It looks like cryptocurrency mining is back in craze, as miners are firing up their old mining hardware from 2022 to cash in. Bitcoin is now north of $72,000, and is dragging up the value of several other cryptocurrencies, one such being Qubic (QBIC). Profitability calculators put 24 hours of Qubic mining on an AMD Ryzen 9 7950X 16-core processor at around $3, after subtracting energy costs involved in running the chip at its default 170 W TDP. "Zen 4" processors such as the 7950X tend to retain much of their performance with slight underclocking, and reducing their power limits; which is bound to hold or increase profitability, while also prolonging the life of the hardware.

And thus, the inevitable has happened—stocks of the AMD Ryzen 9 7950X have disappeared overnight across online retail. With the market presence of the 7950X3D and the Intel Core i9-14900K, the 7950X was typically found between $550-600, which would have added great value considering its low input costs. CPU-based cryptocurrency miners, including the QBIC miner, appear to be taking advantage of the AVX-512 instruction set. AMD "Zen 4" microarchitecture supports AVX-512 through its dual-pumped 256-bit FPU, and the upcoming "Zen 5" microarchitecture is rumored to double AVX-512 performance over "Zen 4." Meanwhile, Intel has deprecated what few client-relevant AVX-512 instructions its Core processors had since 12th Gen "Alder Lake," as it reportedly affected sales of Xeon processors. What about the 7950X3D? It's pricier, but mining doesn't benefit from the 3D V-cache, and the chip doesn't sustain the kind of CPU clocks the 7950X manages to do across all its 16 cores. It's only a matter of time before the 7950X3D disappears, too; followed by 12-core models such as the 65 W 7900, the 170 W 7900X, and the 7900X3D.

A 2019-signed export license has allowed Intel to supply laptop processors to Huawei, under an exclusive deal—this US Government approved arrangement was not viewed favorably by AMD. The rival chipmaker apparently missed out on the securing of a similar trade license back in 2021. According to a new Reuters report, Team Red and a handful of supporters have attempted to revoke Intel's license—worth hundreds of millions of dollars. Two anonymous sources allege that: "Intel has survived an effort to halt chip sales to Huawei...giving one of the world's largest chipmakers more time to sell to the heavily sanctioned Chinese telecoms company." Intel and Huawei's symbiosis is set to end later this year—folks on the inside reckon that the current US administration will not approve a renewal. Reports suggest that Qualcomm is not anticipating a renewal either—Huawei is an approved buyer of Snapdragon chips, but industry whispers indicate an eventual shift to in-house fare.

Intel, Huawei, US Commerce Department and the White House have declined to comment on the aforementioned scenario. Reuters also sent a query to AMD, but the publication did not receive a response. Earlier last year, a government official revealed that "Huawei's licensing policy" was under review, alongside a general push to scrap a number of trade deals. According to insiders, the same government official allegedly told companies—in private—that the US Commerce Department would fix "the licensing discrepancy." Another anonymous source believes that the agency shelved these plans late last year, for reasons unknown—they stressed that there is potential for a revival. Given the upcoming expiry of Intel and Huawei's arrangement—within the year—it makes little sense to implement a drastic change.

Thanks to Phoronix, we have discovered that AMD has recently released initial patches for performance monitoring and events related to their upcoming Zen 5 processors in the Linux kernel. These patches, sent out for review on the kernel mailing list, provide the necessary JSON files for PMU (Performance Monitoring Unit) events and metrics that will be exposed through the Linux perf tooling. As the patches consist of JSON additions and do not risk regressing existing hardware support, there is a possibility that they could be included in the upcoming Linux v6.9 kernel cycle. This would allow developers and enthusiasts to access detailed performance data for Zen 5 CPUs once they become available, helping with optimization and analysis of the next-generation processors.

