AMD today announced a goal to deliver a 30x increase in energy efficiency for AMD EPYC CPUs and AMD Instinct accelerators in Artificial Intelligence (AI) training and High Performance Computing (HPC) applications running on accelerated compute nodes by 2025.1 Accomplishing this ambitious goal will require AMD to increase the energy efficiency of a compute node at a rate that is more than 2.5x faster than the aggregate industry-wide improvement made during the last five years.

Accelerated compute nodes are the most powerful and advanced computing systems in the world used for scientific research and large-scale supercomputer simulations. They provide the computing capability used by scientists to achieve breakthroughs across many fields including material sciences, climate predictions, genomics, drug discovery and alternative energy. Accelerated nodes are also integral for training AI neural networks that are currently used for activities including speech recognition, language translation and expert recommendation systems, with similar promising uses over the coming decade. The 30x goal would save billions of kilowatt hours of electricity in 2025, reducing the power required for these systems to complete a single calculation by 97% over five years.

Remember that Core i9-12900K CPU-Z leak from last week? It had the multi-threaded score blurred out and now we know why. A new CPU-Z screenshot has shown up on Twitter and although the single threaded score is still beating the AMD Ryzen 5950X baseline single core score by a comfortable margin, it's behind when we're switching to the multi-threaded score.

It shouldn't really come as a surprise that eight big and eight small CPU cores doesn't beat AMD's 16 big cores, but this was apparently expected by some. This is not saying that Intel doesn't get close as you can see, but it's also worth keeping in mind that Intel runs on 24 threads vs. AMD's 32 threads. The Core i9-12900K is said to be running on stock clocks, but no other information was provided. Once again, take this for what it is while we wait for the actual launch date and proper benchmarks.

Silicon Lottery is a boutique online store offering prebinned Intel and AMD CPUs that have been tested to reach guaranteed clock speeds when overclocked. The store has been operating for seven years but have recently announced that they will be ceasing operation on October 31st. This news comes as the culmination of multiple factors that have been impacting the store including a shift from Intel to prebin their K-series processors reducing their overclocking potential. This maximization of existing silicon also meant that there was limited opportunity for the store to find enough faster processors to sell. The final influence was the move by Intel to switch from thermal compound under the IHS to solder Tim with their 9th generation CPUs which severely limits the thermal advantages gained from delidding. The general component shortages and delays associated with the COVID-19 pandemic have also impacted the business, the complete statement from Silicon Lottery can be found below.

Finally some potentially good news, as AMD's CEO Lisa Su is bringing hope that the current chip shortage situation might improve over the next 18 to 24 months according to a new piece on CNBC. She's expecting new chip fabs to have come online by then, although no details were mentioned, one would presume it involves TSMC in AMD's case.

Lisa Su is quoted saying "We've always gone through cycles of ups and downs, where demand has exceeded supply, or vice versa, this time, it's different." "The pandemic has just taken demand to a new level". This isn't exactly breaking news by now, but it also seems like the demand for computers has reached its peak and is now plateauing ahead of what will likely be a drop in sales come next year, but that doesn't mean the demand for chips will go down. Lisa Su is also expecting further consolidation in the industry, which has its upsides and downsides, but her take on it is that "if you want to do something very large for the industry, you know, scale is important." AMD should know this better than most companies, since they've scaled their business from the brink of bankruptcy to where they are today.

AMD CEO Dr. Lisa Su last Friday joined an exclusive list of personalities whose contributions to he semiconductor industry have been deemed relevant enough to receive the prestigious Robert N. Noyce medal. The award, attributed by the IEEE and funded by Intel, was awarded to Lisa Su in recognition of her "leadership in groundbreaking semiconductor products and successful business strategies that contributed to the strength of the microelectronics industry." Her current and past actions at AMD have pulled most of the weight behind this recognition, as Dr. Lisa Su has completely turned around a company that was bleeding talent and dollars, reversing its 2$ per share lows from AMD's 2014 up to today's $110 per share.

