Threema becomes open source and thus reveals the source code. “To celebrate, we are lowering the app price by 28. December by 50%”, the Messenger writes -Provider in the announcement. Threema already had documentation of the cryptographic process, now the complete inspection and release of the code follows.
According to Threema, the code is subject to the AGPLv3 license and is therefore generally freely available to users. For the time being, however, it can only be compiled reproducibly under Android. In this way you can ensure that Threema’s Android apps actually use the published code. Apple’s guidelines didn’t make it easy to offer reproducible builds, but they are working on them. The first announcement that Threema should become Open Source was made at the end of the summer, including the announcement of the entry of the investment company Afinum.
Security instead of data collection Since the messenger is still chargeable, it must of course not be simply set up again yourself. A license check prevents new Threema IDs from being created with the self-compiled apps. If you want to use a self-built app, you have to import a backup into the app with an existing ID that was previously created with a purchased Threema app. Only thanks to paying users is it possible to ensure privacy and security of the messenger without data collection or advertising, says Threema.
Threema uses the open source NaCl library for encryption. The correct application of the encryption could already be checked by manually checking the encrypted messages. The messenger uses two ways of encryption; The end-to-end encryption between the communication participants themselves and an additional encryption between the app and the server so that it cannot be found out by tapping network packets who is communicating with whom.
Threema is also working on offering a multi-device function. Since the own security requirements made this very difficult, the development is still taking. The provider has at least already stated that they want to try to use a “mediator server” that takes over the coordination. Several keys should be derived from the user ID in order to prevent the server from knowing which ID belongs to which device.
The Tape Archiver is the mother of all packers and standard on the command line. An introduction to the powerful and at the same time inconspicuous tool.
(Image: Michael Plura)
Introduction to tar: Packing and unpacking data cleverly on the command line First steps with tar Search archives Unpack archives Compress archives Deletion, backups and a packer alternative Software and extensive documents of any kind are usually available for download as a compressed package. A click in the web browser is all it takes to move the package to the download folder. Double-click in the file manager or right-click and “unzip here” to access its content. Even a packed zip archive can be created without much effort: Mark all the required files and directories, right-click again and “compress” in most cases already leads to the desired result.
However, these simple methods fail when it comes to special use cases or even packing in scripts – be it for automatic backups carried out regularly via cron or the provision of generated data on a web server. The classic Unix packer tar is available for this in the GNU / Linux or FreeBSD terminal. This has a big advantage: it is available for almost every operating system and it behaves almost identically everywhere. tar itself cannot compress, which sounds impractical at first, but has advantages because tar works with all common external packers. So tar is first and foremost a sophisticated tool for combining and packing files and directories. Via simple options like z or j tar then uses the compressor of choice such as gzip, bzip2, xz or almost any other data squeezer to evaporate either quickly and moderately or slowly and extremely.
This article shows the basic handling of tar and the common compressors in the terminal under the BSDs and GNU / Linux. macOS behaves like FreeBSD, OpenBSD, and NetBSD thanks to the FreeBSD userland. Under Windows 10 there are different implementations like bsdtar or gzip. We also present tricks and tips that simplify daily work with archives.
Access to all contents of heise + exclusive tests, advice & background information : independent, critically founded c’t, iX, Technology Review , Mac & i, Make, c’t Read photography directly in the browser register once – read on all devices – can be canceled monthly first Month free, then monthly 9, 95 € Weekly newsletter with personal reading recommendations from the editor-in-chief Start FREE month Start your FREE month now heise + already subscribed?
Sign in and read Register now and read the article immediately More information about heise + Introduction to tar: Packing and unpacking data cleverly on the command line First steps with tar Search archives Unpack archives Compress archives Deletion, backups and a packer alternative
Windows Server includes a role that installs Windows over the network. This avoids sneaker administration and saves installation sticks / DVDs.
(Image: Albert Hulm)
The times when a silver disk, be it CD or DVD, was the common installation medium for Windows installations are long gone – today it is mostly USB storage media that the admin plugs from computer to computer. A task that really annoys at the latest when you need to quickly install a delivery of fresh PCs for an entire department or quickly set up notebooks for all new home office employees. An alternative is to install over the network via PXE boot – once set up, the Windows Deployment Services (WDS) of Windows Server save a lot of work.
A machine with Windows Server is required – in Server 2000 is used for this article, but this feature has not changed for years. The server can usually do other work as well, for example as a WSUS server distributing updates to clients. Only rarely will a machine be busy delivering installations around the clock and thus be fully utilized. The machine should definitely have a gigabit connection to the local network; it is better to have several network cards as a so-called NIC team or one 10 – GBit card. After all, the server should deliver Windows packages of several gigabytes in size quickly, sometimes to several computers at the same time.
