Intel today announced AWS customers can access the latest 3rd Gen Intel Xeon Scalable processors via the new Amazon Elastic Compute Cloud (Amazon EC2) M6i instances. Optimized for high-performance, general-purpose compute, the latest Intel-powered Amazon EC2 instances provide customers increased flexibility and more choices when running their Intel-powered infrastructure within the AWS cloud. Today's news is a further continuation of Intel and AWS' close collaboration, giving customers scalable compute instances in the cloud for almost 15 years.

"Our latest 3rd Gen Intel Xeon Scalable processors are our highest performance data center CPU and provide AWS customers an excellent platform to run their most critical business applications. We look forward to continuing our long-term collaboration with AWS to deploy industry-leading technologies within AWS' cloud infrastructure." -Sandra Rivera, Intel executive vice president and general manager, Datacenter and AI Group.

NVIDIA at Bilibili World 2021 showcased some custom RTX 3080 graphics cards - customizations which it commissioned to known modders. In total, three custom graphics cards were shown in a variety of colors, even though all of them shared the same base shroud design. The modders then played with illustrations and color coordination, which resulted in one orange RTX 3080 (featuring the Chengdu Hunters Overwatch clan insignia) and two turquoise/petrol graphics cards with differentiated accents (black and turquoise for the card inspired on Chinese role-playing game The Legend of Sword and Fairy, and white and petrol for the card celebrating the 9th anniversary of the Luo Tianyi anime series).

These graphics cards likely won't be put to sale in any way by NVIDIA; however, they could eventually decide to give them away as a prize for any sort of online competition, much like they did with the gorgeous, Cyberpunk 2077 interpretation of the RTX 2080 Ti some time ago - which the winner eventually sold at auction for $5,200 (at a time where graphics card pricing was actually saner). One could argue that NVIDIA should have just sent these towards the retail channel, but at least we got to see three RTX 3080's at the same time.

As global market changes spur demand for augmented reality (AR) technology, there is an increasing need for innovations that combine the best technologies in CPU, GPU and machine learning into a single SOC (system on chip) to allow for the creation of the most demanding AR experiences while maintaining power efficiency.

Magic Leap today announced it is partnering with AMD on an AR technology solution that includes a semi-custom SOC to enable enterprise users to re-imagine and transform how virtual content and information is visualized and merged with real-world environments.

Video streaming is looking a bit like magic. The uploader sends a video to one platform in one resolution and encoding format, while the viewer requests a video in a specific resolution and encoding format used by the device the video is streamed on. YouTube knows this best, as it represents the world's largest video platform with over 2 billion users visiting the platform each month. That takes a massive load on the server infrastructure over at Google's data centers that host the service. There is about 500 hours worth of video content uploaded to the platform every minute, and regular hardware isn't being enough anymore to handle everything.

That is why YouTube has developed custom chips, ASICs, that are called VCUs or Video (trans)Coding Units. In Google data centers, there is a large problem with transcoding. Each video needs to adapt to the streaming platform and desired specifications, and doing that on regular hardware is a problem. By using ASIC devices, such as VCUs, Google can keep up with the demand and deliver the best possible quality. Codenamed Argos, the chip can deliver 20-33x improvement in efficiency compared to the regular server platform. In data centers, the VCU is implemented as a regular PCIe card, with two chips under the heatsinks.

GALAX, the maker of the popular premium Hall Of Fame (HOF) edition of graphics cards, has recently announced its GeForce RTX 3090 HOF Edition GPU. Designed for extreme overclocking purposes, the card is made with a 12 layer PCB, 26 phase VRM power delivery configuration, and three 8-pin power connectors. Today, we have managed to get the first comprehensive review of the card by a Chinese YouTube channel 二斤自制. However, this wasn't just regular testing being conducted on a card with factory settings. The channel has applied 1000 Watt vBIOS to the GPU and ran it all on the air cooler the GPU comes with.

In the default 420 Watt setting, the card has been running with a GPU clock of 1845 MHz and a temperature of 69 degrees Celsius. However, when the 1000 Watt vBIOS was applied to the card, the GPU core has managed to ramp to 2000 MHz and consume as much as 630 W of power. If you were wondering if the stock cooler was able to handle it all, the answer is yes. The card has reached a toasty 96 C temperature. While GALAX doesn't offer BIOS like this, the ID of the BIOS corresponds to that of a custom XOC 1000 W BIOS for EVGA Kingpin GeForce RTX 3090 GPU, which you can find in our database. When it comes to performance, the gains were very minimal at only 2-3%. That must have been due to the insufficient cooling, and the card could have done much better on water or LN2. The Firestrike Ultra and Firestrike Extreme results are displayed below.

