In addition to the TITAN and GUARDIAN SKUs announced earlier this month, Sparkle is working on several other SKUs. The roadmap includes the low-profile version of the Arc B570, as well as the Arc B580 ROC OC Ultra, which is expected to come with a 2,800 MHz GPU factory overclock and 210 W TBP, both coming next year.

According to the roadmap, Sparkle plans to release the B580 ROC OC Ultra version in February 2025, and this one will be a part of Sparkle's ROC Luna series, featuring an all-white design. As said, it gets a 2,800 MHz GPU factory-overclock, which is 60 MHz higher than the Sparkle Arc B580 TITAN OC. It also has a slightly higher 210 W TBP. Sparkle included a small picture showing a 2.5-slot thick design with a dual-fan cooler. Additional roadmap also confirms the launch of the Arc B570 Low-Profile version, which will feature a lower 170 W TBP and a three-fan cooler, similar to what we have seen from GUNNIR lately.

Thanks to the discovery of VideoCardz, we get a glimpse of MAXSUN's latest Arc 580 GPU with not only a GPU but extra room for two additional M.2 SSDs. The PCIe connector on the Intel Arc B580 has x16 physical pins but runs at PCIe 4.0 x8 speeds. Intel verified it runs only x8 lanes instead of the full x16 slot, leaving x8 lanes unsued. However, MAXSUN thought of a clever way to put the leftover x8 lanes to good use by adding two PCIe x4 M.2 SSDs to thelatest triple-fan iCraft B580 SKU. Power delivery for the M.2 drives comes directly from the graphics card, which is made possible by the GPU's partial PCIe lane utilization. This configuration could prove particularly valuable for compact builds or systems with limited motherboard storage options.

Interestingly, the SSD pair appears to have its own thermal enclosure, which acts as a heatsink. Having constant airflow from the GPU's fans, the M.2 SSD configuration should be able to maximize the full bandwidth of the SSDs without thermals throttling the SSD read/write speeds. The design follows in the footsteps of AMD's Radeon Pro SSG, which introduced integrated storage in workstation cards with PCIe 3.0 M.2 slots. Back then, it was mainly a target for workstation users. However, MAXSUN has envisioned gamers unusually expanding their storage space now. The pricing of the card and availability date remains a mystery.

We have already confirmed that Intel is continuing the development of Arc gaming GPUs beyond the current Xe2 "Battlemage" series, with the new Xe3 "Celestial" architecture in the works. However, thanks to PCWorld's The Full Nerd podcast, Tom Petersen of Intel confirmed that the Xe3 IP has been finished, and the hardware teams are already working on the next Xe4 "Druid" GPU IP. "Our architects are way ahead of us, and they are already working on not the next thing but the next thing after the next thing," said Petersen, adding: "The way I would like to comment is our IP that's kind of called Xe3, which is the one after Xe2, that's pretty much baked, right. And so the software teams have a lot of work to do on Xe3. The hardware teams are off on the next thing, right. That's our cadence, that we need to keep going."

The base IP of next-generation Xe3 "Celestial" GPUs is done. That means the basic media engines, Xe cores, XMX matrix engines, ray tracing engines, and other parts of the gaming GPU are already designed and most likely awaiting trial fabrication. The software to support this Xe3 is also being developed while Intel's team is working on enabling more optimizations for the Xe2 "Battlemage" architecture, which we previewed recently. We assume that Intel's Xe GPU will now follow a stricter cadence of releases, with SKUs getting updated much faster, given that a lot is prepared for the future.

It's been public knowledge for a while now that Lenovo is planning an imminent successor to its Legion Go handheld that has proven rather popular among handheld gamers. Previous leaks and rumors indicated that the Legion Go S 8ARP1, as it will apparently be named, will be a more affordable version of the current Legion Go. Now, thanks to Roland Quandt, Windows Central, and WinFuture, more details about the upcoming Legion Go S have leaked, including images of the device, supposed specifications, and a potential price.