Who really has many clients (maybe even hundreds) have to install in parallel, but don’t have to worry about the network card becoming the bottleneck. The Windows provisioning services can also deliver the data via multicast over the network and thus install almost any number of computers in parallel without adding up the amount of data sent. Since such a multicast environment also requires adjustments to the network, it is not part of this article. In this article, the WDS server is set up so that it can deliver the first Windows images.
Access to all contents of heise + exclusive tests , Advice & background: independent, critically sound c’t, iX, Technology Review, Mac & i, Make, c’t read photography directly in the browser register once – read on all devices – can be canceled monthly first month free, thereafter monthly 9, 95 € Weekly newsletter with personal reading recommendations from the editor-in-chief Start FREE month Start your FREE month now heise + already subscribed?
Sign in and read Register now and read the article immediately More information about heise +
Page 1: Tenda Nova MW 12 in the test: Mesh system at an attractive price
In the mesh network hardware sector, Tenda is unlikely to be a household name for many, even though the company has been active as such for some time. In the past few months there has been more attention in this country – especially with regard to cheap, yet fast mesh hardware. One wants to have left the established providers behind. We have the Tenda Nova MW 11 looked more closely as a beginner’s solution.
Tenda is in the year 2012 partnered with Broadcom and has become a key partner in China. The company has been using Broadcom’s chipsets since then and wants to be able to guarantee a certain degree of compatibility and performance. In addition to the mesh products, routers, switches and much more in the area of network hardware are also available.
The Nova MW 12 with its three stations covers an area of 500 m² from. The set consisting of three stations can be extended to 12 Mesh node can be expanded and then covers up to 2. 000 m² from. With the help of the Tenda WiFi app, the stations can be managed and, if required, a guest network can be set up or a child lock can be activated.
Now we come to the Nova MW 12 or to the technical data of the mesh system:
Tenda Nova MW 12
Model
Tenda Nova MW 12
Street price
from 220 Euro
Homepage
www.tendacn.com/de
Technical specifications
Internet
Dynamic IP, Static I. P
LAN
2x Gigabit Ethernet 1x WAN
Telephony
–
WIRELESS INTERNET ACCESS
IEEE 802. 11 a / b / g / n / ac tri-band WiFi 1x 2.4 GHz: up to 300 Mbit / s 1x 5 GHz: up to 867 Mbit / s 1x 5 GHz: up to 867 Mbit / s (backhaul for the mesh)
Connections
2x Gigabit Ethernet (10 / 100 / 1000 Base-T) 1x Gigabit Ethernet (10 / 100 / 1000 Base- T) WAN for connection to cable / DSL / fiber optic modem or network
Dimensions
100 x 100 x 100 mm
Scope of delivery
3x mesh 12 stations three power supplies Ethernet cable Installation Guide
The starter set consists of three meshes 11 – stations that are all identical. Depending on which node is integrated first, this is the gateway or the central element of the mesh network and is also connected to the rest of the network or the router via Ethernet. The system automatically detects what type of Internet connection it is – for example, whether an IP is assigned dynamically or if this is done manually.
At this mesh node, Two more Gigabit Ethernet connections are available, as the third is used as a WAN connection. Mesh nodes that are only integrated via WLAN therefore have three gigabit connections available. The WLAN is according to 802. 11 ac compatible, but can of course also integrate clients with older standards. The 2.4 GHz network is 300 Mbit / s possible, for 5 GHz networks it should be 802 Mbit / s. Of course, we’ve seen faster network hardware here – at least on paper. WPA-PSK and WPA2-PSK are supported in terms of encryption.
Each station has an SoC with 900 MHz and 220 MB DDR3- RAM installed. This is to enable the network up to 100 Can process devices simultaneously in the network. In addition, the computing power is apparently necessary to control the MU-MIMO, QoS, and Beamforming +. The delivery includes the three mesh 18 stations, each with a power supply unit, an Ethernet cable and a brief instruction that can be helpful during commissioning.
The Mesh 12 – Stations are designed as cubes and have dimensions of exactly 100 x 100 x 100 mm. The mesh hardware is therefore quite compact, because we have also seen solutions that cannot be easily placed on a shelf or the like. Small and inconspicuous, this is how the design of the Nova MW 12 describe. The housings are also simply kept white. A Nova lettering and a light gray, grid-shaped imprint are everything that can be observed from the outside.
