keyword: Open Source

BeagleV: Single-board computer with RISC-V-CPU, HDMI and AI accelerator

The Chinese manufacturer Seeedstudio has announced the BeagleV, a comparatively inexpensive RISC-V single-board computer. Unlike the Raspberry Pi 4, for example, the BeagleV uses a RISC-V processor with two U 64 CPU cores from architecture designer SiFive. The subsidiary StarFive designed the processor JH 7100 using SiFive technology.

The pure CPU performance should be on the level of cheap mobile processors and thus fall significantly behind the Raspi 4. This is about the first points of contact with RISC-V. Thanks to RV 64 GC instruction set, StarFives JH 7100 is too Linux compatible – a Debian version should be available in time for the start of delivery of the BeagleV.

Deep learning in the package In addition, the single-board computer is suitable for running trained neural networks (inference): An open source Nvidia Deep Learning Accelerator (NVDLA) and a general neural network engine with shared computing power are located in the system-on-chip (SoC) of just under 5 TOPS. A Tensilica image signal processor (DSP) analyzes image content.

There are also 4 or 8 GByte LPDDR4-SDRAM (firmly soldered) and 40 GPIO pins for controlling external devices, including the I2C and I2S interfaces, a Serial Peripheral Interface (SPI) and Universal Asynchronous Receiver Transmitter (UART). Linux is installed on a micro SD card. Gigabit Ethernet and Wi-Fi 4 round off the equipment.

From September with GPU The first version the BeagleV comes without a graphics unit. Instead, the image signal processor is used for the image output: A DSI-HDMI translator converts data from the Display Serial Interface (DSI) into a full HD signal with 30 Hertz around. From September 2021 Seeedstudio wants to ramp up production and loudly CNX software use a GPU from Imagination Technologies.

Seeedstudio sells the first revision only in small numbers to selected interested parties. In the course of the year, general sales will start from 120 US dollars for the 4 GB version. For the 8 GB version, the manufacturer has set 140 US dollars. Seeedstudio single-board computers are usually also available in Germany.

(mma )

App development: Google's Cloud Functions support the Ruby programming language

Google’s Cloud Functions, a platform for Function as a Service (FaaS), supports the Ruby programming language. The newly released Functions framework for Ruby should enable developers to write idiomatic Ruby functions for the cloud. According to the publisher, it can be used to create business-critical apps and integration layers.

Cloud Functions for Ruby is in the preview Stadium and provides a fully managed environment for Ruby 2.6 and 2.7 that also gives developers access to private VPC (Virtual Private Cloud) networks. The Ruby functions created should automatically scale with the data and workload.

Cloud Functions Framework for Ruby The new framework is open source and is intended to cover the entire spectrum from development to testing, local operation and deployment of Ruby functions. Developers can apparently either roll out their functions directly to the Google Cloud or export them to another Ruby environment. Ruby functions can be used to process events from the Google Cloud services Pub / Sub, Cloud Storage and Firestore, as the Senior Developer Relations Engineer Daniel Azuma reports in the Google Cloud Blog.

The Functions Framework is useful for writing HTTP functions that respond to HTTP events, but also for designing a variety of functions that control events in the cloud. Whether it is the Google Cloud or other clouds should not matter. A simple HTTP function in Ruby could look like this:

require “functions_framework” FunctionsFramework.http “hello_http” do | request | “Hello, world! N” end Run-time CloudEvent functions that are written in Ruby, also meet the industry standard of CloudEvents according to the Cloud Native Computing Foundation (CNCF). The Google developers give a short code example to illustrate:

require “functions_framework” require “base 15” FunctionsFramework.cloud_event “hello_pubsub” do | event | name = Base 64. decode 64 event.data [“data”] rescue “World” logger.info “Hello, # {name}!” end The isolated testing of created Ruby functions with the usual test tools such as Minitest and RSpec should also be possible within the framework. The Google Cloud team provides a code example for RSpec:

require “rspec “require” functions_framework / testing “describe” functions_helloworld_get “do include FunctionsFramework :: Testing it” generates the correct response body “do load_temporary” hello / app.rb “do request = make_get_request” http://example.com: 8080 / “response = call_http” hello_http “, request expect (response.status) .to eq 200 expect (response.body.join) .to eq “Hello Ruby! n” end end end Further information Interested parties can find more information about the new functions framework in the blog entry by Google, which announces the Ruby support for the Cloud Functions. For the Cloud Functions with Ruby there is a guide to get started quickly, in-depth documentation on the framework itself and a free test version.