With NVIDIA's GeForce RTX 30-series lineup of GPUs, laptop manufacturers are offered a wide variety of GPU SKUs that internally differ simply by having different Total Graphics Power (TGP), which in turn results in different clock speeds and thus different performance. ASUS uses NVIDIA's variant of GeForce RTX 3080 mobile GPU inside the company's ROG Zephyrus Duo (GX551QS) with a TGP of 115 Watts, and Dynamic Boost technology that can ramp up the card to 130 Watts. However, this doesn't represent the maximum for RTX 3080 mobile graphics card. The maximum TGP for RTX 3080 mobile goes up to 150 Watts, which is a big improvement that lets the GPU reach higher frequencies and more performance.

Have you ever wondered what would happen if you manually applied vBIOS that allows the card to use more power? Well, Baidu forum users are reporting a successful experiment of transforming their 115 W RTX 3080 to 150 W TGP card. Using GPU vBIOS from MSI Leopard G76, which features a 150 W power limit, and applying it to the ROG's Zephyrus Duo power-limited RTX 3080 cards is giving results. Users have successfully used this vBIOS to squeeze out more performance from their laptops. As seen on the 3D Mark Time Spy rank list, the entries are now dominated solely by modified laptops. Performance improvement is, of course, present and it reaches up to a 20% increase.

European Union has today processed a declaration that was signed by 17 member states about the development of a 2 nm semiconductor node and an advanced low-power processor. The declaration signed today proposes that the EU puts away 145 billion Euros for the development of the technologies needed to manufacture a 2 nm semiconductor manufacturing process, along with the development of a custom, low-power embedded processor designed for industrial applications. The 17 member countries include Belgium, France, Germany, Croatia, Estonia, Italy, Greece, Malta, Spain, The Netherlands, Portugal, Austria, Slovenia, Slovakia, Romania, Finland, and Cyprus. All of the countries listed are going to join the development of these technologies and will have the funds to do it over the next 2-3 years.

EU DeclarationTo ensure Europe's technology sovereignty and competitiveness, as well as our capacity to address key environmental and societal challenges and new emerging mass markets, we need to strengthen Europe's capacity to develop the next generation of processors and semiconductors. This includes chips and embedded systems that offer the best performance for specific applications across a wide range of sectors as well as leading-edge manufacturing progressively advancing towards 2 nm nodes for processor technology. Using connectivity, where Europe enjoys global lead, as a major use case driver for developing such capacity enables Europe to set the right level of ambition. This will require a collective effort to pool investment and to coordinate actions, by both public and private stakeholders.

Tachyum Inc. today announced that it is signing early adopter customers for the software emulation system for its Prodigy Universal Processor, customers may begin the process of native software development (i.e. using Prodigy Instruction Set Architecture) and porting applications to run on Prodigy. Prodigy software emulation systems will be available at the end of January 2021.

Customers and partners can use Prodigy's software emulation for evaluation, development and debug, and with it, they can begin to transition existing applications that demand high performance and low power to run optimally on Prodigy processors. Pre-built systems include a Prodigy emulator, native Linux, toolchains, compilers, user mode applications, x86, ARM and RISC-V emulators. Software updates will be issued as needed.

Designing a custom processor can be a rewarding thing. You can control your ecosystem surrounding it and get massive rewards in terms of application-specific performance uplift, or lower total cost of ownership. It seems like cloud providers have figured out that at their scale, designing a custom processor can get all of the above with the right amount of effort put into it. If you remember, in 2018, Amazon has announced its Graviton processor based on Arm instruction set architecture. Today, the company has almost 10% of its AWS instances based on the Graviton 1 or 2 processors, which is a massive win for a custom design.