Lisa Su divides her carrier in two parts: the first ten to 15 years where she moved and produced as an MIT-trained electrical engineer, where she earned bachelor's, master's, and doctoral degrees on the subject. The move towards management of research and technological teams actually happened during her stint in IBM; after 11 years at the company, in 2006 she was appointed vice president of IBM's semiconductor research and development center in New York. She then moved on to AMD as senior vice president in charge of the company's global business units, where she was so impressively skilled it only took her two years to become President and CEO of AMD. Her nomination for the Robert N. Noyce Medal paints her as the first woman to have ever received it. In 1993, MIT female graduates where 32% of the total; in 2016, that number increased to 50%.

Stephanie Brenham, Team Lead Programmer for Ubisoft's upcoming AAA Far Cry 6, recently spoke to WCCFTech on the upcoming Far Cry installment. Stephanie went into some detail regarding the graphics and performance options, and an interesting fact that surfaced was that neither Sony's PS5 nor Microsoft's Xbox Series consoles will feature ray tracing enabled on their respective versions of the game. Apparently, ray tracing will be a PC-exclusive feature, as console versions of the game are targeting higher render resolution and more fluid framerates over expensive graphics options such as ray tracing. And even on PC, it'll be a hybrid form of it, and not a full implementation: ray tracing is supported for both shadows and reflections, but Ubisoft opted for a hybrid approach here, marrying traditional rendering with ray tracing so as to improve performance in mainstream PC hardware.

"Ray tracing is a PC-only feature," Stephanie Brenham said. "On console, our objective has been to take advantage of new hardware capabilities, optimizing performance targeting 4K and achieving 60 FPS." This does somewhat fall in the face of performance expectations set by both Sony and Microsoft; both companies made (and still make) extensive use of ray tracing support on the marketing campaigns for their consoles. However, as we've seen in the past, enabling ray tracing comes with severe performance penalties in even the latest and greatest PC hardware (sometimes not to best effect, even), which still outclasses even the latest consoles' powerful innards (compared to their predecessors, of course).

AMD's Fidelity Super Resolution (FSR) tech is being implemented in RPSCS3, one of the foremost emulators for Sony's Playstation 3. The emulator allows PC users to play otherwise PS3-exclusive games via software emulation. The nature of this emulation, however, leads to a couple important aspects. One pertains to performance: emulating non-existent hardware is one of the most resource-hungry workloads one can think of, and is highly dependent on the emulator's coding quality. Another is that since this is a software solution, it does allow to changes in maximum render resolution, for example, or the addition of visual effects or other modifications to the rendering pipeline. One limitation of this approach is that game support has to be added almost manually, checking and correcting the emulators' behaviors according to the software being played.

AMD's FSR tech been received with a rather enthusiastic response. This is in part due to its open source nature, but also because of its apparent ease of implementation and its higher compatibility with graphics cards new, old, and from the competition - unlike NVIDIA's DLSS, which requires specific hardware (Tensor cores) to be present in the GPU chip, locking it to only the latest NVIDIA products. This nature of FSR has led to its integrationn on the RPCS3 emulator, promising a relatively easy to implement performance and image quality increase compared to the original rendering pipeline, including 4K upscaling. Check after the break for a video of the tech in action (spoiler: the quality difference isn't nearly as close as what the thumbnail implies).

An alleged Intel Core i9-12900K "Alder Lake-S" sample is shown beating the 32-core AMD Ryzen Threadripper 2990WX HEDT processor at AMD's favorite benchmark, Cinebench R23, in its multi-threaded (nT) test. At this point it's not known whether the i9-12900K is overclocked, but the CPU-Z instance in the screenshot reads 5.30 GHz, which could very well be the processor's stock Thermal Velocity Boost frequency. The sample scored upward of 30000 points, putting it above the Threadripper 2990WX reference score in Cinebench.

The 2990WX is based on the "Zen+" microarchitecture, and released in 2018, but is a 32-core/64-thread chip that should have ripped through this rendering workload. The i9-12900K, on the other hand, has eight "Golden Cove" performance cores that have HyperThreading, in addition to 8 "Gracemont" efficiency cores that lack HTT. This benchmark was run on Windows 10, which lacks awareness of the Intel Thread Director, a hardware component that optimizes utilization of the two kinds of CPU cores. Windows 11 is known to feature better awareness of hybrid core architectures. The i9-12900K sample is possibly installed on a Gigabyte Z690 AORUS Ultra motherboard, and has 32 GB of DDR5-5200 memory (two modules, logically four 40-bit channels).