At one corner of each mesh station there is an LED that lights up green when everything is OK, while the integration flashes green and then shows with a yellow or red LED how good the reception quality is of the individual mesh nodes.
As a cube, the mesh nodes can in principle be set up in any orientation, but of course there is a bottom, because access to the connections must be granted somewhere. Here you can also see four padded feet, which allow the mesh station to set up non-slip and without scratches. Wall mounting is theoretically possible thanks to the two holes. The otherwise closed case offers some openings on the underside so that the hardware inside can also be cooled a bit.
The connections of each mesh node of the Tenda Nova MW are located on the underside 12. The power supply connection can be seen on the right. Next to it are the three Gigabit Ethernet ports, of which the one on the right can also serve as a WAN port. Wired devices can also be integrated into the network locally and at the installation site of the mesh nodes via the Ethernet connections.
Commissioning of the Tenda Nova MW 12
Mesh networks usually promise a simple set-up that the layman should be able to do without any problems . Often, however, problems arise that can have various causes. Tenda also promises a hassle-free setup that the app should help with. This is available for Google’s Android and Apple’s iOS.
In a first step, one of the mesh stations must be set up and supplied with the network and power. As soon as the LED lights up green, the user can connect to the factory-set WLAN. The necessary access data are on the underside of each mesh station. Then the setup takes place step by step.
So the user chooses his own WLAN name and assigns his own password. The other mesh stations can then be integrated. For this purpose, these should first be set up in the immediate vicinity. When powered, they will automatically search for the primary node or gateway. The connection is established after 1-2 minutes and is confirmed with a green LED or in the app. The mesh stations can then be set up in the desired positions. A green LED then indicates that the connection is fast enough. A yellow LED indicates that the connection quality is not good and it would be better to put the nodes closer together. A red LED indicates that the node is not involved at all.
If you have given the nodes correct names, the app gives you a nice overview of your own network structure. Numbers on the nodes show how many devices are currently connected to it. A display for the downstream and upstream provides information about the load on the WAN connection. Additional detailed information can be called up for each individual node. The LED display can also be deactivated here if it is too bright or disturbs for other reasons. Firmware updates are also possible at this point.
The WLAN settings (SSID and password) can be changed via the settings. A guest WLAN can also be set up here, if you don’t want to invite every user to your primary network. Applications can be prioritized and bandwidth limits set using QoS services. Parental controls allow groups with end devices to be defined that are only allowed to use the Internet at certain times. Some services are also blocked here.
Of course there are also rudimentary functions such as port forwarding, a WPS assistant or the option of entering your own DNS server, the Tenda Nova MW 12 but does not offer the extensive possibilities of some other network hardware. This is also not absolutely necessary in this case, because not only should the setup be simple, but also no unnecessary functions should interfere with operation. Very few users in the private network go into the settings in depth. Set up, connect, start surfing – this is the premise with which most users of such mesh networks approach the topic and usually nothing more is needed.
< > Tenda Nova MW 12 in the test: Mesh system at an attractive price WLAN benchmarks
Ampere Computing is moving purposefully in the direction of a CPU alternative for data center use. The Ampere Altra is based on the Neoverse-N1 design from ARM and offers 80 ARM v8.2 + cores in one socket. Up to 80 cores in a power budget of 250 W should be offered to established manufacturers such as Intel and AMD compete. As Altra Max 2021 even variants with 80 cores should follow.
So far, however, there have only been many announcements and few tangible ones, even if Ampere naturally presented its own benchmarks and wanted to emphasize, among other things, the advantages of a stable clock over dynamic boost clock rates. Now Anandtech presents a first detailed and independent test of two Altra Q 80 – 33 (80 cores with 3.3 GHz) in a dual socket system (Mount Jade). The processor in the test system still includes 256 GB of RAM per socket – in eight DIMMs per socket in order to fully utilize the 8-channel memory interface.
The opponents of the two Altra Q 80 – 33 were two AMD EPYC 7742 each with 64 cores with a TDP of 225 W and two Intel Xeon Platinum 8280 each with 28 Cores at 205 W.
The processors and the system leave a different picture in the benchmarks. In the single-threaded test using Spec 2017 integer, the Altra Q intersects 80 – 33 very good – is on a level with the Xeon Platinum 8280 and ahead of the EPYC 7742. In the floating point area, however, the established processors are stronger. It looks much better if the Altra Q 80 – 33 can use all of its 80 cores.
