort) and a single Ethernet port. ECS LIVA Q1L/Q1D specifications: SoC (one or the other) Intel Apollo Lake Pentium N4200 quad-core processor @ 1.1 / 2.5 GHz with 18EU Intel HD graphics; 6W TDP Intel Apollo Lake Celeron N3450 quad-core processor @ 1.1 …

Support CNX Software – Donate via PayPal or become a Patron on Patreon

Year 2019 in Review – Top 10 Posts and Stats

Happy New Year 2020

2019 is closing to an end, or you may already be into 2020 while reading this post. In any case, that means it’s time to look back at 2019 and look forward to the events and new products to take place next year. While 2018 was a boring year for new processors, 2019 brought us some interesting new chips such as Amlogic S922X / A311D, or the first Arm Cortex-A55 only processors such as Amlogic S905X3. Rockchip RK3399Pro was promising when it was announced last year, but it never really took off. It was a pretty quiet year for Allwinner as well. RISC-V architecture has been ramping up with the first general-purpose RISC-V MCU: GD32V, WCH CH572 Bluetooth LE MCU, the launch of more SiFive RISC-V cores, and Kendryte K210 RISC-V AI processor announced last year has found its way into more and more boards. There have also been the usual launches of mobile Arm processors from Qualcomm, Samsung, and …

Support CNX Software – Donate via PayPal or become a Patron on Patreon

OnLogic Introduces Four Apollo Lake Pico-ITX SBC’s with Integrated Heatsink

We previously covered some Apollo Lake Pico-ITX boards with the likes of GIGABYTE GA-PICO3350 or AAEON PICO-APL3 motherboards, but there are now more options with OnLogic – formerly known as Logic Supply – recently launching four Apollo Lake Pico-ITX SBC’s. OnLogic EPM160, EPM161, EPM162, and EPM163 are powered by either an Intel Celeron N3350 dual-core processor or an Intel Pentium N4200 quad-core SoC with different memory capacities, display and networking options. OnLogic EPM16x specifications: SoC EPM160, EPM161, EPM162 – Intel Celeron N3350 dual-core Apollo Lake processor @ 1.1 GHz / 2.4 GHz (Burst frequency) with Intel HD Graphics 500 EPM163 – Intel Pentium N4200 quad-core Apollo Lake processor @ 1.1 GHz / 2.5 GHz (Burst frequency) with Intel HD Graphics 500 System Memory / Storage EPM160 – 1 GB LPDDR4-2133, MicroSD card slot, mSATA (shared PCIe Mini Card slot) EPM161 – 2 GB LPDDR4-2133, 8GB eMMC flash, MicroSD card slot, mSATA (shared PCIe Mini Card slot) EPM162 – 4 GB …

Support CNX Software – Donate via PayPal or become a Patron on Patreon

UP Xtreme Edge (UPX-Edge) Embedded Computer is Built around UP Xtreme Whiskey Lake SBC

UPX Edge Embedded Computer

AAEON UP Xtreme SBC was launched last June with a choice of 15W Whiskey Lake processors, namely Intel Celeron 4305UE dual-core processor, Intel Core i3-8145UE dual-core 4-thread processor, Intel Core i5-8365UE / Core i7-8665UE quad-core 8-thread processors. The company has now launched a complete embedded computer built around UP Xtreme SBC named UP Xtreme Edge Computing System, or simply UPX-Edge for short, that’s a slight variation of the fanless chassis offered for the SBC during the Kickstarter campaign. UPX-Edge specifications: SoC (one or the other) Intel Celeron 4305UE dual-core processor @ up to 2.2 GHz with 2MB cache, Intel UHD 610 graphics Intel Core i3-8145UE dual-core 4-thread processor @ up to 2.1 GHz / 3.9 GHz (Turbo) with 4MB cache, Intel UHD 620 graphics Intel Core i5-8365UE quad-core 8-thread processor @ up to 1.6 GHz / 4.1 GHz (Turbo) with 6MB cache, Intel UHD 620 graphics Intel Core i7-8665UE quad-core 8-thread processor @ up to 1.9 GHz / 4.8 GHz …

Support CNX Software – Donate via PayPal or become a Patron on Patreon

Ubuntu 18.04 on Beelink Gemini J45 Mini PC (Fix and Review)