The AMD Radeon RX 6600 is expected to launch in October after documents received by VideoCardz reveal that reviews for the card are set to be published on October 13th. The documents reveal that board partners who will be releasing cards for review will need to have informed AMD by September 15th and can begin shipping them to reviewers on September 29th. The AMD Radeon RX 6600 will use the Navi 23 GPU with 4 Compute Units disabled for a total of 28. This will give the card 1792 Stream Processors which will be paired with 8 GB of GDDR6 memory. The documents also show that AMD is not planning to release an RX 6600 reference card so no pricing information was included. We expect that the card will be shortly available after the listed review embargo is lifted on October 13th at 9 AM EST.

AMD today released the latest version of their Radeon Software Adrenalin drivers, guaranteeing the best possible experience for AMD GPU users. The latest driver revision introduces support for World War Z: Aftermath, and Amazon's New World MMORPG. Besides that initial game release support, the drivers also deliver an up to 13% performance improvement in Diablo II: Resurrected. This performance increase seems to solve a rendering bottleneck experienced with the RX 6700 XT graphics card at 4K resolution, with Very High settings enabled.

AMD also fixed four outstanding issues with this driver release. The CPU Additional Metrics section within the Performance tab should now show up for all users; compatibility error messages when trying to load a previously saved tuning profile should no longer appear; water should no longer spontaneously evaporate in Hitman III for RX 6000 series users; CPU Auto Overclock in Radeon Software should no longer be absent for users sporting an AMD 5000 series CPU alongside an RX 6000 series graphics card. Check after the break for the list of outstanding issues; as always, you can grab the drivers from the link below.DOWNLOAD: AMD Radeon Software Adrenalin 21.9.2

The prices for NVIDIA Ampere and AMD RDNA 2 GPUs have increased 6% - 7% in Germany since August according to the latest report from 3DCenter. The report covers the period from January 2021 to September 2021 where prices reached 304% and 214% of MSRP for NVIDIA RTX 30 and AMD RX 6000 cards earlier this year before coming down to 150% and 153% respectively. We had hoped that this trend might continue downwards but these reductions have started to reverse with prices steadily increasing since early August. The latest period from August 29th to September 19th saw NVIDIA RTX 30 series prices increase from 159% of MSRP to 170% while AMD RX 6000 cards rose from 164% to 174%.

These price trends can be fairly closely attributed to the changes in supply as shown with the addition of an availability trend line to the graph in this latest report. The card which has experienced the largest price increase relative to MSRP is the AMD Radeon RX 6800 XT which now sells for double the recommended € 649 price at € 1299. The least affected card is the NVIDIA RTX 3080 Ti retailing for 26% above it's € 1199 MSRP at € 1511.

XFX formally launched the Radeon RX 6900 XT Speedster ZERO WB, a graphics card it teased last month. The company's new flagship product, the Speedster ZERO WB is a graphics card with a factory-fitted full-coverage water-block, for those with DIY liquid-cooling setups. It appears to be based on an all new PCB with a more tuned-up 14+2 phase VRM setup than that of the air-cooled RX 6900 XT Speedster MERC 319, which pulls power from a trio of 8-pin PCIe power connectors.

Under the hood is a 7 nm "Navi 21" XTXH silicon, which is able to sustain 10% higher engine clocks than the standard "Navi 21," and XFX claims that it has unlocked overdrive slider limit (the de facto maximum overclock), beyond 3 GHz. Out of the box, the card comes with a maximum boost frequency of 2525 MHz, compared to 2250 MHz AMD-reference. The water block came about from a collaboration with EK Water Blocks, and combines a nickel-plated copper primary material with a slightly frosted acrylic top that's studded with addressable RGB LEDs. XFX didn't reveal pricing.