The release of these patches follows AMD's publication of performance monitor counter documentation for AMD Family 1Ah Model 00h to 0Fh processors last week, confirming that these models represent the upcoming Zen 5 lineup. While Linux kernel 6.8 already includes some elements of Zen 5 CPU support, the upstream Linux enablement for these next-generation AMD processors is an ongoing process. Upon Phoronix examining the Zen 5 core and uncore events, as well as the metrics and mappings, it appears that they are mainly similar to those found in the current Zen 4 processors. This suggests that AMD has focused on refining and optimizing the performance monitoring capabilities of its new architecture rather than introducing significant changes. As the launch of Zen 5 CPUs draws closer, we await to see the performance and capabilities of these next-generation processors. With performance monitoring also getting a push, this could be a sign that Zen 5 launch is nearing.

Microsoft's DirectX department is scheduled to show off several innovations at this month's Game Developers Conference (GDC), although a late February preview has already spilled their DirectSR Super Resolution API's beans. Today, retail support for Shader Model 6.8 and Work Graphs has been introduced with an updated version of the company's Agility Software Development Kit. Program manager, Joshua Tucker, stated that these technologies will be showcased on-stage at GDC 2024—Shader Model 6.8 arrives with a "host of new features for shader developers, including Start Vertex/Instance Location, Wave Size Range, and Expanded Comparison Sampling." A linked supplementary article—D3D12 Work Graphs—provides an in-depth look into the cutting-edge API's underpinnings, best consumed if you have an hour or two to spare.

Tucker summarized the Work Graphs API: "(it) utilizes the full potential of your GPU. It's not just an upgrade to the existing models, but a whole new paradigm that enables more efficient, flexible, and creative game development. With Work Graphs, you can generate and schedule GPU work on the fly, without relying on the host. This means you can achieve higher performance, lower latency, and greater scalability for your games with tasks such as culling, binning, chaining of compute work, and much more." AMD and NVIDIA are offering driver support on day one. Team Red has discussed the launch of "Microsoft DirectX 12 Work Graphs 1.0 API" in a GPUOpen blog—they confirm that "a deep dive" into the API will happen during their Advanced Graphics Summit presentation. NVIDIA's Wessam Bahnassi has also discussed the significance of Work Graphs—check out his "Advancing GPU-driven rendering" article. Graham Wihlidal—of Epic Games—is excited about the latest development: "we have been advocating for something like this for a number of years, and it is very exciting to finally see the release of Work Graphs."

LG Electronics (LG) announced today pricing and availability of its 2024 UltraGear OLED gaming monitors, designed to provide superior gaming experiences. The lineup is led by the 45GS95QE-B—an immersive 45-inch curved LG OLED monitor with a 21:9 aspect ratio. The expanded lineup includes the highly anticipated CES 2024 Innovation Award-winning 32-inch 4K gaming monitor (model 32GS95UE) equipped with the new Dual Mode feature which allows gamers to switch between 4K (3,840 x 2,160) at 240 Hz and Full-HD (1,920 x 1,080) at 480 Hz with one simple click.

The company is also introducing its 39- and 34-inch models with ultra-wide curved screens: the 39GS95QE and 34GS95QE. All four models are now available for pre-order thru March 17, 2024 at LG.com (shipping expected in late March and throughout April) and available later at LG-authorized retailers. Customers who place a pre-order for the 45GS95QE-B, 39GS95QE or 34GS95QE UltraGear OLED gaming monitors via LG.com are eligible to receive a $200 Virtual Mastercard prepaid card. Terms and conditions apply.

Lately Tiny Corp. CEO—George Hotz—has used his company's social media account to publicly criticize AMD Radeon RX 7900 XTX GPU firmware. The creator of Tinybox, a pre-orderable $15,000 AI compute cluster, has not selected "traditional" hardware for his systems—it is possible that AMD's Instinct MI300X accelerator is quite difficult to acquire, especially for a young startup operation. The decision to utilize gaming-oriented XFX-branded RDNA 3.0 GPUs instead of purpose-built CDNA 3.0 platforms—for local model training and AI inference—is certainly a peculiar one. Hotz and his colleagues have encountered roadblocks in the development of their Tinybox system—recently, public attention was drawn to an "LLVM spilling bug." AMD President/CEO/Chair, Dr. Lisa Su, swiftly stepped in and promised a "good solution." Earlier in the week, Tiny Corp. reported satisfaction with a delivery of fixes—courtesy of Team Red's software engineering department. They also disclosed that they would be discussing matters with AMD directly, regarding the possibility of open-sourcing Radeon GPU MES firmware.