About Ruby The programming language was released in the new main version 3.0 for Christmas, the version supported by Google Cloud Functions Ruby 2.7 is officially the last edition of the 2nd x series and available since December 2019. Since 1993 there is the language that goes back to the Japanese developer Yukihiro “Matz” Matsumoto. Ruby is open source and is used especially for web development: Platforms such as GitHub, Shopify, Stripe and AirBnB actively use Ruby, through the “Ruby on Rails” web framework written in Ruby the language was learned from 15 Years of buoyancy.

At its core, Ruby is object-oriented, but it also supports other paradigms and concepts such as dynamic Typing, reflections, and automatic garbage collection. Programs written in Ruby are interpreted at runtime, and the advantage is that the code is not particularly complex. This is said to have brought some developers frustrated by more complex languages ​​on board.

(sih)

Relocation complete: The Eclipse Foundation is now officially European

As announced, the Eclipse Foundation has moved its legal headquarters to Europe. To this end, it founded the Eclipse Foundation AISBL (Association internationale sans but lucratif), which is known in German as an international non-profit association (IVoG), a legal form used in Belgium for non-profit organizations. Now the foundation has completed all the formalities for the move.

The managing director of the Eclipse Foundation Mike Milinkovich describes in a blog post that since May 2020 Planned move as a labor-intensive process. The first step has been taken with the official establishment of the AISBL, but further tasks await the members and those who contribute to the open source projects.

Strong in Europe According to Milinkovich, the Eclipse Foundation is the largest open source foundation in Europe in terms of employees, projects, developers and members. Total 170 members and over 900 Contributors come from European countries.

Europe has traditionally been a strong region for the Eclipse Foundation.

(Image: Eclipse Foundation)

Numerous large European and specifically German companies are founding members of the Eclipse Foundation AISBL, including Bosch, Daimler TSS and SAP. In addition, the European Commission considers Milinkovich’s open source initiatives to be strategically important for digital and industrial transformation.

Since the beginning, the Eclipse Foundation has been based in North America, despite its many European members. At the time of the announcement of the move, there was talk of 18 percent. The changes in the events for the community were gradual: the European EclipseCon overtook its formerly big US sister over the years. The latter eventually disappeared from the scene, while the EclipseCon still exists.

From the line of fire of the trade war The European establishment has the advantage for the Eclipse Foundation that it is not under direct US influence. This puts it outside the firing line of the trade war between the USA and China that has escalated in recent years. The members of the foundation include the Chinese company Huawei, which is the focus of the dispute. The move to Europe will secure the multi-faceted growth of the Eclipse Foundation in the long term, which, according to Milinkovich, last year 60 have joined new members.

The move of the Foundation’s corporate governance to Europe entails some organizational and legal changes. The European legal entity will now take control of the names, trademarks and brands. In future, membership fees will also be billed in euros.

However, this does not mean that the Eclipse Foundation is withdrawing from the USA. The Eclipse Foundation Inc. in the US state of Delaware remains in existence under the legal guidelines there. The move also does not affect the daily work that should be distributed across the Canadian and European branches.

For those who work on Eclipse projects, little should change initially. In the FAQ on the move, it says “The short version is: Continue as before. In the short term, it will remain business as usual for commiters and contributors of Eclipse projects.” However, as part of the move, the Foundation set up resources for development with GitLab in Europe, and according to the FAQ, those responsible are welcome to migrate their projects to it at any time.

(Image: Eclipse Foundation)

The Eclipse Foundation was founded 2001 and was previously headquartered in Ottawa, Canada. The history goes back three years further: 900 IBM started together with a consortium of software providers such as Red Hat, SuSE and Borland the Eclipse IT project.