According to the leaks, the new affordable handheld gaming PC will feature some substantial hardware changes, including a slightly smaller eight-inch display, this time with a much lower 1920 × 1200p resolution and a slightly lower 120 Hz refresh rate. Gone, too, are the Nintendo Switch-style detachable controllers, with the Legion Go S instead featuring a white unibody design. What's more interesting than the leaked images of the Legion Go S or the hardware changes—detachable controllers or not, the Legion Go is still intended to be used as a handheld—is the new AMD APU that will seemingly power the Go S. The as-yet unannounced AMD Ryzen Z2G looks like it will be an odd core configuration featuring an AMD Radeon 680M iGPU and Zen 3+ cores. Ultimately, the APU seems like it will put the Legion Go S somewhere between the current-generation Legion Go and devices featuring the AMD Ryzen Z1 (non-extreme), which is a good place to be if Lenovo hopes to compete with the likes of the Steam Deck OLED, which will seemingly cost around the same as the Legion Go S, depending on which region you are in.

SPARKLE Intel Arc B-Series Graphics Card offer offers high-resolution gaming with Intel XeSS AI upscaling, ray tracing, and 8K media support, plus accelerated AI features for enhanced creation and editing through Intel AI Playground.

SPARKLE, an Intel official AIB partner, is announcing:

SPARKLE Intel Arc B580 TITAN OC - with Limited GPU Holder
The SPARKLE Intel Arc B580 TITAN OC debuts with 12 GB GDDR6 memory and advanced TORN Cooling 2.0 featuring triple AXL fans, a 2.2-slot design, and a full-metal backplate. With a boost clock of 2740 MHz and 200 W power consumption, it delivers top-tier gaming performance.
A blue breathing light effect adds elegance, while a limited SPARKLE GPU Holder completes this powerhouse package. The TITAN Series continues its legacy of performance and style.

Acer today announced an expansion to its gaming portfolio with the new Nitro Intel Arc B-Series graphics cards, aimed at DIY gamers seeking high-performance gaming and content creation upgrades for their PC setups.

The Nitro Intel Arc B570 OC 10 GB and Nitro Intel Arc B580 OC 12 GB graphics cards, with clock speeds up to 2,740 MHz and up to 12 GB GDDR6 memory, offer gamers an immersive experience and access to the latest AI technologies via the Intel AI Playground application. These graphics cards are equipped with Acer's advanced FrostBlade cooling systems to ensure peak performance.

Today, Intel announced the new Intel Arc B-Series graphics cards (code-named Battlemage). The Intel Arc B580 and B570 GPUs offer best-in-class value for performance at price points that are accessible to most gamers, deliver modern gaming features and are engineered to accelerate AI workloads. The included Intel Xe Matrix Extensions (XMX) AI engines power the newly introduced XeSS 2, comprised of three technologies that together increase performance, visual fluidity and responsiveness.

"The new Intel Arc B-Series GPUs are the perfect upgrades for gamers. They deliver leading performance-per-dollar and great 1440p gaming experiences with XeSS 2, second-generation ray tracing engines and XMX AI engines. We're delighted to be joined by more partners than ever so that gamers have more choice in finding their perfect design." -Vivian Lien, Intel vice president and general manager of Client Graphics[Editor's note: Our preview of the Arc Battlemage Series is now live]

ASRock, the global leading manufacturer of motherboards, graphics cards, mini PCs, gaming monitors and power supply units, today launches the all-new Intel Arc B-Series graphics cards, including Steel Legend and Challenger products based on Intel Arc B580 and Intel Arc B570 Graphics, which are born to shine your PC builds!

Based on the latest Xe2-HPG architecture, Intel Arc B-Series GPUs are designed for high performance gaming at 1440p and 1080p with AI upscaling and ray tracing. They are equipped with cutting-edge features: Intel Xe Super Sampling technology (XeSS), which takes your gaming experience to the next level with AI-enhanced upscaling for higher performance and high image fidelity. Intel XeSS Frame Generation and Intel Xe Low Latency technologies make your games play smoother and more responsive. Intel Xe Matrix eXtensions (XMX) AI engines accelerate AI-enhanced gaming, creation, and media generation. The Advanced Media Engine accelerates your content creation with two full-featured media transcoders to speed up the media exporting across the most popular formats including AV1.