Subsequently, Hotz documented his interactions with Team Red representatives—he expressed 70% confidence in AMD approving open-sourcing certain bits of firmware in a week's time: "Call went pretty well. We are gating the commitment to 6x Radeon RX 7900 XTX on a public release of a roadmap to get the firmware open source. (and obviously the MLPerf training bug being fixed). We aren't open source purists, it doesn't matter to us if the HDCP stuff is open for example. But we need the scheduler and the memory hierarchy management to be open. This is what it takes to push the performance of neural networks. The Groq 500 T/s mixtral demo should be possible on a tinybox, but it requires god tier software and deep integration with the scheduler. We also advised that the build process for amdgpu-dkms should be more open. While the driver itself is open, we haven't found it easy to rebuild and install. Easy REPL cycle is a key driver for community open source. We want the firmware to be easy to rebuild and install also." Prior to this week's co-operations, Tiny Corp. hinted that it could move on from utilizing Radeon RX 7900 XTX, in favor of Intel Alchemist graphics hardware—if AMD's decision making does not favor them, Hotz & Co. could pivot to builds including Acer Predator BiFrost Arc A770 16 GB OC cards.

AMD has caught up with NVIDIA and Intel in the race to get a locally run AI chatbot up and running on its respective hardware. Team Red's community hub welcomed a new blog entry on Wednesday—AI staffers published a handy "How to run a Large Language Model (LLM) on your AMD Ryzen AI PC or Radeon Graphics Card" step-by-step guide. They recommend that interested parties are best served by downloading the correct version of LM Studio. Their CPU-bound Windows variant—designed for higher-end Phoenix and Hawk Point chips—compatible Ryzen AI PCs can deploy instances of a GPT based LLM-powered AI chatbot. The LM Studio ROCm technical preview functions similarly, but is reliant on Radeon RX 7000 graphics card ownership. Supported GPU targets include: gfx1100, gfx1101 and gfx1102.

AMD believes that: "AI assistants are quickly becoming essential resources to help increase productivity, efficiency or even brainstorm for ideas." Their blog also puts a spotlight on LM Studio's offline functionality: "Not only does the local AI chatbot on your machine not require an internet connection—but your conversations stay on your local machine." The six-step guide invites curious members to experiment with a handful of large language models—most notably Mistral 7b and LLAMA v2 7b. They thoroughly recommend that you select options with "Q4 K M" (AKA 4-bit quantization). You can learn about spooling up "your very own AI chatbot" here.

HP Inc. today announced the industry's largest portfolio of AI PCs leveraging the power of AI to enhance productivity, creativity, and user experiences in hybrid work settings.

In an ever-changing hybrid work landscape, workers are still struggling with disconnection and digital fatigue. HP's 2023 Work Relationship Index reveals that only 27% of knowledge workers have a healthy relationship with work, and 83% believe it's time to redefine our relationships with work. Most employees believe AI will open new opportunities to enjoy work and make their jobs easier, but they need the right AI tools and technology to succeed.

AMD Linux engineers have been working on "GFX1150" and "GFX1151" targets for a while—official references to "Strix 1/Strix Point" and "Strix Point Halo" have appeared several times on official development channels. Phoronix's head honcho—Michael Larabel—monitors these activities with keen interest, his latest finding indicates that Team Red is preparing open-source RadeonSI/RADV driver support for the GFX1151 IP. Their MESA 24.1 update merges in GPU enablement for possible high-end "Strix Point Halo" laptop processors—tech tipsters believe that these chiplet variants could sport up to sixteen Zen 5 CPU cores and forty RDNA 3.5 GPU cores.

AMD's enablement of the "GFX1150/Strix Point" GPU appeared online late last month—these monolithic laptop chips are alleged to sit below "Strix Point Halo" in Team Red's product hierarchy. Insiders suggest that the best configurations could house twelve Zen 5 CPU cores and sixteen RDNA 3.5 GPU cores. Phoronix posited that the "RDNA 3 refresh" graphics solution: "is just rumored for select APUs, while ultimately we'll see where this GFX 11.5.1 IP is found if for some further upgraded APU or something more special. In any event the open-source Linux driver support is coming together." According to official product roadmaps, the initial batch of "Strix Point" mobile chips are expected ship later this year—representing a proper next-gen upgrade over current "Hawk Point" offerings.

