Tag: NVIDIA

  • Nvidia and Foxconn to Build 10,000-GPU AI Supercomputer in Taiwan

    Nvidia and Foxconn to Build 10,000-GPU AI Supercomputer in Taiwan

    Key Takeaways

    1. Nvidia and Foxconn are partnering to create an “AI factory” supercomputer in southern Taiwan, integrating 10,000 Nvidia Blackwell GPUs.
    2. The project involves an investment of several hundred million US dollars, making it one of the largest AI investments globally, although smaller than Elon Musk’s Memphis Supercluster.
    3. The computing power will be distributed to local researchers, startups, and businesses to boost AI adoption in Taiwan.
    4. The initiative aims to create a larger AI ecosystem, linking public agencies and private enterprises, with goals to accelerate semiconductor innovation and enhance smart-city services.
    5. This supercomputer aligns with Nvidia’s expansion strategy, including plans for a research center in Shanghai and a fleet of AI servers in the U.S. worth half a trillion dollars.


    Nvidia and Foxconn have broadened their long-term partnership with new plans for an “AI factory” supercomputer in southern Taiwan. This system is set to be delivered by Big Innovation Company, a Foxconn subsidiary and Nvidia Cloud Partner, and will integrate 10,000 Nvidia Blackwell GPUs into one setup.

    Investment Overview

    With an investment of several hundred million US dollars, this project is less massive than Elon Musk’s Memphis Supercluster, which has 200,000 GPUs. However, it still stands as one of the largest AI investments globally. Nvidia is responsible for the hardware supply, whereas Foxconn will take care of the datacenter infrastructure and operational support.

    Local Impact

    The National Science and Technology Council of Taiwan plans to distribute the computing power of this cluster to local researchers, startups, and established businesses, thus boosting AI adoption domestically. Taiwan Semiconductor Manufacturing Company (TSMC) also aims to utilize this machine for advanced research and development tasks, anticipating a performance leap that is several times better than earlier systems.

    Vision for the Future

    Company executives view this project as a stepping stone toward a larger AI ecosystem. Jensen Huang, Nvidia’s chief executive, described AI as the driving force of “a new industrial revolution.” In contrast, Foxconn chair Young Liu emphasized the aim of linking public agencies with private enterprises through shared infrastructure. TSMC chair C.C. Wei sees the cluster as a way to accelerate semiconductor innovation.

    Foxconn plans to use the supercomputer for its internal purposes, such as enhancing smart-city services, refining driver-assistance features for its electric-vehicle platform, and improving manufacturing processes using digital-twin technology. Taiwan’s science minister, Wu Cheng-Wen, has expressed that the larger goal is to create “a smart AI island” that connects citizens, businesses, and the government through cutting-edge computing resources.

    Lastly, the supercomputer aligns with Nvidia’s broader strategy for expansion. Recently, the company announced the opening of a research and development center in Shanghai and indicated intentions to create a fleet of AI servers worth half a trillion dollars in the United States.

  • Nvidia’s Next Big AI Move: Cloud-Connected Humanoid Robots

    Nvidia’s Next Big AI Move: Cloud-Connected Humanoid Robots

    Key Takeaways

    1. Nvidia’s GPUs have driven significant growth in the AI industry, establishing a near-monopoly in AI hardware.
    2. CEO Jensen Huang envisions physical AI and robotics as the next industrial revolution, providing essential tools for robotics development.
    3. Nvidia launched Isaac, its first open-source humanoid robot operating system, allowing developers to enhance robots with trained models.
    4. The training process for robots involves creating videos from single images to teach new tasks using compressed action tokens.
    5. Nvidia supports developers with Universal Blackwell Systems powered by RTX PRO 6000 GPUs, facilitating powerful robot training capabilities.


    The AI industry has experienced huge growth over the past four years, largely due to Nvidia and its advanced GPUs that are specially made for AI tasks. Team Green started investing in AI back in the mid-2010s, but the real impact became noticeable only in recent years. With the hardware segment of AI now well established, and almost monopolized, Nvidia is exploring new opportunities that might increase its value even more. At this year’s Computex, Team Green hinted that one of these opportunities could be in physical AI and humanoid robotics.

    CEO’s Vision for the Future

    During his keynote earlier today, CEO Jensen Huang shared, “Physical AI and robotics will lead to the next industrial revolution. From AI brains for robots to simulated worlds for practice, or AI supercomputers for training foundational models, NVIDIA offers essential tools for each phase of the robotics development process.”