With the 15.6 and 17.3 inch XMG APEX gaming laptops, XMG is positioning a new model series below its own high-end range consisting of the NEO and PRO series. These new laptops combine mobile AMD eight-core processors up to the Ryzen 9 5900HX with NVIDIA GeForce RTX graphics cards up to the RTX 3070. The company is simultaneously introducing the XMG FOCUS, a new product series in the entry-level segment. Intel's Core i7-11800H and an NVIDIA GeForce RTX 3050 Ti offer decent gaming performance, while good connectivity and memory round off the overall package. All four new models feature an IPS display with 144 Hz.

Until now, the XMG APEX 15 in the older E20 generation represented uncompromising desktop CPU performance, with processors up to the Ryzen 9 3950X in a laptop. Although XMG is already working on a direct successor under a slightly different name, it is unleashing the XMG APEX 15 and APEX 17 of the M21 generation for the time being with the currently fastest eight-core mobile processors from AMD. The laptops are available with an AMD Ryzen 7 5800H as well as with the slightly faster Ryzen 9 5900HX from the 54 watt TDP class, as well as with an NVIDIA GeForce RTX 3070 or RTX 3060 in the respective maximum TGP configuration (RTX 3070: 125 watts plus 15 watts Dynamic Boost 2.0; RTX 3060: 115 watts plus 15 watts Dynamic Boost 2.0).

Chinese cooler manufacturer Cool Server have recently listed several upcoming coolers for the AMD Zen 4 AM5 & SP5 sockets. The manufacturer has listed 5 AM5 coolers, and 4 SP5 coolers all targeted towards the enterprise sector. The lineup includes several passive coolers which rely on case airflow while the others feature high-performance fans which can get quite noisy. The AM5 socket will be introduced with the next-generation Zen 4 Ryzen processors while the SP5 (LGA6096) socket has been prepared for the Zen 4 EPYC processors. The complete list of coolers can be found below.

In view of the ongoing production capacity shortage in the semiconductor industry and the resultant price hike of chips, revenue of the top 10 IC design companies for 2Q21 reached US$29.8 billion, a 60.8% YoY increase, according to TrendForce's latest investigations. In particular, Taiwanese companies put up remarkable performances during this period, with both MediaTek and Novatek posting YoY growths of more than 95%. AMD, on the other hand, experienced a nearly 100% YoY revenue growth, the highest among the top 10.

TrendForce indicates that the ranking of the top five companies for 2Q21 remained unchanged from the previous quarter, although there were major changes in the 6th to 10th spots. More specifically, after finalizing its acquisition of Inphi, Marvell experienced a major revenue growth and leapfrogged Xilinx and Realtek in the rankings from 9th place in 1Q21 to 7th place in 2Q21.

AMD Chief Financial Officer Devinder Kumar has recently commented that the company "stands ready to manufacture Arm chips" if demand from business customers exists. These remarks were made at the Deutsche Bank Technology Conference held last week and build on comments made by CEO Dr. Lisa Su earlier this year confirming the companies willingness to create Arm chips. AMD has some experience creating Arm products with its K12 architecture which never reached market and their embedded microcontrollers such as the Platform Security Processors. While AMD hasn't confirmed if they have any current Arm projects underway there appears to be interest from customers which AMD is more than happy to fulfill. The pending sale of Arm to NVIDIA may complicate this if NVIDIA reneges on their pledge to continue Arm's open licensing system.

Thermaltake released its highest-capacity dual-channel (2-module) DDR4 memory kit, the ToughRAM XG RGB R016R432GX2-3600C18A. This kit uses modules that are a swansong of DDR4 in terms of capacity—32 GB x 2. The 32 GB module is dual-rank, and only compatible with some of the later generations of Intel Core and AMD Ryzen processors that feature memory controllers with support for up to 128 GB of memory in total. On its part, Thermaltake claims that the modules were only tested to be compatible with Intel 10th Gen and 11th Gen Core; and AMD Ryzen 2000 series and above. In terms of performance, the ToughRAM XG RGB R016R432GX2-3600C18 packs an Intel XMP 2.0 profile for DDR4-3600 with 18-22-22-42 + CR2T timings, and 1.35 V module voltage. The module is aesthetically identical to the ToughRAM XG RGB series the company debuted in March 2021.