Following Amazon's example, the next company to join the custom server processor race is going to be Microsoft. The Redmond based giant is looking to build a custom lineup of processors that are meant to satisfy Microsoft's most demanding sector - server space. The company's Azure arm is an important part where it has big and increasing revenue. By building a custom processor, it could satisfy the market needs better while delivering higher value. The sources of Bloomberg say that Microsoft is planning to use Arm ISA, and start building independence from the x86 vendors like Intel and AMD. Just like we saw with AWS, the industry cloud giants are starting to get silicon-independent and with their scale, they can drive the ecosystem surrounding the new processors forward rapidly. The sources are also speculating that the company is building custom processors for Surface PCs, and with Windows-on-Arm (WoA) project, Microsoft has laid the groundwork in that field as well.

ASUS has today surprised everyone and decided to launch the first custom design of AMD's Radeon RX 6900 XT graphics card. Previously, the word was that AMD was debating with AIBs whatever to open the GPU to custom designs or keep it AMD PCB exclusive. However, thanks to today's launch, we now know that AMD will allow its partners to design their PCBs and push the Big Navi silicon to its maximum. So when it comes to pushing to maximum, enter the world of ASUS TUF Gaming Radeon RX 6900 XT graphics cards, designed to fit in the line of ASUSes TUF principals.

Featuring a triple fan, triple-slot (2.9 slots more precisely )body, the graphics card is built around Navi 21 XTX GPU. This means that only the best needs to be brought to the card as it is a premium product. That is why the company says that "The TUF GAMING Radeon RX 6900 XT is a tenacious beast with a tough metal exterior, super-efficient cooling, and components that offer enhanced endurance." The built-in cooler offers 0dB cooler technology, meaning that fans will not spin unless the GPU reaches 55 degrees C temperature. The card is powered by two eight-pin connectors, so it seems that power supply requirements are not changed compared to the reference card. While the exact specifications are not known, you can expect the card to boost over the standard 2250 MHz frequency, as it is factory overclocked. Pricing is also not yet confirmed but a slight premium is expected as well.

Just a few days ago AMD has announced its Radeon RX 6000 series of graphics cards based on the new RDNA 2 architecture. While AMD has given out the "Big Navi" chips to its partners to design custom boards and give users designs with better cooling and possibly higher overclocking capabilities, that doesn't seem to extend to the highest-end parts. So far, we have seen custom designs from companies like ASUS, MSI, etc., and all of them have one thing in common - they only do designs for Radeon RX 6800 or RX 6800 XT. So one would wonder where are the highest-end custom Radeon RX 6900 XT designs.

The first wave of the "custom" cards will be on November 18th, when manufacturers will release designs that are MBA (Made-by-AMD), meaning that the PCB is a reference design, just with a custom cooler installed. When it comes to the custom RX 6900 XT cards, AMD is now in talks with its partners whether to keep the biggest "Big Navi" design available for custom designs, or to keep it as AMD exclusive, with the most likely scenario being the AMD exclusivity. AMD partners could carry the models in their stores and offerings, however, the PCB and cooler design would be AMD's. The situation is yet unresolved so we have to wait and see what comes out of it and if we are getting any custom designs of the Radeon RX 6900 XT model.

In a somewhat puzzling trend, this is yet another launch from CORSAIR's Hydro X series of custom watercooling components that has not been marketed by the company. It took a careful set of eyes to see the new CPU block, the Hydro X Series XC5, on their website and evidently this is a new release targeting even more a budget-friendly build compared to the Hydro XC7 and Hydro XC9 blocks. The XC5 block is a cost-down version of the XC7, with support for Intel LGA 115x/1200 and AMD AM4 CPU sockets depending on the SKU you go with. Gone is the transparent top, and instead we see a black nylon top with a patterned cutout for RGB lighting. The nickel-plated copper cold plate is replaced with passivated copper, which may also please all you pure copper folks out there, and the 16 RGB LEDs are now set up in 8 lighting zones (two LEDs per zone) as opposed to fully addressable 16 dRGB LEDs on the XC7. This means that the Hydro XC5 blocks can be had for $60/€60 before applicable taxes, which is $20 less than the XC7 on street pricing at the moment.

Intel today unveiled the suite of new security features for the upcoming 3rd generation Intel Xeon Scalable platform, code-named "Ice Lake." Intel is doubling down on its Security First Pledge, bringing its pioneering and proven Intel Software Guard Extension (Intel SGX) to the full spectrum of Ice Lake platforms, along with new features that include Intel Total Memory Encryption (Intel TME), Intel Platform Firmware Resilience (Intel PFR) and new cryptographic accelerators to strengthen the platform and improve the overall confidentiality and integrity of data.