Steam's monthly hardware and software surveys provide a decent picture of what hardware and software gamers rely on to play their favorite games—at least those on Valve's game platform. Since the launch of Windows 11, it has been a somewhat reliable way to track the adoption of the new Windows version, and, as the official cut-off for Windows 10 support draws near, one would expect Windows 11 to pick up steam, especially among gamers, where Windows is the dominant OS. The results of the November Steam Survey are in, and while not much has changed on the hardware front, it seems like Microsoft is indeed wearing gamers down when it comes to Windows 11 adoption. Despite seeing a decent uptick in Windows 11 installations, the overall Windows market share dropped, even if almost imperceptibly, while Linux and macOS both saw a slight uptick in adoption among Steam gamers. As expected, Windows remained the dominant platform for gamers, but Windows 11, specifically, saw growth of 4.18%, while Windows 10 lost 4.15%, which is almost an exact 1:1 match, indicating that gamers are largely staying on Windows when they finally decide to move on from Windows 10. Overall, Windows lost 0.05% market share, compared to Linux, which gained 0.03% and macOS, which grew by 0.02%.

Valve's SteamOS Holo was the most popular Linux version in the survey, but it, too, slid by 0.28%. Of course, the hardware split for Linux is representative of the software side of things, which is to say: It's mostly just Steam Decks. As expected, most of the video cards and CPUs in the Linux results were AMD GPUs, with well over 36% of the sampled Linux gamers using AMD GPUs, even disregarding the obvious bias introduced by the AMD-powered Steam Deck hardware. The most popular NVIDIA GPU on Linux systems running Steam is currently the GeForce RTX 3060, at a mere 1.46% of the market share. Meanwhile, on Windows side, 5.03% of gamers are using the GTX 3060, with the next most popular GPU being the NVIDIA GeForce RTX 4060 Laptop GPU, at 4.92%.

Intel's upcoming Arc "Battlemage" graphics card lineup has been exposed through a recent ASRock specification sheet leak, showcasing the company's latest products for the discrete GPU market. The leak details two models, the B580 and B570, with the B570 making a first appearance in the rumor section. The B580, positioned as the flagship model we already covered, features 20 Xe2-Cores and comes equipped with 12 GB of GDDR6 memory on a 192-bit interface, capable of reaching bandwidth speeds of up to 456 GB/s. Its slightly lower-spec sibling, the B570, sports 18 Xe2-Cores and 10 GB of GDDR6 memory on a 160-bit interface, delivering 380 GB/s bandwidth. In ASRock's case, both cards will be factory-overclocked, with the B580 reaching speeds of 2.8 GHz and the B570 hitting 2.6 GHz.

The new graphics cards are designed to operate on a PCIe 4.0 x8 interface. Both models will support modern display standards, including DisplayPort 2.1 and HDMI 2.1. Intel has scheduled the official unveiling of the Battlemage series for December 3, with cards expected to hit shelves on December 12. While the B570's pricing remains under wraps, the B580's rumored $249 price tag suggests Intel is making a serious play for the mid-range market segment. This aggressive pricing strategy, combined with the card's promising specifications, indicates Intel's determination to establish itself as a legitimate mid-range competitor in the discrete GPU segment, which NVIDIA and AMD have long dominated.

New rumors emerged via Notebookcheck that point to Intel's intention to continue developing discrete desktop GPUs after the imminent launch of its Arc Battlemage series. Leaker JayKihn confirmed in a comment on X that Intel is actively working on future GPU generations, namely the Arc Celestial and Arc Druid, unlike previous rumors that pointed to a possible cancellation if Battlemage underperforms. Sources say Intel is still committed to its GPU roadmap. With the Arc Battlemage series led by the B580 model, Intel is targeting the budget segment, going in direct competition with NVIDIA's RTX 4060 and RTX 4060 Ti. Price-wise, we can expect graphic cards with Intel Arc Battlemage to go around $250 (for the 12 GB model) and although the performance will not be groundbreaking, it can attract interest from buyers on a tight budget.