    Expanding Software Horizons

    With a focus on physical AI, Nvidia is also broadening its efforts in the software domain. The main element here is the company’s first open-source humanoid robot operating system named Isaac. Developers can build on this by adding trained models like the Gr00t N1.0, which was launched in March and has now been updated to version 1.5, introducing the Dreams component. Version 1.0, which featured the Mimic component, served as a foundation for training robot reasoning and actions. With the Dreams component, Nvidia is unveiling a framework that can produce large amounts of synthetic motion data (neural trajectories) that physical AI developers can use to teach robots various motor skills, including adapting to different environments.

    Training Process Explained

    The training method is similar to text-to-image and video models. Initially, the robots undergo post-training with Cosmos Predict world foundation models (WFMs). By using just one image as input, GR00T-Dreams can create videos of the robot doing new tasks in unfamiliar settings. The blueprint extracts action tokens (compressed data pieces easily handled by the robot’s neural network), which instruct the robot on how to carry out new actions.

    Nvidia’s Isaac GR00T is strongly connected to the Omniverse and Cosmos platforms to generate training data. The N1.5 update, along with models such as Isaac Sim 5.0 and Isaac Lab 2.2, will soon be downloadable from the Hugging Face repository. Innovative companies already utilizing the Isaac N models for humanoid robots include AeiRobot, Foxlink, Foxconn, Lightwheel, NEURA Robotics, Boston Dynamics, Agility Robotics, and XPENG Robotics.

    Developer Support and Hardware

    In addition, Nvidia is providing its Universal Blackwell Systems powered by RTX PRO 6000 GPUs, like the DGX Cloud infrastructure, enabling developers to access significant processing power for robot training with ease.

    Source:
    Link

  • Nvidia Launches RTX 5060 for $300 Amid Testing Driver Concerns

    Nvidia Launches RTX 5060 for $300 Amid Testing Driver Concerns

    Key Takeaways

    1. Nvidia is launching the RTX 5060 Blackwell GPU for entry-level desktops and laptops, featuring the same GB206 GPU as the RTX 5060 Ti but with fewer CUDA cores and 8 GB of GDDR7 VRAM.

    2. The desktop version has 3,840 CUDA cores and a boost clock of 2,497 MHz, priced starting at $299, while the laptop version has 3,328 CUDA cores and a maximum boost of 1,455 MHz, with configurations ranging from 45 W to 100 W TGP.

    3. The RTX 5060 can deliver over 100 fps at 1080p max settings, particularly with DLSS 4 and multi-frame generation in modern AAA games.

    4. Eight new laptops featuring the RTX 5060 Laptop GPU are introduced, starting at $1,099, including models from brands like Acer, Asus, and Alienware.

    5. Many media outlets couldn’t provide day 0 reviews of the RTX 5060 desktop GPU due to Nvidia’s decision to release press drivers simultaneously with the public, causing frustration among reviewers.


    At Computex 2025, Nvidia is set to launch the RTX 5060 Blackwell GPU aimed at entry-level desktops and laptops. Both the desktop and laptop versions of the RTX 5060 utilize the same GB206 GPU found in the RTX 5060 Ti, though they feature fewer CUDA cores and come equipped with 8 GB of 128-bit GDDR7 VRAM.

    Desktop Specifications

    The desktop version of the RTX 5060 boasts 3,840 CUDA cores, 120 texture mapping units (TMUs), 120 tensor cores, 48 raster operations pipelines (ROPs), and 30 ray tracing (RT) cores. It has a default boost clock of 2,497 MHz and a memory clock of 1,750 MHz (28 Gbps effective). Nvidia’s add-in board partners will sell the RTX 5060 starting at a price of $299.

    Laptop Features

    In contrast, the RTX 5060 Laptop GPU has even lower specifications to meet the needs of notebooks. This version comes with 3,328 CUDA cores, 104 TMUs and tensor cores, 32 ROPs, and 26 RT cores. OEMs can configure the GPU with a total graphics power (TGP) ranging from 45 W to 100 W, with maximum boosts reaching up to 1,455 MHz.

    Nvidia claims that the RTX 5060 can achieve over 100 fps at 1080p max settings, especially when using DLSS 4 and multi-frame generation in numerous modern AAA games.

    New Laptop Designs

    Moreover, Nvidia has introduced eight new laptops featuring the RTX 5060 Laptop GPU, which includes models like the Acer Predator Triton 14 AI, Asus TUF Gaming A14, Alienware 16X Aurora, Gigabyte A16 Pro, HP Omen Transcend 14, Lenovo Legion 9i, MSI VenturePro A14 AI+, and Razer Blade 14. These laptops support Blackwell Max-Q and will start at a price of $1,099.