AMD in an announcement on Tuesday, announced changes to its AMD FreeSync display requirements. The updated requirements should vastly reduce the flood of 1080p and 1440p displays with lower refresh rates, such as 75 Hz, from boasting of FreeSync support. AMD has three tiers of FreeSync based on the refresh-rate and features such as HDR. The base tier FreeSync feature now calls for a desktop monitor or TV to come with a refresh rate of at least 144 Hz, if its horizontal resolution is under 3440 pixels. This would cover everything 2560 x 1600, to 2560 x 1440, to the popular 1920 x 1080, and certain ultrawide resolutions such as 2560 x 1080.

To qualify for the FreeSync Premium tier, which customers use to identify high refresh-rate displays; a desktop monitor or TV with a horizontal resolution under 3440 pixels now needs at least 200 Hz of refresh rate. Displays with over 3440 pixels (which should include 4K Ultra HD), now need at least 120 Hz to qualify for FreeSync Premium. AMD isn't clear on what refresh-rate it takes for a 4K UHD display to qualify for the base FreeSync tier. Moving on, the top FreeSync Premium Pro tier, which helps customers identify displays that not only have high refresh-rates but also HDR; nothing is changed—the display must feature HDR on top of meeting the new requirements for FreeSync Premium. The story is different for notebook displays. For the base FreeSync tier, a refresh-rate of 40-60 Hz will do; and for FreeSync Premium, you need at least 120 Hz. HDR on top of this gives you FreeSync Premium Pro. All displays will need FreeSync hardware support (which essentially builds on top of VESA Adaptive Sync).

AMD's 12-core/24-thread Ryzen 9 7900X3D processor that comes with 3D Vertical Cache technology, is seeing a slew of price-cuts. The SKU appears to be trapped between the 8-core 7800X3D, which remains the fastest desktop processor for gaming; and the 16-core 7950X3D, which is AMD's flagship. A Newegg listing at $408 with a $20 discount coupon see its effective retail price drop to as low as $389, which is just $20 more than the 7800X3D.

Would you pay $20 for four more cores? The choice is not so simple. While the 7900X3D is a 12-core processor, it features a dual-CCD design, with a 6+6 core arrangement between the two CCDs. Just like with the 7950X3D, only one of the two CCDs has the 64 MB 3D Vertical Cache, or 96 MB of L3 cache; while the other is a regular "Zen 4" CCD with 32 MB of on-die L3 cache. AMD's chipset drives use UEFI CPPC preferred-core flagging to guide gaming workloads to the CCD with the 3D V-cache. While we haven't had a chance to test the 7900X3D, this chip is tested by Tom's Hardware to still be faster than the Core i9-13900K at gaming.

The TinyBox AI server system attracted plenty of media attention last week—its creator, George Hotz, decided to build with AMD RDNA 3.0 GPU hardware rather than the expected/traditional choice of CDNA 3.0. Tiny Corp. is a startup firm dealing in neural network frameworks—they currently "write and maintain tinygrad." Hotz & Co. are in the process of assembling rack-mounted 12U TinyBox systems for customers—an individual server houses an AMD EPYC 7532 processor and six XFX Speedster MERC310 Radeon RX 7900 XTX graphics cards. The Tiny Corp. social media account has engaged in numerous NVIDIA vs. AMD AI hardware debates/tirades—Hotz appears to favor the latter, as evidenced in his latest choice of components. ROCm support on Team Red AI Instinct accelerators is fairly mature at this point in time, but a much newer prospect on gaming-oriented graphics cards.