AMD today released the latest version of Radeon Software Adrenalin. Version 21.9.1 is feature-packed with stuff enthusiasts are bound to love. To begin with, it introduces Smart Access Memory (PCI Resizable BAR) support for AMD Radeon RX 5000 series (RDNA) graphics processors, when used with a compatible platform. Next up, it adds optimization for "Call of Duty Vanguard Open Beta," and "Deathloop." The drivers also add a handful Vulkan API extensions, including VK_KHR_zero_initialize_workgroup_memory, VK_KHR_shader_integer_dot_product, VK_EXT_shader_atomic_float2, and VK_EXT_load_store_op_none. Perhaps the biggest new feature addition is AMD Auto Overclock.

On machines with AMD Ryzen 5000 series processors and Radeon RX 6000 series graphics, the Auto-Overclock feature detects CPU load, and automatically overclocks it in increments of 50 MHz, beyond the processor's rated maximum boost frequency. Just like manual overclocking, this comes with sizable costs to power consumption and cooling. The feature also automatically overclocks your RX 6000 series GPU. Using the feature is as easy as pulling up the Tuning tab in Radeon Software, and clicking on "Auto Overclock" in the Tuning Control tab. Lastly, Adrenalin 21.9.1 adds support for the upcoming Windows 11 operating system. The software is released alongside AMD's latest 3.09.01.140 Chipset Software, which also add Windows 11 support. A handful bugs have also been fixed, as described below.

DOWNLOAD: AMD Radeon Software Adrenalin 21.9.1 | AMD Chipset Software 3.09.01.140

Intel and AMD released the first Radeon Software drivers for the exotic Radeon Vega M graphics found in Intel NUC8i7HNK and NUC8i7HVK, in over 16 months. The Intel Download Center just added version 21.10.03.11 of Radeon Software for the graphics solution, dated 09/09/2021. The previous drivers dated all the way back to February 2020, and were based on Adrenalin 20.2.

The drivers are based on a release candidate of Adrenalin 21.10, which is 21.10RC1, but does not include the security updates AMD could be bundling with the upcoming 21.10 drivers (October 2021). "Radeon RX Vega M Graphics Driver version 21.10.03.11, which is based on 21.10RC1, does not include the latest functional and security updates. An update is targeted to be released in March of 2022 and will include additional functional and security updates. Customers should update to the latest version as it becomes available," says Intel in the release notes of the drivers. The drivers do include optimization for the latest games, including "Metro Exodus" (DX12), and "Resident Evil Village," besides adding support for Microsoft PlayReady AV1 decode.

DOWNLOAD: AMD Radeon Vega M drivers for Intel NUC8i7HNK and NUC8i7HVK from Intel

AMD is slowly preparing the launch of the latest and greatest Ryzen processor family based on the Zen 4 CPU core design. Among various things that are getting an overhaul, the Raphael processor generation is now getting revamped temperature reading and better power management circuitry. According to an Igor's Lab report, AMD has prepared a few new improvements that will make temperature reading and power management easier for PC enthusiasts. Currently, the reported CPU temperature is called Tcontrol (Tctl), which is what the cooling solution sees. If Tctl is high, the fans spin up and cool the system. If Tctl is low, the fans slow down to reduce noise.

With Raphael, the CUR_TEMP (current temperature) output part of Tctl has been upgraded to reflect a much smoother curve, and avoid jittering with fans as they are not spiking so suddenly anymore. This is helping contribute to the noise output and has made it run at a consistent fan speed in the system. Another note about Raphael is a new power management technique. AMD has designed the AM5 platform to avoid sudden power spikes, to maintain maximum efficiency over time. It is a design decision made from the very start, and the CPU will try to constrain itself in the TDP range that it is configured for. For more details about the circuitry, please head over to the Igor's Lab article.