Data is a critical asset both in terms of the business value it may yield and the personal information that must be protected, so cybersecurity is a top concern. The security features in Ice Lake enable Intel's customers to develop solutions that help improve their security posture and reduce risks related to privacy and compliance, such as regulated data in financial services and healthcare.

We first noticed a silent update to CORSAIR's Hydro X GPU blocks with the upcoming RTX 30-series block, with a third screw and a new I/O terminal added to reflect the design of their Dominator Platinum RGB memory sticks. The third screw is no doubt to help quench any concerns over potential leaks of coolant with the older design, which a few people noted was a possibility when under duress, as opposed to a typical use case necessarily. As it turns out, dealing with any bad PR was worth even retroactively making the change with the older RTX 20-series block too, which we now see on their website and is pictured below.

In addition, the company has been adding white color options of their more popular components for people to use as a change from the typical black. Following the Hydro X theme here, we see the company is now selling white Hydro XR5 radiators with a matte white polyurethane coating over the Hardware Labs Black Ice Nemesis LS radiators they are based off. The larger XR7 radiators will not get the same treatment unfortunately, given they are not as popular clearly as the thinner XR5, although you can now also get both pump/reservoir units in white too. Their cooling configurator has been updated with the white parts as well.

TSMC has hit a jackpot with its newer nodes like 7 nm and now 5 nm, as the company is working with quite good yields. To boast, TSMC has seen all of its capacity of 7 nm being fully booked by customers like AMD, Apple, and NVIDIA. However, it seems like the company's next-generation 5 nm node is also getting high demand. According to the latest report from DigiTimes, TSMC's N5 5 nm node is fully booked to the end of 2020. And the biggest reason for that is the biggest company in the world - Apple. Since Apple plans to launch the next-generation iPhone, iPad, and Arm-based MacBook, the company has reportedly booked most of the 5 nm capacity for 2020, meaning that there are lots of chips that Apple will consume. TSMC can't be dependent only on one company like Apple, so the smaller portion of capacity went to other customers as well.

CORSAIR announces more new products each year than most companies, and this often means that marketing resources get devoted to the high-ticket items more often than not. Their custom watercooling line, the Hydro X Series, is not always one of them and so we end up seeing quiet updates without a press release or even a direct note from their representatives to the media. With the new NVIDIA GeForce RTX 3000-series cards coming in three different categories as far as custom watercooling goes (reference, Founders Edition, AIC-variants), we took the weekend to briefly check what the usual custom watercooling suspects were up to. A page on the CORSAIR website confirms, among other things, that Hydro X GPU blocks are on the way for at least the reference PCB, if not more.

More interestingly, we noted that the Hydro XC7 and XC9 CPU blocks had been updated in more ways than one. We have detailed reviews of both (XC7, XC9) for those who want to see, and our review of the iCUE Nexus also showed a rebranding in the making for the company. The updated CPU blocks carry the new wordmark design for the logo on all 5 SKUs and, more importantly, change the mounting mechanism to allow for a captive system with new thumb screws that should make installation even easier on all Intel and AMD CPU sockets. This should bring with it increased compatibility with crowded m-ITX motherboards as well, and we see it reflected in the additional 719 motherboards in the compatibility list relative to before. Finally, a motherboard ARGB LED adapter is now included in the packaging, so you can use the integrated lighting with motherboard LED headers and control software, should you not want to pony extra for CORSAIR's ecosystem. Pricing for the Hydro XC7 is now $5 higher and the same as the XC9, which is confusing given it uses fewer fins on the cold plate and, the launch versions at least, employed a plastic trim compared to the metal on the XC9. We have contacted the company for clarification on this, and will update the post as we receive more information.

[Update, September 16: CORSAIR has a new photo for their example GPU block that looks suspiciously like it is meant for the FE RTX 3000-series cards. Note also the revised I/O port terminal which should help address some concerns people had about the previous design.]