Tiny Corporation's unusual leveraging of Radeon RX 7900 XTX GPUs in a data center configuration has already hit a developmental roadblock. Yesterday, the company's social media account expressed driver-related frustrations in a public forum: "If AMD open sources their firmware, I'll fix their LLVM spilling bug and write a fuzzer for HSA. Otherwise, it's not worth putting tons of effort into fixing bugs on a platform you don't own." Hotz's latest complaint was taken onboard by AMD's top brass—Dr. Lisa Su responded with the following message: "Thanks for the collaboration and feedback. We are all in to get you a good solution. Team is on it." Her software engineers—within a few hours—managed to fling out a set of fixes in Tiny Corporation's direction. Hotz appreciated the quick turnaround, and proceeded to run a model without encountering major stability issues: "AMD sent me an updated set of firmware blobs to try. They are responsive, and there have been big strides in the driver in the last year. It will be good! This training run is almost 5 hours in, hasn't crashed yet." Tiny Corp. drummed up speculation about AMD open sourcing GPU MES firmware—Hotz disclosed that he will be talking (on the phone) to Team Red leadership.

According to the latest report from Bloomberg, AMD has hit a roadblock in offering its top-of-the-line AI accelerator in the Chinese market. The newest AI chip is called Instinct MI309, a lower-performance Instinct MI300 variant tailored to meet the latest US export rules for selling advanced chips to China-based entities. However, the Instinct MI309 still appears too powerful to gain unconditional approval from the US Department of Commerce, leaving AMD in need of an export license. Originally, the US Department of Commerce made a rule: Total Processing Performance (TPP) score should not exceed 4800, effectively capping AI performance at 600 FP8 TFLOPS. This rule ensures that processors with slightly lower performance may still be sold to Chinese customers, provided their performance density (PD) is sufficiently low.

However, AMD's latest creation, Instinct MI309, is everything but slow. Based on the powerful Instinct MI300, AMD has not managed to bring it down to acceptable levels to acquire a US export license from the Department of Commerce. It is still unknown which Chinese customer was trying to acquire AMD's Instinct MI309; however, it could be one of the Chinese AI labs trying to get ahold of more training hardware for their domestic models. NVIDIA has employed a similar tactic, selling A800 and H800 chips to China, until the US also ended the export of these chips to China. AI labs located in China can only use domestic hardware, including accelerators from Alibaba, Huawei, and Baidu. Cloud services hosting GPUs in US can still be accessed by Chinese companies, but that is currently under US regulators watchlist.

JEDEC Solid State Technology Association, the global leader in the development of standards for the microelectronics industry, is pleased to announce the publication of JESD239 Graphics Double Data Rate (GDDR7) SGRAM. This groundbreaking new memory standard is available for free download from the JEDEC website. JESD239 GDDR7 offers double the bandwidth over GDDR6, reaching up to 192 GB/s per device, and is poised to meet the escalating demand for more memory bandwidth in graphics, gaming, compute, networking and AI applications.

JESD239 GDDR7 is the first JEDEC standard DRAM to use the Pulse Amplitude Modulation (PAM) interface for high frequency operations. Its PAM3 interface improves the signal to noise ratio (SNR) for high frequency operation while enhancing energy efficiency. By using 3 levels (+1, 0, -1) to transmit 3 bits over 2-cycles versus the traditional NRZ (non-return-to-zero) interface transmitting 2 bits over 2-cycles, PAM3 offers higher data transmission rate per cycle resulting in improved performance.

"The Thaumaturge" the action RPG with turn-based combat, is getting support for AMD FSR 3 at launch, including frame generation, which provides a near-doubling in frame-rates. As of this writing, AMD hasn't released a new Adrenalin driver release with any optimizations as such, which means that the game is already optimized for AMD hardware. This is because AMD lists this as part of its Featured Games, which means that the game is natively optimized for AMD Radeon RX GPUs, and Ryzen CPUs. Will you need FSR 3, or even DLSS 3 FG game-optimization that NVIDIA just released? We doubt—the game can make do with a Radeon GPU as old as the RX 580 "Polaris," and its recommended GPU calls for an RX Vega 64, which means even an RX 7600 should max out the game at 1080p thru 1440p.