Since Intel has reportedly canceled all plans to launch discrete laptop Battlemage cards, Arc Druid and Arc Celestial could follow the same path. Although details regarding Arc Celestial and Arc Druid are scattered, confirmation of their development is good news for the PC graphics card market. What we do know now is that Arc Celestial models will arrive in 2025 or early 2026, which could coincide with the launch of Intel's Panther Lake CPUs. The GPUs are expected to take advantage of the new Xe3 graphics architecture (Arc B580 will feature Intel's Xe2-HPG architecture). However, given Intel's latest challenges, the ultimate success of these next-generation GPUs remains to be seen, while we still believe that the success of Arc Battlemage will be decisive for future Intel GPU development.

HWiNFO has been updated to version v8.16, which brings several new improvements and fixes, as well as adds enhanced support for next-generation AMD CPUs and APUs, as well as next-generation Intel GPUs. While it does not specify any models, it is clear that it gets support for AMD's upcoming Kraken Point and Strix Halo APUs, as well as Intel's upcoming Battlemage GPUs.

In addition, the new HWiNFO v8.16 also improves support for Intel Lunar Lake, adds reporting of number of NPU tiles, adds support for Cybenetics Powenetics PMD (Power Measurements Device), enhances sensor monitoring on MSI's Z890, B860, and H810 series motherboards, enhances reporting of CUDIMM CKD parameters, NGU clock, and drive letter, and improves reporting of V/F curve settings, among other things. You can check out the full release notes below.

During NVIDIA's recent Q3 earnings call, CFO Colette Kress cautioned about potential GPU supply constraints in the fourth quarter despite strong gaming sector performance. The gaming division posted impressive results, with $3.2 billion in revenue, representing a 15% increase from the previous year. However, Kress indicated that fourth-quarter gaming revenue might see a decline due to supply limitations, though she reassured that supply should stabilize in early 2025. The company is scaling back RTX 40-series production as it prepares for the anticipated launch of its next-generation Blackwell architecture, which is expected to debut at CES 2025. The RTX 50-series GPU lineup, particularly the flagship RTX 5090 and RTX 5080 models, is rumored to be unveiled during the January event.

"Gaming, although sell-through was strong in Q3, we expect fourth-quarter revenue to decline sequentially due to supply constraints." For consumers, this could mean limited availability and higher prices for gaming GPUs during the holiday shopping season. The shortage is expected to primarily affect RTX 40-series cards, with a particular impact on laptop GPU availability. However, NVIDIA plans to continue producing select RTX 40 mobile chips alongside the upcoming RTX 50 series, suggesting a slow transition between generations. The holiday season is upon us, so this shortage of current-gen models could cost the company some additional customers, as the customer spending usually holds until holidays and holiday discounts.

For over half a century, general-purpose processors have been built on the Tomasulo algorithm, developed by IBM engineer Robert Tomasulo in 1967. It's a $500B industry built on specialized CPU, GPU and other chips for different computing tasks. Hardware startup Ubitium has shattered this paradigm with a breakthrough universal processor that handles all computing workloads on a single, efficient chip - unlocking simpler, smarter, and more cost-effective devices across industries - while revolutionizing a 57-year-old industry standard.

Alongside this, Ubitium is announcing a $3.7 million in seed funding round, co-led by Runa Capital, Inflection, and KBC Focus Fund. The investment will be used to develop the first prototypes and prepare initial development kits for customers, with the first chips planned for 2026.

Thermaltake, a leading PC DIY brand for premium hardware solutions, launches The Tower 250 Mini Tower Chassis, the latest Mini-ITX case in The Tower Series, featuring incredible hardware compatibility, an iconic vertical body design with an octagonal prism shape, and LCD customization. Available in Black, Snow, Matcha Green and Hydrangea Blue finishes, The Tower 250 is built for optimized cooling performance and enhanced capabilities.