    Nvidia also compared the RTX 5060 Laptop’s performance against the RTX 4060 Laptop and the RTX 3060 Laptop. The results, however, seem exaggerated since they used a 100 W TGP variant and enabled multi-frame generation, a feature not available in older GPUs.

    Review Access Issues

    Regrettably, many media outlets, including ours, couldn’t provide day 0 reviews of the RTX 5060 desktop GPU because Nvidia decided not to supply press drivers under embargo, which is standard procedure. This means that reviewers will receive RTX 5060 drivers at the same time as the public, which is later today.

    Day 0 reviews are crucial for helping potential buyers make informed choices, so many reviewers expressed their frustration regarding Nvidia’s decision to not allow embargoed reviews for the RTX 5060. During a recent pre-brief call at Computex 2025, Nvidia explained that it wanted to ensure the best experience for users on the availability date but did not elaborate further.

    Some early review sites like GameStar Tech and Ascii.jp managed to get access but only under Nvidia’s strict conditions. These conditions included comparisons of the RTX 5060 solely with the RTX 3060 and the RTX 2060 Super, and limiting game performance reviews to only select titles like Avowed, Doom: The Dark Ages, Marvel Rivals, Cyberpunk 2077, and Hogwarts Legacy, with features like DLSS or FSR frame generation enabled.


  • Hold Off on RTX 5060: Hardware Unboxed Critiques Nvidia’s Control

    Hold Off on RTX 5060: Hardware Unboxed Critiques Nvidia’s Control

    Key Takeaways

    1. Nvidia faced backlash for controlling the coverage and comparisons of the RTX 5060, allowing limited access to select media outlets.
    2. Previews were restricted to comparing the RTX 5060 only with the RTX 3060 and RTX 2060 Super, using specific games and settings to showcase its performance.
    3. The performance metrics presented may be misleading, as the RTX 5060’s impressive FPS figures were achieved under conditions that older models do not support.
    4. Hardware Unboxed warned gamers against rushing to purchase the RTX 5060, highlighting the lack of diverse and independent reviews due to Nvidia’s control.
    5. Gamers are advised to wait for more comprehensive reviews or consider alternative GPUs like the RX 9070 or older Nvidia models for better value.


    Nvidia’s efforts to manage the coverage of the RTX 5060 have led to intense backlash online. The company allowed certain media outlets to access the GPU ahead of its launch for performance previews, but with strict limitations. Responding to what it considers manipulation of the story, Hardware Unboxed has openly criticized Nvidia’s approach and is urging gamers to hold off on purchasing the RTX 5060 at its release.

    Controlled Comparisons

    For example, GameStar Tech, a German publication, noted in its preview that Nvidia required the RTX 5060 to be compared solely with the RTX 3060 and the RTX 2060 Super. Additionally, Nvidia instructed the site to limit tests to just five games.

    Nvidia even went so far as to mandate that the previews only feature 1080p resolution, showcasing performance metrics that included DLSS, Ray tracing, Ray Reconstruction, and Multi Frame Generation 4X (MFG). By insisting on the inclusion of MFG 4X, Nvidia allowed the RTX 5060 to achieve impressive triple-digit FPS, unlike the older RTX 3060 and RTX 2060 Super that do not support this technology.

    Misleading Performance Metrics

    For instance, in Game Rant’s testing, the RTX 5060 achieved approximately 223 FPS in Doom: The Dark Ages using the specified graphics settings. In contrast, the RTX 3060 and RTX 2060 Super could only reach around 63 and 52 average FPS, respectively, as per GameStar’s findings.

    The mandated settings clearly make the RTX 5060 appear to be a significant advancement in performance. Gamers who might not be very knowledgeable about technology could be tempted to buy the RTX 5060 based on these impressive figures.

    Warning Against Hasty Purchases

    To caution gamers against making a hasty decision, Hardware Unboxed produced a 32-minute video exposé that discusses Nvidia’s previous attempts to influence and manage its coverage. In this video, HU questions the reliability of the RTX 5060 previews due to their strict control by Nvidia.

    It appears that Nvidia only provided review units of the RTX 5060 to those publications that agreed to publish the previews. This tactic seems to reward compliant sites while sidelining those that did not comply. HU was among the latter, as they managed to obtain the RTX 5060 weeks early but were unable to release their reviews on launch day due to Nvidia not supplying them with press drivers.