AMD today announced the AMD Spartan UltraScale+ FPGA family, the newest addition to the extensive portfolio of AMD Cost-Optimized FPGAs and adaptive SoCs. Delivering cost and power-efficient performance for a wide range of I/O-intensive applications at the edge, Spartan UltraScale+ devices offer the industry's highest I/O to logic cell ratio in FPGAs built in 28 nm and lower process technology, deliver up to 30 percent lower total power consumption versus the previous generation, and contain the most robust set of security features in the AMD Cost-Optimized Portfolio.

"For over 25 years the Spartan FPGA family has helped power some of humanity's finest achievements, from lifesaving automated defibrillators to the CERN particle accelerator advancing the boundaries of human knowledge," said Kirk Saban, corporate vice president, Adaptive and Embedded Computing Group, AMD. "Building on proven 16 nm technology, the Spartan UltraScale+ family's enhanced security and features, common design tools, and long product lifecycles further strengthen our market-leading FPGA portfolio and underscore our commitment to delivering cost-optimized products for customers."

AMD announced that Thomas Zacharia has joined AMD as senior vice president of strategic technology partnerships and public policy. Zacharia will lead the global expansion of AMD public/private relationships with governments, non-governmental organizations (NGOs) and other organizations to help fast-track the deployment of customized AMD-powered AI solutions to meet rapidly growing number of global projects and applications targeting the deployment of AI for the public good.

"Thomas is a distinguished leader with decades of experience successfully creating public/private partnerships that have resulted in consistently deploying the world's most powerful and advanced computing solutions, including the world's fastest supercomputer Frontier," said AMD Chair and CEO Lisa Su. "As the former Director of the U.S.'s largest multi-program science and energy research lab, Thomas is uniquely positioned to leverage his extensive experience advancing the frontiers of science and technology to help countries around the world deploy AMD-powered AI solutions for the public good."

V-COLOR today introduced a series of overclocking memory RDIMM kits for workstations powered by AMD Ryzen Threadripper 7000WX processors on the WRX90 platform that features 8-channel DDR5 memory. The kits include 8 RDIMMs, with densities ranging between 16 GB per RDIMM (128 GB per kit), to 96 GB per RDIMM (768 GB per kit); and comes in speeds ranging between DDR5-5600 and DDR5-7200. The best part? These modules feature AMD EXPO profiles, which should make enabling their advertised speeds as easy as a couple of clicks in the motherboard's UEFI setup program.

An EXPO profile not just applies the kit's memory speed, timings, and voltages, but also several sub-timings and settings that are specific to the AMD platform, which are not found on Intel. V-COLOR has tested its overclocking RDIMMs on popular AMD WRX90 chipset motherboards, namely the ASRock WRX90 WS EVO, ASUS PRO WS WRX90E-SAGE SE, and certain unreleased WRX90 workstation motherboards by Supermicro. Although the RDIMMs lack heatspreaders for the DRAM chips, V-COLOR is including what it calls "micro heatsinks" for the PMIC and RCDs. The RCD in particular is crucial to get Threadrippers to operate at speeds such as DDR5-7200. The kits should be available starting today, with all models available from mid-March. The company didn't reveal pricing.

AMD CTO Mark Papermaster confirmed that AMD is working on a new upscaling technology that leverages AI. A key technological difference between AMD FSR and competing solutions NVIDIA DLSS and Intel XeSS, has been AMD's remarkable restraint in implementing AI in any part of the upscaler's pipeline. Unlike FSR, both DLSS and XeSS utilize AI DNNs to overcome temporal artifacts in their upscalers. AMD Radeon RX 7000 series GPUs and Ryzen 7000 CPUs are the first with accelerators or ISA that speed up AI workloads; and with the RX 7000 series capturing a sizable install-base, AMD is finally turning to AI for the next generation of its FSR upscaling tech. Papermaster highlighted his company's plans for AI in upscaling technologies in an interview with No Priors.

To a question by No Priors on exploring AI for upscaling, Papermaster responded: "2024 is a giant year for us because we spent so many years in our hardware and software capabilities for AI. We have just completed AI-enabling our entire portfolio, so you know cloud, edge, PCs, and our embedded devices, and gaming devices. We are enabling gaming devices to upscale using AI and 2024 is a really huge deployment year." In short, Papermaster walked the interviewer through the 2-step process in which AMD is getting into AI, with a hardware-first approach.