ASUS Republic of Gamers (ROG) today announced an all-new lineup of gaming gear at the Meta Buffs online launch event. Aligning with the event theme, the new products are all meta buffs: status enhancements that are lighter, faster and more powerful for taking gaming experiences to a new level. The lineup features the new ROG Strix GeForce RTX 30 Series graphics cards, which unleash an amped-up thermal design and chart-topping performance with the NVIDIA Ampere architecture. The new NVIDIA GeForce RTX 30 Series GPUs, the 2nd generation of RTX, feature new RT Cores, Tensor Cores and streaming multiprocessors, bringing stunning visuals, amazingly fast frame rates and AI acceleration to games and creative applications. Powered by the NVIDIA Ampere architecture, which delivers increases of up to 1.9 times performance-per-watt over the previous generation, the RTX 30 Series effortlessly powers graphics experiences at all resolutions, even up to 8K at the top end. The GeForce RTX 3090, 3080 and 3070 represent the greatest GPU generational leap in the history of NVIDIA.

ROG also introduced the ROG Swift PG329Q, ROG Swift 360 Hz PG259QN and ROG Swift 360 Hz PG259QNR gaming monitors for ultrafast and ultrasmooth visuals; the ROG Strix XF 120 fan for low-noise cooling performance and versatility; ROG Rapture GT-AXE11000, the world's first Wi-Fi 6E router; the ROG Strix Scope RX gaming keyboard, which debuts the responsive, reliable and durable ROG RX Red Optical Mechanical Switches; the ROG Keris Wireless mouse for compact and nimble control; the ROG Delta S headset for crystal-clear audio; and ROG Scabbard II, an extended gaming mouse pad with a protective nano-coating that repels water, oil and dust.

When Apple announced their transition form Intel processors to Apple Silicon, we were left wondering how the silicon will perform and what characteristics will it bring with it. According to the latest report from The China Times, the Apple custom GPU found inside the new Apple Silicon will bring better performance and energy efficiency compared to Intel iGPU it replaces. The 5 nm GPU manufactured on TSMC's N5 semiconductor manufacturing node is supposedly codenamed "Lifuka" and it brings Apple's best to the table. Planned to power a 12-inch MacBook, the GPU will be paired with a custom CPU based on Arm ISA as well. The same chips powering iPhone and iPad devices will go into MacBook devices, with the TDP increased as MacBook will probably have much higher cooling capacity. The first Apple Silicon MacBook will come in H2 of 2021.Here is the copy of a full report from The China Times below:

Here's the first render of a finished custom-design GeForce RTX 3090 graphics card, courtesy of a leak by HD Tecnologia. The ZOTAC GeForce RTX 3090 Trinity Holo is shown featuring a conventional axial-flow fan-heatsink design. The card is taller than standard by a good inch or so, over 2 slots thick, and appears to feature an acrylic diffuser with RGB LED elements. The RGB embellishments cover both sides of the card.

A recurring theme with "Ampere" cards appears to be thicker-than-standard backplates. On this card, the backplate looks chunky, with RGB elements all over. As for the cooling solution, it features a triple-fan setup ventilating a large heatsink, and possibly a baseplate underneath. We can see the new-generation multi-GPU connector. Variable fan-speeds for individual fans could be a new design trend with this generation. More renders of the RTX 3090 Trinity Holo surfaced, revealing three 8-pin PCIe power connectors. Renders of the RTX 3080 Holo by ZOTAC also surfaced, revealing a smaller product design with just two fans, and two 8-pin PCIe power inputs. A see-through render shows that NVIDIA is planning to use two grades of the same GA102 silicon on both the RTX 3090 and RTX 3080, differing in core-configuration, memory bus, and clock speeds.Update Aug 30th: VideoCardz scored a family shot of possibly all of ZOTAC's upcoming custom RTX 30-series board designs.

Intel has been focusing its resources on data center and high-performance computing lately and the company has made some interesting products. Today, Intel has released its latest creation - mOS operating system. Created as a research project, Intel has made an OS made for some extreme-scale HPC systems, meaning that the OS is created for hyper scalers and ones alike. The goal of mOS is to deliver a high-performance environment for software with low-noise, scalability, and the concept of lightweight kernels (LWK) that manage the system.. Being based on the Linux kernel, the OS is essentially another distribution, however, it has been modified so it fits the HPC ecosystem the best way. The mOS is a product in the pre-alpha phase, however, it can already be used in supercomputers like ASCI Red, IBM Blue Gene, and others. Intel is aiming to develop a stable release by the time the Aurora exascale system is ready so it can deploy mOS there.