Windows Mini PC Benchmarks Comparison

When I recently reviewed the Beelink J45 (aka Beelink Gemini J), a mini PC that uses the slightly older Intel Apollo Lake Pentium J4205 processor, whilst Windows 10 Pro ran fine it was unsuitable for Ubuntu because after installation the system became unstable and problems were encountered when running anything that loaded the system. The main issue was that when connected via wired-ethernet performing a command like ‘sudo apt upgrade’ would cause the ethernet to drop after which only a reboot would restore the connection. At the time it, was unclear what the cause was however a solution to the issue was posted by ‘gambetta’ on the Beelink forum. Basically it consists of installing the r8168 module which is the Linux device driver released for RealTek RTL8168B/8111B, RTL8168C/8111C, RTL8168CP/8111CP, RTL8168D/8111D, RTL8168DP/8111DP and RTL8168E/8111E Gigabit Ethernet controllers with PCI-Express interface. To paraphrase the ‘README.Debian’ file, you use ‘r8168-dkms because the in-kernel r8169 does not support your NIC or is not working …

Support CNX Software – Donate via PayPal or become a Patron on Patreon

A Look at Ubuntu on MINIX NEO G41V-4 and J50C-4 Mini PCs

Mini PC Comparison

MINIX Technology Limited recently released two new Gemini Lake mini PCs running Windows 10 Pro namely the MINIX NEO J50C-4 actively-cooled mini PC with an Intel Pentium Silver J5005 processor and the MINIX NEO G41V-4 fanless mini PC powered by an Intel Celeron N4100 processor. Whilst each mini PC comes with 64GB of eMMC with pre-installed Windows 10 Pro together with 4GB of RAM they also support the addition of an optional 2280 M.2 drive and the MINIX NEO J50C-4 allows optional memory upgrades. Prior to testing their performance under Ubuntu, I established a comparison baseline by updating Windows to version 1903 and then running my standard set of benchmarking tools first with the default configuration of each mini PC and then repeated having installed the official MINIX 2280 M.2 240GB drives for each device together with adding an extra 4GB RAM to the MINIX NEO J50C-4. The results can then compared them with other Intel mini PCs: Then for …

Support CNX Software – Donate via PayPal or become a Patron on Patreon

Intel Austin Beach is a Modular, Fanless NUC Powered by Intel NUC Compute Elements

Intel Austin Beach NUC + Chandler Bay NUC Compute Element

Unveiled last May, Intel NUC Compute Element is a family of cards that should cost less than the failed Computer Cards, and easy to handle than your typical computer-on-module. The first Compute Elements showcased at Computex 2019 were powered by a 15-Watt Core U-series processor, and now various models will find their way into Intel Austin Beach fanless NUC with “Chandler Bay” NUC Compute Elements powered by Whiskey Lake processors. Intel Austin Beach specifications: Chandler Bay Compute Element (one or the other, all 15-W Whiskey Lake) BKCM8CCB4R – Intel Celeron 4305U 2-core/2-thread processor with 4GB LPDDR3, 64GB eMMC storage BKCM8PCB4R – Intel Pentium Gold 5405U 2-core/4-thread processor with 4GB LPDDR3, 64GB eMMC storage BKCM8i3CB4N – Intel Core i3-8145U 2-core/4-thread processor with 4GB LPDDR3 BKCM8i5CB8N – Intel Core i5-8265U 4-core/8-thread processor with 8GB LPDDR3 BKCM8v5CB8N – Intel Core i5-8365U 4-core/8-thread vPro processor with 8GB LPDDR3 BKCM8i7CB8N – Intel Core i7-8565U 4-core/8 thread processor with 8GB LPDDR3 BKCM8v7CB8N – Intel Core i7-8665U …

Support CNX Software – Donate via PayPal or become a Patron on Patreon

Hacarus Embedded AI Computing Kit Leverages Sparse Modeling Technology

Hacarus AI Computing Kit Sparse Modeling Technology

AI training often requires thousands of samples to become accurate, and it can be costly and time-consuming, for example, if you want to train a model to detect manufacturing defects you’d need to provide images with both defective samples and good samples. Japanese AI experts at Hacarus have been working on a solution called Sparse Modeling which requires about 50 samples or even less for training, and worked with Congatec to provides an embedded AI computing kit leveraging the technology. Sparse Modeling Technology Hacarus does not go into great detail but explains Sparse Modeling technology is using a data modeling approach that focuses on identifying unique characteristics, in a way that humans recognize friends and family without having to look at everything from feet to head. That means algorithms based on Sparse Modeling do not need as much data as traditional AI solutions, leading to much smaller AI footprint suitable for fanless low-power systems operating 24/7, and/or that may a …

Support CNX Software – Donate via PayPal or become a Patron on Patreon