The ARM processor is really good in the LLVM compiling and the NAMD performance. This is where the old Q 80 – 33 in 1S and 2S configuration on par with one or two EPYC 7742. In the chosen benchmark in 1S configuration, these two processors are even faster than two Xeon Platinum 8280.
Across all tests it can be said that the Altra Q 80 – 33 not only quickly, but also efficiently are. Although the TDP provides for a power budget of 250 W, the system does not use this for the 3.3 GHz clock rate. This must also be taken into account when assessing performance. Price also plays a role. Ampere calls for the Altra Q 80 – 33 a list price of 4. 050 US dollars, an EPYC 7742 costs around 5 in this country 200 Euro and a Xeon Platinum 8280 at least 8. 600 Euro, albeit one should note here that these prices in the free trade have very little to do with what server providers are offered at reduced prices.
Whether and where amps can be compared with the Altra processors in the data center Segment will be able to establish itself remains to be seen. Hyperscalers like Amazon have already developed their own processors. AMD is nibbling on Intel’s market share, but the chip giant is supposed to strike back 2021 / 22 . First, however, AMD will launch the 3rd generation EPYC processors based on the Zen 3 architecture.
currently working to shift its Azure servers from Nvidia’s processors to its own in-house chips, it’s clear that major tech industry players are keen to bring more and more of their production lines as close to home as possible. And according to a new report from
Bloomberg
, Microsoft’s gearing up to be the next company to transition to in-house chips. At least for some of its products.
“Microsoft Corp. is working on in-house processor designs for use in server computers that run the company’s cloud services,” Bloomberg writes. “It’s also exploring using another chip that would power some of its Surface line of personal computers.”
Like Amazon’s Graviton 2 chips, these processors will be Arm-based, which makes sense.
Windows on Arm
already exists, so Microsoft’s familiar with it. But since Nvidia owns Arm, the goal here is probably to reduce reliance on Intel rather than to make the company’s computing perfectly bespoke.
This also isn’t the first time Microsoft’s explored using self-developed chips, though they’ve mostly been made in collaboration with other companies until now. Microsoft already uses an Arm-based chip that it’s co-developed with Qualcomm, the SQ2, in its Surface Pro X tablet. The Surface Laptop 3 also uses a custom Ryzen chip, and the company’s long used custom processors co-developed with other manufacturers in its game consoles, like the Xenon chip it made with IBM.
But using custom chips in its servers is a first for the company.
“Because silicon is a foundational building block for technology, we’re continuing to invest in our own capabilities in areas like design, manufacturing and tools,” Microsoft Communications Chief Frank Shaw told The Verge, “while also fostering and strengthening partnerships with a wide range of chip providers.”
Which also points to another potential reason for the move: as companies recover from the pandemic, they’re facing pressure to become more self-reliant.
Still, according to Bloomberg, this move has been in the works for a while, as Microsoft’s spent the last few years recruiting more and more
processor engineers
.
Intel’s long been able to rely on its business sales when its consumer chips fall behind, but according to Bloomberg, Intel’s stock is now down 21% this year as more and more companies distance themselves from it.
Microsoft is designing its own ARM-based processors for servers and possibly a future Surface device, according to Bloomberg News. The processors will be used in servers for Microsoft’s Azure cloud services and be based around ARM designs, according to the report. Microsoft is also reportedly “exploring” using another chip for some of its Surface devices, but it’s not clear if this will progress into a final product.
Microsoft currently uses Intel-based processors for the majority of its Azure cloud services, and most of the company’s Surface lineup run on Intel chips, too. Microsoft has worked with AMD and Qualcomm for custom chips for its Surface Laptop 3 and Surface Pro X devices, showing a willingness to move away from just Intel.
Microsoft’s custom chips for Surface devices.Photo by Amelia Holowaty Krales / The Verge
Microsoft co-engineered an ARM-based SQ1 processor for the Surface Pro X last year and followed this up with an SQ2 variant a couple of months ago. AMD also worked with Microsoft to create a custom version of its Ryzen processor for the Surface Laptop 3.
A move to ARM on the server side is certainly more significant, particularly for Intel. Apple has already signaled its move away from Intel chips for its Mac products, with its own M1 silicon based on ARM designs. Intel’s Xeon server chips currently rule the server market, and AMD has already been chipping away at this lucrative market with its own EPYC processors.
We’ve reached out to Microsoft to comment on the report, and we’ll update you accordingly.
