GALAX is known to go overboard with its top flight graphic card models—the introduction of GeForce RTX 40 SUPER models has further lengthened the manufacturer's naming conventions. A Hall of Fame (HOF) OC LAB Master Edition card based on NVIDIA's freshly launched GeForce RTX 4070 SUPER GPU has been introduced via a Galaxy BBS blog post. As befits such a fancily named card, GALAX has rolled out the red carpet with a very special cooling solution that is designed to temper a substantial (211 MHz) overclock over Team Green's reference settings. Its 2685 MHz spec sits atop the customized GeForce RTX 4070 SUPER GPU pile—VideoCardz notes that another high-end option—GIGABYTE's AORUS RTX 4070 SUPER MASTER model—trails by a 30 MHz margin.

The GALAX GeForce RTX 4070 SUPER Hall of Fame OC LAB Master Edition graphics card sports an almost all-white design, even its PCB is outfitted in a pale hue. This Ada Lovelace AD104-350-A1-driven flagship is specced with a 250 W TGP as standard, user-adjustable up to 320 W. A three-pin to sixteen-pin adapter is supplied by default; ensuring that more than enough juice is supplied. GALAX states that the card features a 12+3 phase design, coupled with a power section is controlled by an XDPE10281 PWM. The Chinese manufacturer hints that more white Master Edition SUPER models are incoming (see below). We hope to see further announcements, and full product pages uploaded to their web site(s), but GALAX is unlikely to sell these top flight cards outside of their native market.

According to new data from Chinese Customs, China's imports of integrated circuits suffered their steepest annual drop on record in 2023, falling 15.4% to $349.4 billion. The decline marks the second straight year of falling chip imports and can be attributed to economic uncertainty and US export controls on advanced semiconductors. Shipment volumes of imported chips also saw a substantial 10.8% year-over-year decrease as demand within China stagnated. The country's important tech manufacturing sector has struggled under strict zero-Covid policies and a lackluster recovery post-pandemic. Flagship manufacturing companies like TSMC recorded modest declines in 2023 sales, though TSMC still forecasts overall growth this year.

Sentiment plunged further when the Biden administration heightened restrictions on China's access to cutting-edge AI-capable chips from NVIDIA and other top American suppliers. The escalating US export controls have choked off China's pipeline to advanced semiconductors needed for AI and supercomputing applications. However, early positive signs for global semiconductor demand have emerged, with worldwide chip sales rising for the first time in over a year this past November. The increase was driven by growing demand for AI and other emerging technologies that rely on advanced computing chips. While the US seeks to limit China's progress in this key strategic area, an inflection point for the battered global chip sector may be nearing.

Intel has reportedly rolled out its China market-exclusive Core i7-14790F "Black Edition" CPU—MEGAsizeGPU tweeted out an initial glimpse of the decidedly not-very-Team-Blue outer packaging. A Core i7-14700F desktop processor was included in an announced list at CES 2024 and launched via global retail channels a few days ago, but its somewhat similarly monikered "Black Edition" appears to be region specific (also notably absent from last week's trade show). Intel's Core i7-14790F's nomenclature would suggest that its sits just a little bit above the i7-14700F in Raptor Lake Refresh's hierarchy, but the "Black Edition" specifications paint an odd picture. As noted by Tom's Hardware: "This CPU has an odd configuration since it has fewer cores and more cache than the Core i7-14700F, contrary to what its SKU would imply to be a higher-spec variant. "

The Core i7-14790K is currently priced at 2909 RMB (on JD.com) which converts to ~$410 (USD). The Core i7-14700F's asking price is roughly ten dollars more, which nets you a 16-core processor that packs 8 performance cores, 12 efficiency cores and 28 threads. Chinese buyers are presented with the slightly cheaper Core i7-14790F and its curiosity specs: 16 total cores comprised of 8 performance cores, 8 efficiency cores and 24 threads. Both F-models share the same list of clock speeds, but the Black Edition does slightly better with an overall total of 36 MB Intel Smart Cache (versus 33 MB)—in turn it loses out in L2 cache designation (24 vs. 28 MB). Tom's Hardware has kindly assembled a comparison table (see below).