Some of the earliest signs are emerging that AMD's mobile processor or desktop APU silicon that succeeds "Strix Point" being codenamed "Sound Wave." AMD tends to come up with quirky internal codenames for upcoming projects, mostly to zero in on the source of leaks, so "Sound Wave" as a codename is subject to change with time. While the upcoming 4 nm "Strix Point" and "Strix Halo" chips implement the "Zen 5" CPU microarchitecture and RDNA 3+ graphics architecture, besides XDNA 2 based NPU with a generational tripling in AI TOPS; Wccftech believes that "Sound Wave" could be an AMD processor of comparable class to "Strix Point," which implements the "Zen 6" CPU microarchitecture, which AMD has planned for a 2025-26 timeframe.

Perhaps the most interesting aspect of this leak is the foundry node, with the original source over at Korean tech blog Gamma0burst referencing 3 nm. This is the final node family from TSMC to implement FinFET transistors before the foundry transitions to nanosheets with N2. It's likely that AMD chooses one of the more advanced variants of TSMC's 3 nm nodes, such as the N3P or N3X, because 2025-26 will see rival Intel get close to introducing the Intel 20A foundry node for mass-production. Not much else is known about "Sound Wave" besides the "Zen 6" CPU cores at this point.

AMD announced three new entries to the growing list of games that support FSR 3.0, "The Last of Us Part 1," "Remnant II," and "RoboCop: Rogue City." The three will soon get patches from their developers that add FSR 3 support. "The Last of Us Part 1" is a particularly important title in the list. FSR 3 entails updates to the upscaling algorithm that provide more image quality at a given preset compared to FSR 2, but more importantly, it introduces frame-generation, with an interpolation technology that nearly doubles framerates. FSR 3 support will be available to those with Radeon RX 6000 series and RX 7000 series GPUs.

AMD recently tried to add support for key HDMI 2.1 features like 4K@120 Hz and 5K@240 Hz to their open-source Linux graphics driver called AMDGPU. They invested engineering resources over several months to prototype the necessary code internally before publishing. The goal was to showcase HDMI 2.1 capabilities and get the implementation approved by the HDMI Forum. Unfortunately, the Forum ultimately rejected AMD's request, blocking Linux users of new AMD Radeon GPUs from utilizing those cutting-edge display features over HDMI. In comments, AMD stated: "The HDMI Forum has rejected our proposal unfortunately. At this time an open source HDMI 2.1 implementation is not possible without violating HDMI Forum requirements." This outcome comes as a major disappointment given the time and effort AMD expended aiming to satisfy the Forum's guidelines. The months of work now feel wasted with this outright rejection. As reasoning, the HDMI Forum cited legal and compliance rules around not enabling open-source HDMI 2.1 code.

Legal issues and compliance are major problems for open-source HDMI developers, as HDMI Forum has decided to make the HDMI specification private in 2021. This directly translates into the newest open-source driver developments, where the latest features will probably remain behind a closed-source binary. Consequently, AMD is advising Linux gamers to use DisplayPort if they want access to features like 4K 120 Hz gaming. Meanwhile, Windows AMD users still get full HDMI 2.1 capabilities. This dichotomy spotlights the ongoing obstacles around open-source driver development. The rejection also strains the AMD - HDMI Forum relationship. AMD hoped spearheading open-source HDMI 2.1 drivers would position them as leaders in the open-source community. Instead, their flexibility plea was denied by the rigid HDMI Forum requirements. Ultimately, whether Linux-based AMD owners can ever utilize next-gen HDMI 2.1 displays fully remains to be determined. For now, AMD continues pushing open-source as the best approach, while the HDMI Forum refuses to budge on compliance demands. Both sides seem firmly entrenched, leaving consumers caught in the middle.