Collabora developer Alexandros Frantzis introduces an experimental Wayland driver for the Windows compatibility layer Wine. For the graphical display of apps in Linux, Wine usually interacts as a client with the X server. However, Linux desktop environments increasingly rely on the Wayland protocol instead of X 11. For this reason, Wine previously had to use Xwayland as an intermediate step in order to pass content through to the Wayland compositor. The Wayland driver for Wine now presented makes this step superfluous
The Wayland driver for Wine is at an early stage. On the Collabora website, Frantzis gives an overview of the development status. Up until now, GDI and OpenGL / DirectX applications have run on one display. Users can set the windows to full screen, resize and maximize, but not minimize them. In addition to mouse inputs, keyboard inputs are also possible, but so far only in the QWERTY layout.
Well-received These first successes prompted Frantzis to present the driver on the “wine-devel” mailing list and to ask for feedback and assessments. The Wine developers initially welcomed the contribution and are now advising whether and how the driver can be integrated into the Wine code.
A short video on the Collabora Youtube channel shows some programs that already work with the Wayland driver, including Windows Explorer, Gimp, Firefox and the game “The Batttle for Wesnoth”. Frantzis announces the intention to cooperate with another open source project that works on the compatibility of Wine with the graphic API Vulkan in Wayland mode.
Interested parties can find more information in the announcement on the Collabora website. The discussion about the Wayland driver can be followed on the Wine mailing list.
Home/Component/Cooling/G.Skill introduces Enki Series AIO CPU coolers featuring ARGB
João Silva 35 mins ago Cooling, Featured Tech News, Tech News
G.Skill is a very prominent name in the memory market, but it looks like the company wants to diversify its portfolio of PC components and expand into the CPU cooler market. The new Enki AIO CPU cooler series is a custom-designed solution available in 240mm, 280mm, and 360mm radiator sizes, featuring ARGB lighting on the pump and a sleek black design that it’s easy to fit in most systems.
Designed to offer high-cooling performance, the Enki AIO series eliminates “heat-transfer bottlenecks” while effectively transferring the heat from the CPU. This is achieved by careful selection of the cooling solution components, which include low thermal resistance thermal paste, custom-tuned convex copper cold plate, high-density radiator widened, low-evaporation coolant tubing, and 9-blade hydro-bearing cooling fans. Additionally, the Enki CPU coolers also feature ARGB lighting and an easy-to-mount system for a trouble-free installation process.
The custom convex cold plate reduces the total travel distance of heat from the CPU, which goes through the thermal paste before reaching it. Once it gets to the cold plate, the heat is then transferred to the stair-shaped high-density micro-fins that it will then pass it on to the coolant. The coolant travels through high flow-rate coolant tubes with 8mm inner diameter, made of thick rubber to prevent evaporation and braided with nylon for increased durability. The radiator features high-density piping to increase the surface area and efficiently dissipate the heat from the coolant into the radiator. The 240mm and 360mm radiators feature 3-4 more radiator pipes than common radiators, while the 280mm one come with 6x more pipes.
All Enki AIO CPU coolers come equipped with 9-blade hydro-bearing PWM cooling fans designed to produce high static pressure. The accessory pack included with the Enki AIO coolers include a tube of server-grade thermal paste, which stimulates a “quick heat transfer from the CPU to the copper cold plate”.
G.Skill Enki AIO series will be available in 2021, backed by a 5-year warranty.
KitGuru says: Are you looking for a new CPU cooler for your system? If you were to choose an Enki AIO, which size would it be?
Become a Patron!
Check Also
The 2020 Game Awards nearly doubles viewership to 83 million livestreams
Each year, following the conclusion of The Game Awards, the show’s creator shares details on …
Microsoft and its partners have responded to a malware operation known as the SolarWinds attack, including the so-called with sinkhole tactics.
FireEye recently unveiled an extensive offensive operation in which attackers suspected of being state-run attacked a Trojan dubbed Sunburst (FireEye) and Solorigate (Microsoft) into the popular SolarWinds Orion enterprise software. Behind Sunburst is believed to be a group of hackers known as the APT 29 and Cozy Bear, suspected of being linked to the Russian intelligence service. The attack itself began as early as last spring, although it has only just survived.
Now Microsoft has outlined what it will do to counter the attack. The company said it manually removed the digital certificates used by the Trojan-infected files as soon as the attack came to light. In practice, this means that no Windows configuration agrees to run these files because it sees them as untrusted.
In addition, Microsoft updated Windows Defender to detect and alert users of a used Trojan. On Wednesday, however, the company decided to change the standard functionality of Windows Defender for Solorigate from mere warning to automatic quarantine. Automatically quarantining infected files is effective, but at the same time it can also cause a system crash, for example.