The first review of a Chinese-exclusive "RTX 4090D" GPU model hit the internet last week—Expreview received a sample GALAX RTX 4090 D Metal Master model not long ago, and their testing team proceeded to find out whether the nerfed version of NVIDIA's flagship gaming GPU was truly compromised in terms of performance. Effective October 2023, the US Federal Trade Commission placed restrictions on Team Green—thus blocking trade of units based on the "Ada Lovelace" AD102-300 GPU in China. In turn, a variant—AD102-250-A1—was prepared in order to confirm to new policies.

NVIDIA's China-specific GeForce RTX 4090D launched officially right at the end of 2023. Board partner GALAX seems to be leading the pack, with customized versions being sent out for evaluation. The GeForce RTX 4090D GPU arrives with a lesser configuration: 14,592 CUDA, 456 Tensor, and 114 RT cores—but the first review indicates that this only trails behind its uncompromised sibling by roughly 5 to 6% across sixteen games. It lags behind in Stable Diffusion benchmarks—an AI workload at 512x512 resolution shows a 10% difference, although the gaps narrows at 768x768 and 1024x1024.

According to the latest teardown from TechInsights, China's biggest technology maker, Huawei, has been shipping laptops with technology supposedly sanctioned by the United States. As the teardown shows, TechInisights has discovered that Huawei's Kirin 9006C processor is manufactured on TSMC's 5 nm semiconductor technology. Originally, the United States have imposed sanctions on Huawei back in 2020, when the government cut off Huawei's access from TSMC's advanced facilities and forbade the use of the latest nodes by Huawei's HiSilicon chip design arm. Today's findings show signs of contradiction, as the Qingyun L540 notebook that launched in December 2023 employs a Kirin 9006C chipset manufactured on a TSMC 5 nm node.

TechInsight's findings indicate that Kirin 9006C assembly and packaging occurred around the third quarter of 2020, whereas the 2020 Huawei sanctions started in the second quarter. Of course, the implication of the sanctions likely prohibited any new orders and didn't prevent Huawei from possibly stockpiling millions of chip orders in its warehouse before they took place. The Chinese giant probably made orders beforehand and is using the technology only now, with the Qingyun L540 laptop being one of the first Kirin 9006C appearances. Some online retailers also point out that the laptop complies with the latest security practices required for the government, which means that they have been in the works since the chip began the early stages of design, way before 2020. We don't know the stockpile quantity, but SMIC's domestic efforts seem insufficient to supply the Chinese market alone. The news that Huawei is still using TSMC chips made SMIC's share go for a 2% free fall on the Hong Kong stock exchange.

According to the report from a journal called Fundamental Research, researchers from the Institute of Computing Technology at the Chinese Academy of Sciences have developed a 256-core multi-chiplet processor called Zhejiang Big Chip, with plans to scale up to 1,600 cores by utilizing an entire wafer. As transistor density gains slow, alternatives like multi-chiplet architectures become crucial for continued performance growth. The Zhejiang chip combines 16 chiplets, each holding 16 RISC-V cores, interconnected via network-on-chip. This design can theoretically expand to 100 chiplets and 1,600 cores on an advanced 2.5D packaging interposer. While multi-chiplet is common today, using the whole wafer for one system would match Cerebras' breakthrough approach. Built on 22 nm process technology, the researchers cite exascale supercomputing as an ideal application for massively parallel multi-chiplet architectures.

Careful software optimization is required to balance workloads across the system hierarchy. Integrating near-memory processing and 3D stacking could further optimize efficiency. The paper explores lithography and packaging limits, proposing hierarchical chiplet systems as a flexible path to future computing scale. While yield and cooling challenges need further work, the 256-core foundation demonstrates the potential of modular designs as an alternative to monolithic integration. China's focus mirrors multiple initiatives from American giants like AMD and Intel for data center CPUs. But national semiconductor ambitions add urgency to prove domestically designed solutions can rival foreign innovation. Although performance details are unclear, the rapid progress shows promise in mastering modular chip integration. Combined with improving domestic nodes like the 7 nm one from SMIC, China could easily create a viable Exascale system in-house.

Global semiconductor capacity is expected to increase 6.4% in 2024 to top the 30 million *wafers per month (wpm) mark for the first time after rising 5.5% to 29.6 wpm in 2023, SEMI announced today in its latest quarterly World Fab Forecast report.