Corsair today introduced the Hydro X series iCUE custom water cooling kits. These include the Hydro X Series iCUE XH305i, and the Hydro X Series iCUE XH303i. Both kits are nearly identical, differing with the included radiator fans and reservoirs. The XH305i kit includes an XC7 RGB CPU water block that supports AM4, LGA115x, and LGA1200 CPU socket types; an XR5 360 radiator, an XD5 pump+reservoir combo device, XL5 clear coolant, 3 m XT Hardline 10 mm and 14 mm tubing, eight Hardline 14 mm fittings, a pair of 90° rotary adapters, a 250 ml filling bottle, a Hardline tubing bending toolkit, a trio of QL 120 RGB fans, iCUE Commander PRO, and RGB Fan LED Hub. The XH303i is mostly similar to the XH305i, except that it includes a trio of SP 120 RGB PRO fans, and XD3 pump+radiator combo device. Corsair is pricing the Hydro X Series iCUE XH305i at 579.90€, while the XH303i is going for 499.90€. Corsair is also selling the Hardline Bending Toolkit for 49.90€.

Marvell today announced a unique custom ASIC offering that addresses the stringent requirements of next generation 5G carriers, cloud data centers, enterprise and automotive applications. Marvell's comprehensive custom ASIC solution enables a multitude of customization options and a differentiated approach with best-in-class standard product IP including Arm -based processors, embedded memories, high-speed SerDes, networking, security and a wide range of storage controller and accelerators in 5 nm and beyond. By partnering with Marvell, customers gain enhanced performance, power and area resulting in accelerated time-to-market and providing optimal returns on investment.

Traditionally, data infrastructure manufacturers and cloud data center operators have had to choose between securing standard products or a full custom silicon solution designed in-house, while developing or licensing foundational IP as needed. Now, for the first time, Marvell is offering full access to its broad and growing portfolio of industry-leading data infrastructure standard product IP and technologies, which can be integrated and enabled in custom ASIC solutions at the most advanced technology nodes.

Both Sony and Microsoft are expected to soon launch their next-generation consoles, which are highly anticipated among console gamers as they will bring some of the latest advancements in graphics, like ray tracing. The current situation in the world is that everything is slowed down due to the COVID-19 outbreak. Companies across the world have mostly slowed down their production lines due to the less demand, however, AMD has done exactly the opposite for their custom SoC customers. In wake of expected high demand, AMD has ramped up its production lines so it can supply Sony and Microsoft for their next-generation consoles—PlayStation 5 and Xbox Series X. This is an encouraging sign that the demand is going to be strong - and that AMD is up to the task of delivering.

Today, Arm announced significant additions to its artificial intelligence (AI) platform, including new machine learning (ML) IP, the Arm Cortex -M55 processor and Arm Ethos -U55 NPU, the industry's first microNPU (Neural Processing Unit) for Cortex-M, designed to deliver a combined 480x leap in ML performance to microcontrollers. The new IP and supporting unified toolchain enable AI hardware and software developers with more ways to innovate as a result of unprecedented levels of on-device ML processing for billions of small, power-constrained IoT and embedded devices.

European Processor Initiative (EPI) is a Europe's project to kickstart a homegrown development of custom processors tailored towards different usage models that the European Union might need. The first task of EPI is to create a custom processor for high-performance computing applications like machine learning, and the chip prototypes are already on their way. The EPI chairman of the board Jean-Marc Denis recently spoke to the Next Platform and confirmed some information regarding the processor design goals and the timeframe of launch.

Supposed to be manufactured on TSMC's 6 nm EUV (TSMC N6 EUV) technology, the EPI processor will tape-out at the end of 2020 or the beginning of 2021, and it is going to be heterogeneous. That means that on its 2.5D die, many different IPs will be present. The processor will use a custom ARM CPU, based on a "Zeus" iteration of Neoverese server core, meant for general-purpose computation tasks like running the OS. When it comes to the special-purpose chips, EPI will incorporate a chip named Titan - a RISC-V based processor that uses vector and tensor processing units to compute AI tasks. The Titan will use every new standard for AI processing, including FP32, FP64, INT8, and bfloat16. The system will use HBM memory allocated to the Titan processor, have DDR5 links for the CPU, and feature PCIe 5.0 for the inner connection.