In addition, Microsoft and its partners have isolated the so-called avsvmcloud.com site used by the attackers. with sinkhole tactics. Sinkhole refers to the way in which a server is isolated by means of DNS servers from the rest of the Internet to direct traffic destined for them elsewhere. As a result, the malware loses the ability to communicate with its host server and is left without instructions
Yesterday in Milan and province an earthquake of magnitude 3.9 was warned that the INGV or the National Institute of Geophysics and Volcanology immediately reported. Google, however, has done the same but not using the classic seismographs but through the motion sensors of smartphones. This is how experimental technology works.
by Bruno Mucciarelli published 18 December 2020 , at 00: 21 in the channel Telephony Google Android
At 16.59 of yesterday, 17 December 2020, a earthquake was felt by many inhabitants of the Milan and surrounding areas . A roar that initially was not actually considered as an earthquake but which was then immediately reported by the INGV or the National Institute of Geophysics and Volcanology which mentioned an earthquake between the 3.8 and 4.4 degree of magnitude which later revealed itself precisely to 3.9 with epicenter near Pero. Clearly, not only the citizens but also the specialized seismographs of the Institute, but not only, were able to detect the shock.
In fact, news of this earthquake in the province of Milan concerns the report by the Google search engine which in this case received the signal not from the seismographs but from the movement sensors present in the Android smartphones. Yes, because in this case Google has been using the motion sensors of users’ smartphones for some time to understand if there may actually be anomalous movements due to the earthquake.
Google ” feels ” the earthquake via smartphones
The Google system at the moment is clearly experimental . The Mountain View company has well thought out to exploit the motion sensors or the accelerometers present in all Android smartphones to understand whether during the day or night there could be such tremors as to associate them with an earthquake. The company had already announced it a few months ago: the system underlying the technology would not have involved anything in the smartphones of users who would not have had to install any tools or applications but only would have had to use the smartphone as they do every day.
How does this new technology work? The system relies on integrated accelerometers as mentioned in all smartphones to detect signals of an earthquake. If the phone detects an earthquake, it sends a signal to a Google server, along with information on the location of the device. The server combines information received from multiple Android phones to figure out if an earthquake is indeed taking place. In case you actually know a positive feedback on the problem, here is that the system automatically sends alert notices to Android devices, so that the people can find shelter or make themselves safe .
Interestingly, although a normal accelerometer is not obviously as precise as a seismograph, hundreds of thousands of accelerometers distributed in the space of a few hundred square kilometers can still serve as an alarm bell for a rudimentary but quick-to-use anti-seismic system. This is what happened yesterday in the province of Milan by searching for Earthquake on Google: users in the affected area were given a warning message that advised to consult the official sources for more information on the subject . Not only because a question was also asked if the earthquake had actually been felt in that area.
The rush to the world’s largest chip order manufacturer TSMC is said to be so high that the company wants to cancel the previously granted discounts for bulk buyers from 2021. This is reported by the Taiwanese news agency Central News Agency, according to which TSMC is said to have granted a three percent discount so far. That doesn’t sound like a lot at first, but with tens of thousands of wafers per month, the savings quickly add up.
At TSMC, new generations of production are currently the first available in high capacities. Among other things, AMD ensures high utilization of Ryzen 3000 processors, Epyc in the 7 nanometer generation , Radeon-RX – 6000 – graphics cards and console systems-on-chips for the Playstation 5 and Xbox Series X. Apple takes TSMC the first large quantities 5- nm wafer for the A 14 of the iPhone 12 and the M1 of the ARM Macs. Intel also wants to have TSMC manufactured from 2021. Means: the chip order maker doesn’t have to worry about finding customers at the moment.
From 2021 noticeable When asked, TSMC explained that they did not comment on prices in their own production. If the Central News Agency is right, higher production costs should be (slightly) noticeable in the business figures of manufacturers, more precisely in the margin, within a year at the latest. Initially, the contracts concluded so far continue to run with fixed purchase quantities and prices, which are regularly negotiated between chip order manufacturers and chip manufacturers. All graphics chips from the GA 102 roll off the production line at the South Koreans, which means the complete GeForce series RTX 3000 includes. Only the server accelerator A 100 with GA 100 – GPU lets Nvidia produce with 7 nm technology at TSMC. Rumors are said to have tried Nvidia with Samsung as a second manufacturer to put price pressure on TSMC.