The 2024 growth will be driven by capacity increases in leading-edge logic and foundry, applications including generative AI and high-performance computing (HPC), and the recovery in end-demand for chips. The capacity expansion slowed in 2023 due to softening semiconductor market demand and the resulting inventory correction.

According to the latest round of reports, AMD has decided not to include China in the initial global launch of its upcoming Radeon RX 7600 XT graphics card. The RX 7600 XT, featuring 16 GB of memory and based on AMD's next-generation RDNA 3 architecture, was expected to launch soon at a price of around $300. However, the company is currently re-evaluating its Chinese GPU launch strategy due to the runaway success of its existing Radeon RX 6750 Golden Rabbit Edition (GRE) series in the region. The RX 6750 GRE cards with 10 GB and 12 GB configurations retail between $269-$289 in China, offering exceptional value compared to rival NVIDIA RTX models. AMD seems hesitant to risk undercutting sales of its popular RX 6750 GPUs by launching the newer 7600 XT.

While the RX 7600 XT promises more raw performance thanks to advanced RDNA 3 architecture, 6750 GRE, with its RDNA 2 design, seemingly remains efficient enough for most Chinese mainstream gamers. With the RX 6750 GRE still selling strongly in China, AMD has postponed the RX 7600 XT introduction for this key market. Final launch timelines for the 7600 XT in China and globally remain unconfirmed by AMD at time of writing. The company appears to be treading cautiously amidst the shifting competitive landscape.

NVIDIA this October was barred by the US Federal Trade Commission from selling the GeForce RTX 4090 in the Chinese domestic market. The company scampered to create a China-specific variant of the RTX 4090 that would comply with the FTC's limits on the AI inference performance, while mostly being as fast as the regular RTX 4090 at gaming. The resulting product is the RTX 4090D. This card has significantly lower AI inference performance, probably due to a lower Tensor core count as VideoCardz reports, or a firmware based performance limiter similar to LHR (lite hashrate) GPUs during the peak of the crypto-scalping GPU shortage. This card goes on sale today. To prevent modding of this GPU back to a regular RTX 4090, it probably comes with a different device ID, and other barriers that prevent video BIOS from regular RTX 4090 from being transplanted over; and overclocking of the GPU is disabled beyond the maximum boost frequency advertised. The TGP has been lowered to 425 W, too.

The Chinese Government regulators overseeing the country's video game industry approved 105 new game titles since last week's curbs that prevent online games from rewarding daily gameplay. The scope of the curbs were so profound that they sent company stocks of the country's two leading game publishers, Tencent and NetEase, crashing down by approximately 16% and 24%, respectively, with tens of billions in investor value erased. China's domestic online gaming revenues were last assessed at around $42 billion a year.

The Press and Publications Administration (PPA), the overarching regulatory body behind all information and content creation and circulation in China, in its official WeChat account, commended the Game Working Committee of China Music and Digital Association, the body that approves publication of new games, for the successful implementation of the new curbs, and the approval of the 105 new games, stating that the approvals "send positive signals that support the prosperity and healthy development of the online game industry." At any given time, the Chinese regulators have a queue of games and their content additions (eg: DLCs) awaiting government approval. 105 is the highest number of monthly approvals passed by the government, and is hence seen as a move aimed at appeasing investors after last week's crash of game publisher stocks.

Intel's software team has added support in its open-source Linux drivers for an unannounced Habana "Gaudi2C" AI accelerator variant. Little is documented about the mystery Gaudi2C, which shares a core identity with Intel's flagship Gaudi2 data center training and inference chip, otherwise broadly available. The new revision is distinguished only by a PCI ID of "3" in the latest patch set for Linux 6.8. Speculations circulate that Gaudi2C may be a version tailored to meet China-specific demands, similar to Intel's Gaudi2 HL-225B SKU launched in July with reduced interconnect links. With US export bans restricting sales of advanced hardware to China, including Intel's leading Gaudi2 products, creating reduced-capability spinoffs that meet export regulations lets Intel maintain crucial Chinese revenue.

Meanwhile, Intel's upstream Linux contributions remain focused on hardening Gaudi/Gaudi2 support, now considered "very stable" by lead driver developer Oded Gabbay. Minor new additions reflect maturity, not instability. The open-sourced foundations contrast NVIDIA's proprietary driver model, a key Intel competitive argument for service developers using Habana Labs hardware. With the SynapseAI software suite reaching stability, some enterprises could consider Gaudi accelerators as an alternative to NVIDIA. And with Gaudi3 arriving next year, the ecosystem will get a better competitive advantage with increased performance targets.