Compatible with a wide range of hardware gadgets, The Tower 250 can house a 360 mm high-end GPU, a 360 mm AIO radiator and a standard ATX or SFX power supply up to 200 mm, along with eight 120 mm or five 140 mm fans. Additionally, this Mini-ITX case excels in thermal efficiency with two CT120 fans pre-installed on the top of the case, and room for a 360 mm/280 mm all-in-one CPU cooler on the right. Behind the motherboard tray, two 2.5" SSDs or one 3.5" HDD can be installed, offering maximum storage support.

Intel has prepared "Jaguar Shores," its "next-next" generation AI and HPC accelerator, successor to its upcoming "Falcon Shores" GPU. Revealed during a technical workshop at the SC2024 conference, the chip was unveiled by Intel's Habana Labs division, albeit unintentionally. This announcement positions Jaguar Shores as the successor to Falcon Shores, which is scheduled to launch next year. While details about Jaguar Shores remain sparse, its designation suggests it could be a general-purpose GPU (GPGPU) aimed at both AI training, inferencing, and HPC tasks. Intel's strategy aligns with its push to incorporate advanced manufacturing nodes, such as the 18A process featuring RibbonFET and backside power delivery, which promise significant efficiency gains, so we can expect to see upcoming AI accelerators incorporating these technologies.

Intel's AI chip lineup has faced numerous challenges, including shifting plans for Falcon Shores, which has transitioned from a CPU-GPU hybrid to a standalone GPU, and cancellation of Ponte Vecchio. Despite financial constraints and job cuts, Intel has maintained its focus on developing cutting-edge AI solutions. "We continuously evaluate our roadmap to ensure it aligns with the evolving needs of our customers. While we don't have any new updates to share, we are committed to providing superior enterprise AI solutions across our CPU and accelerator/GPU portfolio." an Intel spokesperson stated. The announcement of Jaguar Shores shows Intel's determination to remain competitive. However, the company faces steep competition. NVIDIA and AMD continue to set benchmarks with performant designs, while Intel has struggled to capture a significant share of the AI training market. The company's Gaudi lineup ends with third generation, and Gaudi IP will get integrated into Falcon Shores.

More than 600 games and applications feature RTX technologies, and each week new games integrating NVIDIA DLSS, NVIDIA Reflex and advanced ray-traced effects are released or announced, delivering the definitive PC experience for GeForce RTX players. This week, Industry Giant 4.0, Microsoft Flight Simulator 2024 and S.T.A.L.K.E.R. 2: Heart of Chornobyl all launch with day-one DLSS 3 support, LEGO Horizon Adventures is out now with DLSS 3, and Proton users can now use DLSS 3 Frame Generation on Linux to accelerate performance in Proton-compatible games.

S.T.A.L.K.E.R. 2: Heart of Chornobyl Launches November 20th with DLSS 3 & Reflex
GSC Game World's S.T.A.L.K.E.R. 2: Heart of Chornobyl is a brand-new entry in the legendary series, enjoyed by millions of players worldwide. The unique combination of first-person shooter, immersive sim, and horror is back. With unprecedented scale, advanced graphics, freedom of choices, and the thickest atmosphere of a deadly adventure, it's going to be the ultimate S.T.A.L.K.E.R. experience.

NVIDIA and Microsoft today unveiled product integrations designed to advance full-stack NVIDIA AI development on Microsoft platforms and applications. At Microsoft Ignite, Microsoft announced the launch of the first cloud private preview of the Azure ND GB200 V6 VM series, based on the NVIDIA Blackwell platform. The Azure ND GB200 v6 will be a new AI-optimized virtual machine (VM) series and combines the NVIDIA GB200 NVL72 rack design with NVIDIA Quantum InfiniBand networking.

In addition, Microsoft revealed that Azure Container Apps now supports NVIDIA GPUs, enabling simplified and scalable AI deployment. Plus, the NVIDIA AI platform on Azure includes new reference workflows for industrial AI and an NVIDIA Omniverse Blueprint for creating immersive, AI-powered visuals. At Ignite, NVIDIA also announced multimodal small language models (SLMs) for RTX AI PCs and workstations, enhancing digital human interactions and virtual assistants with greater realism.