The foundation, which welcomes both companies and individuals, was created with the aim of creating a vendor-independent community. The projects are publicly available under the Eclipse Public License

One of the well-known projects is the Eclipse IDE. In the last few years there has been a strong increase in the enterprise Java environment. The Java EE successor Jakarta EE is under the umbrella of the foundation, and Eclipse MicroProfile is primarily aimed at the development of microservices. The OSGi Alliance recently handed over its work to the Foundation. In addition, the Foundation is traditionally well positioned in the Internet of Things (IoT) sector.

Overall, there are currently 18 Working groups for the individual areas.

More details about the move can be found in the Eclipse blog and the FAQ as well as a presentation on the Eclipse Foundation AISBL in PDF format. In addition, the Foundation has published a white paper on digital transformation in Europe through global open source collaboration.

(rme)

graphic-interface:-vulkan-officially-available-for-apple-platforms

Graphic interface: Vulkan officially available for Apple platforms

The Khronos Group has released an update to the Software Development Kit (SDK) for Vulkan, which officially covers Apple’s platforms. It is based on Vulkan Portability and also covers Apple Silicon. This allows developers to write cross-platform applications on Vulkan instead of implementing them natively for Apple’s own Metal graphic architecture. Vulkan applications also run on Apple Silicon via Universal Binaries.

The Vulkan stands on metal In September the Khronos Group published the Vulkan Portability Extension 1.0, which enables Vulkan SDK calls on other than the natively supported hardware platforms. The main focus was on implementing macOS, iOS and tvOS, since Apple does not allow Vulkan drivers for its own operating systems.

The open source project MoltenVK is a central component of the portability layer. It offers an implementation of Vulkan 1.1 for Apple Metal and thus allows development on Vulkan and execution on Metal. There is also the open-source library gfx-portability, which offers portability via the abstraction layer gfx-hal.

A wide base enables the execution of Vulkan applications on Apple platforms.

(Image: Khronos Group)

Layers of the volcano As part of the updated SDK, LunarG has officially published its Device Simulation Layer (DevSim), which was previously only available for Linux and Windows, for macOS as well. This layer enables cross-platform development on more or less any source and target platform, for example to write a Vulkan application for certain iPhone models on Windows. The respective configurations for the properties of the hardware are stored in JSON files.

In addition, the properties of the hardware on Apple systems can be determined via the Hardware Capability Viewer, the entries for the public database gpuinfo. org created. In addition, the official macOS SDK for Vulkan contains the Vulkaninfo tool for the automatic creation of DevSim configuration files. It also comes with pre-configured configurations for certain Mac, iPad and iPhone devices.

Further details on the updated Vulkan SDK can be found on the Khronos blog. An updated overview of the Vulkan Portability Initiative can be found in a separate presentation in PDF format.

(rme)

jpeg-xl-is-said-to-inherit-jpeg,-png,-gif-and-webp

JPEG XL is said to inherit JPEG, PNG, GIF and Webp

The Joint Photographic Experts Group (JPEG) has fixed the format of its new standard JPEG XL. In addition to the description of the code stream, the Freeze Release format also contains an open source library for reading and writing the new format as well as an encoder and a decoder as reference implementations.

JPEG XL uses a psychovisual model which – analogous to MP3 for audio – allows a stronger compression of the image material with significantly fewer artifacts than with the classic JPEG format. JPEG XL can compress images both lossy and lossless. This means that the format should be just as suitable for photos as it is for graphics which, due to the formation of artifacts on sharp edges, are not saved in lossy formats such as JPEG, but in PNG or other lossless formats. Like GIF, JPEG supports XL animation.

(Image: Jon Sneyers)

Different from the Google format Webp, which can also compress losslessly and lossy and also outperforms JPEG when it comes to the compromise between compression rate and image quality, JPEG XL is not limited to 8 bits per color channel. The new format also masters a progressive mode: the more data in a JPEG-XL file is processed, the better the resolution of the image. This is a particular advantage for responsive (web) applications that require image material in different resolutions depending on the output device.

Conversion without loss of quality Existing image material in the formats JPEG, PNG8, PNG 24 and GIF should be convertible into JPEG-XL files without any loss of quality, whereby the JXL files should be smaller than the original files. Further information on JPEG XL is available from the JPEG Consortium, a (perhaps not completely impartial) comparison with other image formats is provided by Jon Sneyers, co-chair of the JPEG XL group.