Chinese chipmaker Montage Technology has unveiled new data center processors under its Jintide brand based on Intel's latest Emerald Rapids Xeon architecture. The 5th generation Jintide lineup offers anywhere from 16-core to 48-core options for enterprise customers needing advanced security specific to China's government and enterprise requirements. Leveraging a long-running joint venture with Intel, Jintide combines standard high-performance Xeon microarchitectures with added on-die monitoring and encryption blocks, PrC (Pre-check) and DSC (Dynamic Security Check), which are security-hardened for sensitive Chinese use cases. The processors retain all core performance attributes of Intel's vanilla offerings thanks to IP access, only with extra protections mandated by national security interests. While missing the very highest core counts, the new Jintide chips otherwise deliver similar Emerald Rapids features like 8-channel DDR5-5600 memory, 80 lanes of speedy PCIe 5.0, and elevated clock speeds over 4.0 GHz at peak. The Jintide processors have 2S scaling, which allows for dual-socket systems with up to 96 cores and 192 threads.

Pricing remains unpublished but likely carries a premium over Intel list prices thanks to the localized security customization required. However, with Jintide uniquely meeting strict Chinese government and data regulations, cost becomes secondary for target customers needing compliant data center hardware. After matching lockstep with Intel's last several leading Xeon generations, Jintide's continued iteration highlights its strategic value in enabling high-performance domestic infrastructure as China eyes IT supply chain autonomy. Intel gets expanded access to the growing Chinese server market, while Chinese partners utilize Intel IP to strengthen localized offerings without foreign dependency. It manifests the delicate balance of advanced chip joint ventures between global tech giants and rising challengers. More details about the SKUs are listed in the table below.

After years of delays, Chinese chipmaker Zhaoxin has finally launched its long-awaited KX-7000 series consumer CPUs, only one of its kind in China, based on the licensed x86-64 ISA. Zhaoxin claims the new 8-core processors based on "Century Avenue" uArch deliver double the performance of previous generations. Leveraging architectural improvements and 4X more cache, the KX-7000 represents essential progress for China's domestic semiconductor industry. While still likely lagging behind rival AMD and Intel chips in raw speed, the KX-7000 matches competitive specs in areas like DDR5 memory, PCIe 4.0, and USB4 support. For Chinese efforts to attain technological independence, closing feature gaps with foreign processors is just as crucial as boosting performance. Manufactured on a 16 nm process, the KX-7000 does not use the best silicon node available.

Other chip details include out-of-order execution (OoOE), 24 PCIe 4.0 lanes, a 32 MB pool of L3 cache and 4 MB L2 cache, a base frequency of 3.2 GHz, and a boost clock of 3.7 GHz. Interestingly, the CPU also has VT-x, BT-d 2.5, SSE4.2/AVX/AVX2 support, most likely also licensed from the x86 makers Intel and/or AMD. Ultimately, surpassing Western processors is secondary for China next to attaining self-reliance. Instructions like SM encryption catering to domestic data protection priorities underscore how the KX-7000 advances strategic autonomy goals. With its x86 architecture license giving software compatibility and now a vastly upgraded platform, the KX-7000 will raise China's chip capabilities even if it is still trailing rivals' speeds. Ongoing progress closing that performance gap could position Zhaoxin as a mainstream alternative for local PC builders and buyers.

The United States is softening restrictions on the significant GPU maker NVIDIA, selling artificial intelligence chips to China. While still limiting advanced chip exports deemed strategically threatening, Commerce Secretary Gina Raimondo clarified this week that NVIDIA could supply some AI processors to Chinese commercial companies. Previously, Raimondo had sharply criticized NVIDIA for attempting to sidestep regulations on selling powerful GPUs abroad. Her comments followed rumors that NVIDIA tweaked chip designs to avoid newly imposed export controls narrowly. However, after discussions between Raimondo and NVIDIA CEO Jensen Huang, the Commerce Department says NVIDIA and other US firms will be permitted to export AI chips to China for general commercial use cases. Exports are still banned on the very highest-end GPUs that could enable China to train advanced AI models rivaling American developments.