Hypertec proudly announces the launch of its latest breakthrough product, the TRIDENT iG series, an immersion-born GPU server line that brings extreme density, sustainability, and performance to the AI and HPC community. Purpose-built for the most demanding AI applications, this cutting-edge server is optimized for generative AI, machine learning (ML), deep learning (DL), large language model (LLM) training, inference, and beyond. With up to six of the latest NVIDIA GPUs in a 2U form factor, a staggering 8 TB of memory with enhanced RDMA capabilities, and groundbreaking density supporting up to 200 GPUs per immersion tank, the TRIDENT iG server line is a game-changer for AI infrastructure.

Additionally, the server's innovative design features a single or dual root complex, enabling greater flexibility and efficiency for GPU usage in complex workloads.

Since its introduction, the NVIDIA Hopper architecture has transformed the AI and high-performance computing (HPC) landscape, helping enterprises, researchers and developers tackle the world's most complex challenges with higher performance and greater energy efficiency. During the Supercomputing 2024 conference, NVIDIA announced the availability of the NVIDIA H200 NVL PCIe GPU - the latest addition to the Hopper family. H200 NVL is ideal for organizations with data centers looking for lower-power, air-cooled enterprise rack designs with flexible configurations to deliver acceleration for every AI and HPC workload, regardless of size.

According to a recent survey, roughly 70% of enterprise racks are 20kW and below and use air cooling. This makes PCIe GPUs essential, as they provide granularity of node deployment, whether using one, two, four or eight GPUs - enabling data centers to pack more computing power into smaller spaces. Companies can then use their existing racks and select the number of GPUs that best suits their needs. Enterprises can use H200 NVL to accelerate AI and HPC applications, while also improving energy efficiency through reduced power consumption. With a 1.5x memory increase and 1.2x bandwidth increase over NVIDIA H100 NVL, companies can use H200 NVL to fine-tune LLMs within a few hours and deliver up to 1.7x faster inference performance. For HPC workloads, performance is boosted up to 1.3x over H100 NVL and 2.5x over the NVIDIA Ampere architecture generation.

According to The Information, NVIDIA's latest "Blackwell" processors are reportedly encountering significant thermal management issues in high-density server configurations, potentially affecting deployment timelines for major tech companies. The challenges emerge specifically in NVL72 GB200 racks housing 72 GB200 processors, which can consume up to 120 kilowatts of power per rack, weighting a "mere" 3,000 pounds (or about 1.5 tons). These thermal concerns have prompted NVIDIA to revisit and modify its server rack designs multiple times to prevent performance degradation and potential hardware damage. Hyperscalers like Google, Meta, and Microsoft, who rely heavily on NVIDIA GPUs for training their advanced language models, have allegedly expressed concerns about possible delays in their data center deployment schedules.

The thermal management issues follow earlier setbacks related to a design flaw in the Blackwell production process. The problem stemmed from the complex CoWoS-L packaging technology, which connects dual chiplets using RDL interposer and LSI bridges. Thermal expansion mismatches between various components led to warping issues, requiring modifications to the GPU's metal layers and bump structures. A company spokesperson characterized these modifications as part of the standard development process, noting that a new photomask resolved this issue. The Information states that mass production of the revised Blackwell GPUs began in late October, with shipments expected to commence in late January. However, these timelines are unconfirmed by NVIDIA, and some server makers like Dell confirmed that these GB200 NVL72 liquid-cooled systems are shipping now, not in January, with CoreWave GPU cloud provider as a customer. The original report could be using older information, as Dell is one of NVIDIA's most significant partners and among the first in the supply chain to gain access to new GPU batches.

GIGABYTE TECHNOLOGY Co. Ltd, a leading manufacturer of premium gaming hardware, today launched the cutting-edge GIGABYTE AMD Radeon PRO W7800 AI TOP 48G. GIGABYTE has taken a significant leap forward with the release of the Radeon PRO W7800 AI TOP 48G graphics card, featuring AMD's RDNA 3 architecture and a massive 48 GB of GDDR6 memory. This significant increase in memory capacity, compared to its predecessor, provides workstation professionals, creators, and AI developers with incredible computational power to effortlessly handle complex design, rendering, and AI model training tasks.