When AMD unveiled its Ryzen Threadripper Pro 3000WX-series processors and WRX80 chipset in July, the company said that the platform will be available only from select workstation suppliers and there were no plans to bring it to do-it-yourself (DIY) enthusiasts. Officially, AMD’s stance has not changed and Threadripper Pro are CPUs for professional workstations, but Gigabyte has developed a branded motherboard for Threadripper Pro processors that will likely have broad availability. A promotional picture of Gigabyte’s WRX80-SU8 motherboard was published by VideoCardz website, which may indicate that the product is either about to be launched or is in final stages of its development.
Gigabyte’s platform is called the WRX80 SU8 IPMI Server Motherboard and it features a server form-factor (most likely EATX) and carries a baseboard management controller (BMC). Meanwhile, the platform supports all AMD Ryzen Threadripper Pro 3000WX-series processors in sWRX8 form-factor up to the 64-core Pro 3995WX. The platform has eight memory slots to use Threadripper Pro’s main feature: eight-channel memory support.
(Image credit: VideoCardz)
The motherboard has eight PCIe 4.0 x16 slots (not all of them support 16 lanes at all times though), at least two M.2-2280 slots for SSDs and SAS/SATA ports. Gigabyte’s WRX80 SU8 also carries two 10GbE connectors, two GbE ports for management, several USB Type-A connectors, at least one display output (for BMC), and a 7.1-channel audio subsystem, an indicator that this platform is intended both for servers and for workstations.
AMD’s Ryzen Threadripper Pro 3000WX-series CPUs do not support overclocking, which makes them somewhat less appealing to enthusiasts. But at the same time support for up to 2TB of eight-channel DDR4-3200 memory make the processors very attractive to professionals.
At present, AMD only sells Ryzen Threadripper Pro 3000WX-series CPUs to select makers of workstations. Given the combination of server and workstation features the Gigabyte WRX80 SU8 motherboard has, it may address workstations by boutique PC shops as well as smaller server builders. If this is the case, AMD is about to start selling its Ryzen Threadripper Pro not only to select makers of professional workstations, but also to distributors and ultimately end users. Unless, of course, the platform was designed for a specific customer with certain requirements and meant exclusively to one client that has access to AMD’s exotic Threadripper Pro chips.
Network storage: four NAS empty housings for self-assembly in the test Practical aspects Connections Performance Individual meetings Conclusion Test table Article in c’t 1 / 2021 read The data storage in the LAN is not only useful for centrally stored files and PC backups: it can also act as a media server for music and videos Deliver to the smart TV. If you want to break away from Google, the NAS with software extensions (apps from the manufacturer’s repositories) can also be used as a cloud server for contacts, appointments and notes that can be accessed from the Internet. The central backup of photos and videos from family smartphones via app is no longer a bonus.
You will find an overview of the functions of common NAS operating systems in another article. This is about the hardware equipment and performance of inexpensive NAS empty housings with two storage spaces for self-installation of hard drives or SSDs. We brought four of them into the laboratory: Netgear ReadyNAS 212, QNAP TS – 220, Synology DS 220 j and Western Digital (WD) My Cloud EX2 Ultra. The TerraMaster F2 – 210 is also in the 150 – Euro price range. An Asustor AS 5202 T for around 300 Euro gives an outlook on the added value that a slightly larger investment will buy.
We equipped each of the empty housings with two NAS hard drives Seagate ST 4000 VN 008 of 4 terabytes each. This is the capacity you get most of the bang for your buck right now. In the RAID-1 configuration (mirroring, protects against hard disk defects), there is also a net capacity of 4 terabytes. The devices also support configurations for twice the memory size, but you should avoid that.
Access to all contents of heise + exclusive tests, advice & background: independent , critically founded c’t, iX, Technology Review, Mac & i, Make, c’t read photography directly in the browser register once – read on all devices – can be canceled monthly first month free, then monthly 9 , 95 € Weekly newsletter with personal reading recommendations from the editor-in-chief Start FREE month Start the FREE month now Heise + already subscribed?
Sign in and read Register now and read the article immediately More information about heise +
Diablo Immortal didn’t make a great first impression. It was announced at Blizzcon in 2018 when most fans were expecting the next mainline Diablo title, not a free-to-play spinoff for smartphones. Since then, Blizzard has been very quiet about Immortal (though Diablo IV was eventually announced).
That’s starting to change. Ahead of a technical alpha that’s set to kick off soon, I was able to play through the early stages of the game. And while I can’t speak to every aspect, one thing was immediately clear: this feels like a proper Diablo game.