Raimondo said NVIDIA will collaborate with the US to comply with the export rules. Huang reaffirmed the company's commitment to adherence. The clarification may ease pressures on NVIDIA, as China accounts for up to 25% of its revenue. While optimistic about recent Chinese approvals for US joint ventures, Raimondo noted frustrations linger around technology controls integral to national security. The nuanced recalibration of restrictions illustrates the balances the administration must strike between economic and security interests. As one of the first big US technology exporters impacted by tightened restrictions, NVIDIA's ability to still partly supply the valuable Chinese chip market points to a selective enforcement approach from regulators in the future.

GALAX is a much older brand than it looks, dating back to 2003, when it did business as "Galaxy," selling custom-design GeForce 400 series graphics cards with its signature shade of deep blue for the PCB, and Zalman-sourced coolers. The company is celebrating its 20th anniversary, at least in its largest market of China, where it launched a special edition GeForce RTX 4070 Ada graphics card, called simply the 20th Anniversary Edition.

The card features an all-white look, and an interesting cooler airflow design. The card is no more than 2 slots thick, and features a slender aluminium fin-stack heatsink. Two of its three fans are located on the obverse side of the card. There is an intake where you'd expect the third fan to be, which is actually on the reverse side of the card, blowing outward (pulling air through the heatsink, much like NVIDIA's dual-axial flow-through cooler in some of its Founders Edition cards. The card uses a single 8-pin PCIe power connector, which has been tucked away behind a magnetic cover. The card offers factory overclocked speeds of 2610 MHz maximum boost, compared to 2475 MHz reference (+5.5%). The company didn't reveal pricing.

Huawei's subsidiary HiSilicon has made significant strides in the independent R&D of AI chips, launching the next-gen Ascend 910B. These chips are utilized not only in Huawei's public cloud infrastructure but also sold to other Chinese companies. This year, Baidu ordered over a thousand Ascend 910B chips from Huawei to build approximately 200 AI servers. Additionally, in August, Chinese company iFlytek, in partnership with Huawei, released the "Gemini Star Program," a hardware and software integrated device for exclusive enterprise LLMs, equipped with the Ascend 910B AI acceleration chip, according to TrendForce's research.

TrendForce conjectures that the next-generation Ascend 910B chip is likely manufactured using SMIC's N+2 process. However, the production faces two potential risks. Firstly, as Huawei recently focused on expanding its smartphone business, the N+2 process capacity at SMIC is almost entirely allocated to Huawei's smartphone products, potentially limiting future capacity for AI chips. Secondly, SMIC remains on the Entity List, possibly restricting access to advanced process equipment.

NVIDIA is preparing to launch the GeForce RTX 4090 D, or "Dragon" edition, designed explicitly for China. Circumventing the US export rules of GPUs that could potentially be used for AI acceleration, the GeForce RTX 4090 D is reportedly cutting back on overclocking as a feature. According to BenchLife, the AD102-250 GPU used in the RTX 4090 D will be a stranger to overclocking, as the card will not support it, possibly being disabled by firmware and/or physically in the die. The information from @Zed__Wang suggests that the Dragon version will be running at 2280 MHz base frequency, higher than the 2235 MHz of AD102-300 found in the regular RTX 4090, and 2520 MHz boost, matching the regular version.

Interestingly, the RTX 4090 D for China will also feature a slightly lower Total Graphics Power (TGP) of 425 Watts, down from the 450 Watts of the regular model. With memory configuration appearing to be the same, this new China-specific model will most likely perform within a few percent of the original design. Higher base frequency probably indicates a lack of a few CUDA cores to comply with the US export regulation policy and serve the Chinese GPU market. The NVIDIA GeForce RTX 4090 D is scheduled for rollout in January 2024 in China, which is just a few weeks away.

NVIDIA's GeForce RTX 40 series lineup of graphics cards has been supporting massive cooler designs ranging from three to four slots in thickness, with gamers rarely even getting a standard two-slot solution available. However, GALAX recently announced a novel entry into this lineup: the RTX 4060 Ti Max 16 GB Unparalleled Max, a GPU noted for its unprecedented single-slot design and exceptionally thin 20 mm profile. This model, previously previewed on GALAX's China website, stands out with its unique vapor chamber cooling system paired with a copper heatsink, diverging from the typical multi-fan setups seen in the market. Measuring at 267x111x20 mm, the design is very friendly towards smaller cases with room for only a single slot cooler.