⁠GIGABYTE stands as the AMD professional graphics partner in the market, with a proven ability to design and manufacture the entire Radeon PRO series. Our dedication to quality products, unwavering business commitment, and comprehensive customer service empower us to deliver professional-grade GPU solutions, expanding user's choices in workstation and AI computing.⁠

We know that NVIDIA's latest "Blackwell" GPUs are fast, but how much faster are they over the previous generation "Hopper"? Thanks to the latest MLPerf Training v4.1 results, NVIDIA's HGX B200 Blackwell platform has demonstrated massive performance gains, measuring up to 2.2x improvement per GPU compared to its HGX H200 Hopper. The latest results, verified by MLCommons, reveal impressive achievements in large language model (LLM) training. The Blackwell architecture, featuring HBM3e high-bandwidth memory and fifth-generation NVLink interconnect technology, achieved double the performance per GPU for GPT-3 pre-training and a 2.2x boost for Llama 2 70B fine-tuning compared to the previous Hopper generation. Each benchmark system incorporated eight Blackwell GPUs operating at a 1,000 W TDP, connected via NVLink Switch for scale-up.

The network infrastructure utilized NVIDIA ConnectX-7 SuperNICs and Quantum-2 InfiniBand switches, enabling high-speed node-to-node communication for distributed training workloads. While previous Hopper-based systems required 256 GPUs to optimize performance for the GPT-3 175B benchmark, Blackwell accomplished the same task with just 64 GPUs, leveraging its larger HBM3e memory capacity and bandwidth. One thing to look out for is the upcoming GB200 NVL72 system, which promises even more significant gains past the 2.2x. It features expanded NVLink domains, higher memory bandwidth, and tight integration with NVIDIA Grace CPUs, complemented by ConnectX-8 SuperNIC and Quantum-X800 switch technologies. With faster switching and better data movement with Grace-Blackwell integration, we could see even more software optimization from NVIDIA to push the performance envelope.

With the latest industry boom in AI, the demand for more compute power is greater than ever, and the recent industry forecast predicts that the global GPU market will exceed $98.5 billion in value by the year 2024. This staggering projection, outlined in the 2024 supply-side GPU market summary report by Jon Peddie Research (JPR), shows how far the GPU market has come. Once primarily associated with powering consumer gaming rigs with AMD or NVIDIA inside, GPUs have become a key part of our modern tech stack, worth almost $100 billion in 2024 alone. Nowadays, GPUs are found in many products, from smartphones and vehicles to internet-connected devices and data centers.

"Graphics processor units (GPUs) have become ubiquitous and can be found in almost every industrial, scientific, commercial, and consumer product made today," said Dr. Jon Peddie, founder of JPR. "Some market segments, like AI, have grabbed headlines because of their rapid growth and high average selling price (ASP), but they are low-volume compared to other market segments." The report also shows the wide range of companies that are actively participating in the GPU marketplace, including industry giants like AMD, NVIDIA, and Intel, as well as smaller players from China like Loongson Zhongke, Siroyw, and Lingjiu Micro. Besides the discrete GPU solutions, the GPU IP market is very competitive, and millions of chips are shipped with GPU IP every year. Some revenue estimates of Chinese companies are not public, but JPR is measuring it from the supply chain side, so these estimates are pretty plausible.

Today, Sony Interactive Entertainment expands the PlayStation 5 (PS5) family of products with the release of the new PlayStation 5 Pro (PS5 Pro) console - the company's most advanced and innovative gaming console to date. PlayStation 5 Pro was designed with deeply engaged players and game creators in mind and includes key performance features that allow games to run with higher fidelity graphics at smoother frame rates.

"With PlayStation 5 Pro, we wanted to make sure that the most dedicated gamers, as well as game creators, could utilize the most advanced console technology, taking the PlayStation 5 experience even farther," said Hideaki Nishino, CEO Platform Business Group, Sony Interactive Entertainment. "This is our most advanced PlayStation to date, and it gives our community of players the opportunity to experience games the way that developers intended for them to be. Players will be thrilled with how this console enhances some of their favorite titles, while opening avenues to discover new ones."