A mobile version of Diablo makes a lot of sense. There are already countless clones and Diablo-inspired titles on smartphones, while the Nintendo Switch version of Diablo IIIproved just how fun the series is to play on the go. There’s just something about the hack and slash gameplay that makes it a great fit for a handheld experience. It’s the kind of game that can suck you in for hours and yet, paradoxically, is also something that doesn’t always require your full attention. I spent part of my time previewing Immortal while I was also in a Zoom meeting. (Don’t tell Nilay.)
Immortal takes place between the events of the last two Diablo games, and it’s surprisingly story-heavy for a mobile title. There are brief cutscenes and lots of voice acting, though the actual narrative is typical Diablo fare: you’re dropped into a small village overrun with the undead and cultists, while Deckard Cain tells you to collect magical shards that can stop some world-ending calamity — you know, regular Diablo stuff.
More important is how the game plays. Immortal features entirely touch-based controls (there’s no gamepad support right now) and they work surprisingly well. The left side of the screen is a virtual joystick so you can move around, while the right features all of your powers. It works just like a regular Diablo title for the most part. You have a main attack and will quickly unlock special abilities, each with their own cooldowns. You just tap on the abilities to use them, and some require you to actually aim. The only big addition is a new class-specific “ultimate” ability, which is reminiscent of Overwatch; it’s an incredibly powerful attack, but one that builds up slowly, forcing you to use it sparingly.
The early-game dungeons I played through were all relatively small, which made them great for mobile, as I could squeeze them in when I just had 20 minutes or so to play. They were also relatively straightforward: I didn’t find myself doing much exploration in search of loot or other secrets. That said, I really enjoyed the boss fights and raids, which do a good job of increasing the scale with massive monsters and big waves of enemies.
Really, the fact that the game feels so familiar is arguably the best-case scenario. Blizzard seems to have done an admirable job of transporting the spirit of its loot-heavy roleplaying game to a smartphone. But there are also a few elements of the experience — ones that will likely be key to its longevity — that I couldn’t test out in this preview. For one thing, Immortal is billed as a massively multiplayer game, one where tens of thousands of players exist simultaneously on one server. I saw a few players running around dungeons and graveyards while I played, but I couldn’t test out any of the actual social features.
Perhaps the biggest question mark, though, is how Blizzard will monetize the game. The developer says all character classes and story content will be entirely free, and that gear isn’t something you can buy. Instead, you have to collect it out in the world by defeating monsters or completing quests.
But there’s still plenty to buy, including stones that let you reforge an item, “crests” that add better rewards when you complete a randomized dungeon, and a battle pass that Blizzard says will refresh monthly. There’s even a player-to-player marketplace — though the developer is quick to say it’s nothing like the infamous auction house from Diablo III.
Here’s how Blizzard explains it:
Before we get into the details let’s be clear — this is not the Diablo III real-money auction house. Player exchange has been a part of Diablo’s history, and we’ve learned many lessons–both good and bad–along the way. In Diablo Immortal we want players to have the ability to participate in a fair, healthy, player-driven economy that doesn’t circumvent gameplay.
The market connects anonymous buyers and sellers, with no ability to cash-out. There are limits to what can and cannot be put on the market. While it will be a possible source of certain materials, supplementary items, and legendary gems, it is not a place to acquire gear. It’s also a way for players who wish to remain free-to-play to exchange items they find for additional in-game currency, beyond what they’re able to earn through play.
Those are some important question marks, though it’s likely it’ll still be a while before we have a complete picture of what Diablo Immortal is. Blizzard is opening the game up to a small amount of Android users in Australia as part of this alpha, but there’s no word on when it will be available more broadly, as the game still doesn’t have a release date.
We use cookies on our website to give you the most relevant experience. By clicking “Accept”, you consent to the use of ALL the cookies.
This website uses cookies to improve your experience while you navigate through the website. Out of these, the cookies that are categorized as necessary are stored on your browser as they are essential for the working of basic functionalities of the website. We also use third-party cookies that help us analyze and understand how you use this website. These cookies will be stored in your browser only with your consent. You also have the option to opt-out of these cookies. But opting out of some of these cookies may affect your browsing experience.
Necessary cookies are absolutely essential for the website to function properly. This category only includes cookies that ensures basic functionalities and security features of the website. These cookies do not store any personal information.
Any cookies that may not be particularly necessary for the website to function and is used specifically to collect user personal data via analytics, ads, other embedded contents are termed as non-necessary cookies. It is mandatory to procure user consent prior to running these cookies on your website.