The RTX 4060 Ti Max is set to operate at a default clock speed of 2535 MHz, with a power target of 165 Watts, suggesting a solid performance base for all GPU-intensive sessions. Currently, GALAX has yet to indicate the availability of an 8 GB version or the inclusion of a non-Ti model with this cooler, as only a 16 GB version has been shown. Interestingly, GALAX has made overclocking the card possible; however, the voltage regulation module setup is 6+2 VRMs placed on a six-layer PCB, not providing an ideal overclocking setup. Additionally, while feasible, overclocking the GPU with such a tiny single-slot cooler should be approached cautiously.More images, along with specification table (in Chinese), can be seen below.

NVIDIA is giving final touches to the new GeForce RTX 4090 D, a graphics card SKU specific to the Chinese market, aimed squarely at gamers. The card fills the void for gamers shopping in the enthusiast segment, as all inventory of the regular RTX 4090 has been bought up by Chinese companies to accelerate AI, and controls are in place that prevent NVIDIA from selling the card in its current form in the Chinese market.

What sets this SKU apart is that it is designed to comply with U.S. export controls of GPUs that have the dual use as a high compute-density AI accelerator. In other words, its performance with AI will be artificially limited. This is being done by lowering the card's TPP (total processing performance), which could mean that it ends up slower than the regular RTX 4090. This is somewhat similar in concept to LHR (lite hash rate) GPUs NVIDIA designed for gamers as their regular GPUs were being heaped up by crypto currency miners, although LHR wasn't created due to government policy, but in response to market demand. The RTX 4090 D is expected to retail for RMB 13,000, which is similar to the baseline price of the RTX 4090.

The Chinese chipmaker Loongson has launched its newest desktop processors, the 4-core, 8-thread 3A6000 series, based on the company's LoongArch microarchitecture. We have previously reported that the company wants to match Intel's "Willow Cove" and AMD's Zen 3 instruction per clock (IPC) levels with its 3A6000 CPU series, and today we have the first preview of the performance. Powered by the LA664 cores, 3A6000 is built on a 14/12 nm manufacturing process, with clock speeds going from 2.0 to 2.5 GHz and power consumption of up to 50 Watts. It features 256 KB of L2 cache and 16 MB of L3 cache in total.

While several hardware partners are announcing new Loongson-powered solutions, ASUS China's "Uncle Tony" managed to get his hands on one of them and overclocker the CPU to 2.63 GHz on air cooling. In overclocking tests using liquid nitrogen cooling, a 3A6000 processor reached 3.0 GHz, though there are indications that there is still overhead. In standard out-of-the-box configuration, the 3A6000 performs similarly to Intel's Core i3-10100 four-core CPU, an achievement for Loongson but still behind Intel's latest offerings that clock nearly twice as high. This rapid development of Loongson IP has led to a massive performance increase, matching the IPC of modern CPUs. We are still left to see more information about these 3A6000 series SKUs; however, early benchmarks suggest a significant improvement. You can see the CPU benchmarks below, which include UnixBench and SPEC CPU 2006.

The recent U.S. government restrictions on AI hardware exports to China have significantly impacted several key semiconductor players, including NVIDIA, AMD, and Intel, restricting them from selling high-performance AI chips to Chinese land. This ban has notably affected NVIDIA's GeForce RTX 4090 gaming GPUs, pushing them out of mainland China due to their high computational capabilities. In anticipation of these restrictions, NVIDIA reportedly moved a substantial inventory of its AD102 GPUs and GeForce RTX 4090 graphics cards to China, which we reported earlier. This could have contributed to the global RTX 4090 shortage, driving the prices of these cards up to 2000 USD. In an interesting turn of events, insiders from the Chinese Baidu forums have disclosed that specialized factories across China are repurposing these GPUs, which arrived before the ban, into AI solutions.

This transformation involves disassembling the gaming GPUs, removing the cooling systems and extracting the AD102 GPU and GDDR6X memory from the main PCBs. These components are then re-soldered onto a domestically manufactured "reference" PCB, better suited for AI applications, and equipped with dual-slot blower-style coolers designed for server environments. The third-party coolers that these GPUs come with are 3-4 slots in size, whereas the blower-style cooler is only two slots wide, and many of them can be placed in parallel in an AI server. After rigorous testing, these reconfigured RTX 4090 AI solutions are supplied to Chinese companies running AI workloads. This adaptation process has resulted in an influx of RTX 4090 coolers and bare PCBs into the Chinese reseller market at markedly low prices, given that the primary GPU and memory components have been removed.Below, you can see the dismantling of AIB GPUs before getting turned into blower-style AI server-friendly graphics cards.