With JPEG 2000 the Joint Photographic Experts Group had already made an attempt to establish a successor to JPEG. However, the format did not catch on, mainly because initially no open source libraries were available for it and license fees were feared due to the unclear patent situation. The JPEG XR successor, developed by Microsoft, was just as unsuccessful as the JPEG XT format of the JPEG consortium.

(odi)

RISC-V Powered BeagleV Board Announced Via New Collaboration

(Image credit: Seeed, BeagleBoard, StarFive)

The RISC-V CPU architecture is making progress across a plethora of devices. From soldering irons and watches to desktop computers, the open source RISC-V CPU looks set to make an impact in 2021. Seeed Studio and Beagleboard have today announced an official collaboration with StarFive, a leading RISC-V solutions provider. The focus of this announcement is the Beagle V, an affordable single board computer (SBC) which runs Linux and is powered by a dual core RISC-V.

Image 1 of 2

(Image credit: Seeed, BeagleBoard, StarFive)

Image 2 of 2

(Image credit: Seeed, BeagleBoard, StarFive)

Powering Beagle V is a dual core 64-bit U74 RISC-V CPU clocked at 1.5 GHz and up to 8GB of LPDDR4 RAM. From a CPU perspective Beagle V looks to be slightly faster than a Raspberry Pi 4, but looking at the StarFive JH7100 System on Chip (SoC) we can see a Tensilica-VP6 used in computer vision projects, an Nvidia Deep Learning Accelerator (NVDLA) used for artificial intelligence and finally a Neural Network Engine. All of these point towards a board that looks on paper to be quite the AI / machine learning powerhouse. Best of all Beagle V is a purely open source product, with both the hardware and the software offered under an open source licence.

Processor SiFive U74 RISC-V Dual core with 2MB L2 cache @ 1.5GHz
Vision DSP Tensilica-VP6 for computing vision
NVDLA Engine (configuration 2048 MACs@800MHz )
Neural Network Engine (1024MACs@500MHz)
Memory 8GB LPDDR4 (2 x 4GB LPDDR4 SDRAM)
Video Processing Video Decoder/Encoder(H264/H265) up to 1 channel 4K@60FPS or 8 channel 1080p@30FPS
Dual channels of ISP, each channel support up to 4K@30FPS
2 x MIPI-CSI, 1 x MIPI-DSI
1 x HDMI support up to 1080P@30FPS
Support MIPI-CSI TX for video output after ISP and AI processing
JPEG Encoder/Decoder
Peripherals 4 x USB 3.0 Ports
1 x Gigabit Ethernet
1 x 3.5mm Audio jack
40 Pin GPIO Header (28 x GPIO, I2C, I2S, SPI, UART)
MicroSD card slot for operating system and data storage
1 x Wi-Fi 2.4GHz b/g/n and Bluetooth 4.2
Power with USB Type-C (5V@3A)
1 x Reset button and 1 x Power Button

At its heart, Beagle V is still an SBC and as such it has what we come to think as the “standard complement” of ports. With four USB 3.0 ports, Gigabit Ethernet, Micro SD card and a single HDMI port, capable of 1080P 30fps. A 40 pin GPIO suggests compatibility with Raspberry Pi HATs, and a casual glance at the pinout shows that the standard digital I/O, I2C and SPI seem to be mapped correctly. Dual camera (CSI) and a single display connector (DSI) suggest that Beagle V may be compatible with the official Raspberry Pi camera and display. Wireless connectivity is provided via 2.4 GHz WiFi and Bluetooth 4.2. A shame that 5 GHz WiFi and Bluetooth 5.0 are not present, but we can live without it. Power is provided via a USB-C port, and from the datasheet we can see that Beagle V requires 5V at 2A but a 3A supply would be a real benefit.

Starting from $119 for the 4GB model, or $149 for the 8GB, Beagle V is slightly more expensive than a Raspberry Pi 4 but for RISC-V and AI enthusiasts this is a small price to pay for a RISC-V machine of such power. With an anticipated early access release date of March, with larger availability from September 2021, Beagle V looks to be a worthy contender to the Nvidia Jetson for AI and machine learning projects. It may not be a direct competitor to the Raspberry Pi, but it will be interesting to put the two against one another to see both of their pros and cons.