Microsoft is potentially looking to refresh the Xbox Series S in late 2022 with an upgraded 6 nm AMD APU according to Moore's Law is Dead. The upgraded processor would be manufactured on TSMC's 6N process which boasts higher yields and could allow Microsoft to enable all 24 Compute Units on the APU compared to the 20 they currently enable. This increase in Compute Units and a clock speed boost could potentially increase the console's performance by 50%. This updated model would come in at close to 350 USD representing a 50 USD premium however the existing model would be retained and see a price cut to 189-249 USD. The rumor also claims that Microsoft will refresh the Xbox Series X in 2023 or later.

AMD's "Zen 3" CCD, or compute complex die, the physical building-block of both its client- and enterprise processors, possibly has a core count limitation owing to the way the various on-die bandwidth-heavy components are interconnected, says an AnandTech report. This cites what is possibly the first insights AMD provided on the CCD's switching fabric, which confirms the presence of a Ring Bus topology. More specifically, the "Zen 3" CCD uses a bi-directional Ring Bus to connect the eight CPU cores with the 32 MB of shared L3 cache, and other key components of the CCD, such as the IFOP interface that lets the CCD talk to the I/O die (IOD).

Imagine a literal bus driving around a city block, picking up and dropping off people between four buildings. The "bus" here resembles a strobe, the buildings resemble components (cores, uncore, etc.,) while the the bus-stops are ring-stops. Each component has its ring-stops. To disable components (eg: in product-stack segmentation), SKU designers simply disable ring-stops, making the component inaccessible. A bi-directional Ring Bus would see two "vehicles" driving in opposite directions around the city block. The Ring Bus topology comes with limitations of scale, mainly resulting from the latency added from too many ring-stops. This is precisely why coaxial ring-topology faded out in networking.

BIOSTAR, a leading brand of motherboards, graphics cards, and storage devices, today announced product support for the latest AMD Ryzen 5000G series Cezanne processors. AMD's next-generation Ryzen 5000G series desktop processors codenamed "Cezanne" are ready to invade the global market. The new 5000G series processors are based on Zen 3 architecture, AMD's Ryzen 5000 series of desktop APUs based on the Zen 3 CPU and Vega GPU microarchitectures succeeding the Ryzen 4000 "Renoir" series.

Extreme performance enabled for personal computing with up to 8 cores fueled by the world's most advanced 7 nm processor core technology, the AMD Ryzen 5000 G-series desktop processors with Radeon graphics deliver ultra-fast responsiveness and multi-threaded performance for any use case.

Despite the intensifying COVID-19 pandemic that swept Taiwan in 2Q21, the domestic OSAT (outsourced semiconductor assembly and test) industry remained largely intact, according to TrendForce's latest investigations. Global sales of large-sized TVs were brisk thanks to major sporting events such as the Tokyo Olympics and UEFA Euro 2020. Likewise, the proliferation of WFH and distance learning applications propelled the demand for IT products, while the automotive semiconductor and data center markets also showed upward trajectories. Taking into account the above factors, OSAT companies raised their quotes in response, resulting in a 26.4% YoY increase in the top 10 OSAT companies' revenue to US$7.88 billion for 2Q21.

TrendForce indicates that, in light of the ongoing global chip shortage and the growing production capacities of foundries/IDMs in the upstream semiconductor supply chain, OSAT companies gradually increased their CAPEX and expanded their fabs and equipment in order to meet the persistently growing client demand. However, the OSAT industry still faces an uncertain future in 2H21 due to the Delta variant's global surge and the health crisis taking place in Southeast Asia, home to a significant number of OSAT facilities.

Chinese news site MyDrivers have recently reported on increasing graphics card prices in China over the last month. These price increases come as a result of reduced GPU shipments especially of the NVIDIA RTX 3060 whose supply is down 50% with improvements not expected for several weeks. The site has tracked the prices for popular cards from ASUS, Gigabyte, and MSI with increases across the board ranging from a 12 USD increase on the Gigabyte GTX 1050 Ti up to a 92 USD increase on the ASUS RTX 3070 Ti. These supply impacts have impacted NVIDIA to a greater extent than AMD but the general price increases are likely indicative of the global trend over the coming weeks.