AMD's lineup of Radeon and Instinct GPUs, including the flagship RX 7900 XTX/XT, the professional-grade PRO W7900, and the upcoming Instinct MI300, are facing sales prohibitions in China, according to an alleged sales advisory guide from Dell. This restriction mirrors the earlier ban on NVIDIA's RTX 4090, underscoring the increasing export limitations U.S.-based companies face for high-end semiconductor products that could be repurposed for military and strategic applications. Notably, Dell's report lists several AMD Instinct accelerators, which are integral to data center infrastructure, and Radeon GPUs, which are widely used in PCs, indicating the broad impact of the advisory.

The ban includes discrete GPUs like AMD's Radeon RX 7900 XTX and 7900 XT, which, despite their data-center potential, may still be sold under specific "NEC" eligibility. This status allows for continued sales in restricted regions like sales of NVIDIA's RTX 4090. However, the process to secure NEC eligibility is lengthy, potentially leading to supply shortages and increased GPU prices—a trend already observed with the RX 7900 XTX in China, where it's become a high-end alternative in light of the RTX 4090's scarcity and inflated pricing. The Dell sales advisory also lists that sales of the aforementioned products are banned in 22 countries, including Russia, Iran, Iraq, and others listed below.

International Data Corporation (IDC) has upgraded its Semiconductor Market Outlook by calling a bottom and return to growth that accelerates next year. IDC raised its September 2023 revenue outlook from $518.8 billion to $526.5 billion in a new forecast. Revenue expectations for 2024 were also raised from $625.9 billion to $632.8 billion as IDC believes the U.S. market will remain resilient from a demand standpoint and China will begin recovering by the second half of 2024 (2H24).

IDC sees better semiconductor growth visibility as the long inventory correction subsides in two of the largest market segments: PCs and smartphones. Automotive and Industrials elevated inventory levels are expected to return to normal levels in 2H24 as electrification continues to drive semiconductor content over the next decade. Technology and large flagship product introductions will drive more semiconductor content and value across market segments in 2024 through 2026, including the introduction of AI PCs and AI Smartphones next year and a much-needed improvement in memory ASPs and DRAM bit volume.

The 62nd edition of the TOP500 reveals that the Frontier system retains its top spot and is still the only exascale machine on the list. However, five new or upgraded systems have shaken up the Top 10.

Housed at the Oak Ridge National Laboratory (ORNL) in Tennessee, USA, Frontier leads the pack with an HPL score of 1.194 EFlop/s - unchanged from the June 2023 list. Frontier utilizes AMD EPYC 64C 2GHz processors and is based on the latest HPE Cray EX235a architecture. The system has a total of 8,699,904 combined CPU and GPU cores. Additionally, Frontier has an impressive power efficiency rating of 52.59 GFlops/watt and relies on HPE's Slingshot 11 network for data transfer.

Thanks to the TMTPost interview with the Global Vice President and China President of ASML, Shen Bo, the Dutch semiconductor equipment manufacturer has revealed that around 1,400 of its deep ultraviolet (DUV) lithography and metrology machines are currently installed in China. The company is expected to achieve a global output of 600 DUV equipment units by the end of 2025. Shen Bo stated that the company aims to install 500-600 units of DUV machinery in China by late 2025 or early 2026. The growth in ASML's Chinese revenues was notably high, with China contributing 46% of the company's system sales in 3Q 2023, representing an 82% revenue increase from the previous quarter.

China plans to build 25 12-inch wafer fabs in the next five years, covering logic wafers, DRAM, and MEMS production. ASML currently has a substantial presence in China, with 16 offices, 12 warehouses, distribution centers, development centers, training centers, and maintenance centers. The company employs over 1,600 people for its China operations. Despite the export restrictions imposed by the US government, ASML anticipates that the new measures will have little impact on its financial outlook for 2023 as it strives to meet the growing demand for semiconductor manufacturing equipment in the global market.