Microsoft has just announced that their new DirectSR Super Resolution API for DirectX will provide a unified interface for developers to implement super resolution in their games. This means that game studios no longer have to choose between DLSS, FSR, XeSS, or spend additional resources to implement, bug-test and support multiple upscalers. For gamers this is huge news, too, because they will be able to run upscaling in all DirectSR games—no matter the hardware they own. While AMD FSR and Intel XeSS run on all GPUs from all vendors, NVIDIA DLSS is exclusive to Team Green's hardware. With their post, Microsoft also confirms that DirectSR will not replace FSR/DLSS/XeSS with a new upscaler by Microsoft, rather that it builds on existing technologies that are already available, unifying access to them.

While we have to wait until March 21 for more details to be revealed at GDC 2024, Microsoft's Joshua Tucker stated in a blog post: "We're thrilled to announce DirectSR, our new API designed in partnership with GPU hardware vendors to enable seamless integration of Super Resolution (SR) into the next generation of games. Super Resolution is a cutting-edge technique that increases the resolution and visual quality in games. DirectSR is the missing link developers have been waiting for when approaching SR integration, providing a smoother, more efficient experience that scales across hardware. This API enables multi-vendor SR through a common set of inputs and outputs, allowing a single code path to activate a variety of solutions including NVIDIA DLSS Super Resolution, AMD FidelityFX Super Resolution, and Intel XeSS. DirectSR will be available soon in the Agility SDK as a public preview, which will enable developers to test it out and provide feedback. Don't miss our DirectX State of the Union at GDC to catch a sneak peek at how DirectSR can be used with your games!"

TechPowerUp's resident GPU reviewer extraordinaire—W1zzard—has grappled with a handful of custom design AMD Radeon RX 7900 GRE 16 GB models. Team Red and its board partners are pushing a proper/widespread Western release of the formerly China market-exclusive "Golden Rabbit Edition" GPU. TPU's initial review selection of three Sapphire cards and a lone ASRock Steel Legend OC variant garnered two Editor's Choice Awards, and two Highly Recommended badges. Sapphire's Radeon RX 7900 GRE Nitro+ was also honored with a "...But Expensive" tag, due to its MSRP of $600—the premium tier design was one of last year's launch day models in China. Western reviewers have latched onto a notable GRE overclocking limitation—all of TPU's review samples were found to have "overclocking artificially limited by AMD." Steve Walton of Hardware Unboxed has investigated whether the GRE's inherent heavily limited power specification was less of an issue on Sapphire's Nitro+ variant—check out his "re-re-review" video below.

The higher board power design—305 W OC TGP limit and 351 W total board power—is expected to exhibit "up to 10% higher performance than Radeon RX 7800 XT" according to VideoCardz, but falls short. TPU's W1zzard found the GRE Nitro+ card's maximum configurable clock of 2803 MHz: "Overclocking worked quite well on our card, we gained over 8% in real-life performance, which is well above what we usually see, but less than other GRE cards tested today. Sapphire's factory OC eats into OC potential, and maximizes performance out of the box instead. Unfortunately AMD restricted overclocking on their card quite a lot, probably to protect sales of the RX 7900 XT. While NVIDIA doesn't have any artificial limitations for overclockers, AMD keeps limiting the slider lengths for many models, this is not a gamer-friendly approach. For the GRE, both GPU and memory overclocking could definitely go higher based on the results that we've seen in our reviews today." An AMD representative has contacted Hardware Unboxed, in reaction to yesterday's Update review—the GRE's overclocking limitation is a "bug," and a fix is in the works. This situation is a bit odd, given that the Golden Rabbit Edition is not a brand-new product.

TrendForce underscores that the primary momentum for server shipments this year remains with American CSPs. However, due to persistently high inflation and elevated corporate financing costs curtailing capital expenditures, overall demand has not yet returned to pre-pandemic growth levels. Global server shipments are estimated to reach approximately. 13.654 million units in 2024, an increase of about 2.05% YoY. Meanwhile, the market continues to focus on the deployment of AI servers, with their shipment share estimated at around 12.1%.

Foxconn is expected to see the highest growth rate, with an estimated annual increase of about 5-7%. This growth includes significant orders such as Dell's 16G platform, AWS Graviton 3 and 4, Google Genoa, and Microsoft Gen9. In terms of AI server orders, Foxconn has made notable inroads with Oracle and has also secured some AWS ASIC orders.