    For the typical gamer looking to get the best GPU for their budget, the pre-launch previews of the RTX 5060 could be quite misleading. The absence of reviews from respected sources like HU and Gamers Nexus suggests that Nvidia may not want consumers to make informed choices based on a variety of reputable reviews.

    Consider Other Options

    Thus, if you’re a gamer on a tight budget searching for the best value GPU, it’s advisable to wait for a range of RTX 5060 reviews to be released. If you can’t hold off, a wise alternative might be to consider the RX 9070 or older Nvidia models like the RTX 4060 Ti, if they’re available.

    Source:
    Link


  • Nvidia’s Media Tactics: Fallout from Gamers Nexus Exposé

    Nvidia’s Media Tactics: Fallout from Gamers Nexus Exposé

    Key Takeaways

    1. Nvidia’s Pressure Tactics: Gamers Nexus (GN) alleges that Nvidia has pressured them to include specific performance metrics (MFG4X) in GPU reviews, even for unsupported models, claiming this is misleading and unethical.

    2. Control Over Media Access: Nvidia reportedly threatened to restrict access to key engineers if GN did not comply with their editorial demands, indicating a desire to control media coverage rather than simply influence it.

    3. Erosion of Trust: GN argues that Nvidia’s tactics have eroded trust in media coverage of their products, making it difficult for audiences to discern whether metrics were included voluntarily or under pressure.

    4. Manipulative Corporate Culture: The article highlights concerns over Nvidia’s corporate culture, where executives may shift blame and use access to engineers as leverage against media outlets, impacting editorial independence.

    5. Commitment to Integrity: Gamers Nexus emphasizes their commitment to maintaining editorial independence, even at the risk of losing future access to Nvidia products, urging other media outlets to resist similar pressures.


    As Nvidia CEO Jensen Huang shares news about artificial intelligence, Gamers Nexus has taken this opportunity to reveal what it calls Nvidia’s growing manipulative behavior towards the media. In an editorial characterized by Steve Burke’s distinctive tone, the piece highlights a troubling trend of editorial pressure, coercive tactics based on access, and retaliatory measures that the outlet claims exceed standard industry practices.

    Nvidia’s Demands

    Gamers Nexus alleges that for the past six months, Nvidia has pressured its team to include Multi-Frame Generation 4X (MFG4X) performance metrics in GPU reviews, even in cases where the graphics cards tested do not support this feature. GN refused, asserting that such inclusion would be misleading, unethical, and fundamentally dishonest.

    Threats and Access Control

    Instead of relenting, Nvidia reportedly intensified its approach by threatening to deny access to key internal engineers — notably thermal engineer Malcolm and latency expert Gamm0 — both of whom resonate well with GN’s audience. Even though these engineers have no direct link to MFG technology, their availability was allegedly made contingent on GN agreeing to Nvidia’s editorial stipulations.

    The main allegation from GN is that Nvidia is not just trying to influence coverage but is actively seeking to control it. This includes urging reviewers to depict the RTX 5070 as comparable to the significantly more powerful RTX 4090 by highlighting synthetic MFG-enhanced figures. Such comparisons risk misleading consumers by stripping away essential context.

    The Nature of Communication

    Burke argues that this behavior shifts Nvidia’s once-open communication with the media towards manipulation. While Nvidia has historically provided technical insights and briefings, GN contends that this access is now being leveraged to enforce compliance with corporate messaging. Once viewed as a hallmark of transparency, access to Nvidia’s engineers is increasingly seen as conditional and transactional.

    Gamers Nexus is not isolated in voicing concerns. Burke mentions that the outlet contacted other reviewers and media representatives globally, revealing a pattern of behind-the-scenes pressure to influence editorial content. This includes explicit or implied expectations regarding how products should be framed and which benchmarks should be highlighted. According to GN, these strategies indicate a broader company-wide approach that goes beyond regional public relations teams.

    Nvidia’s History of Controversy

    Nvidia has faced its share of media influence controversies. In 2020, it temporarily cut off Hardware Unboxed’s GPU sampling access due to the outlet’s unwillingness to emphasize ray tracing metrics. A similar situation reportedly surfaced again, with Nvidia pressuring GN to modify its editorial stance on MFG and DLSS coverage.

    Burke claims that Nvidia informed Gamers Nexus that integrating MFG into reviews was necessary to “secure budget” for interviews with engineers, even though GN incurs costs for traveling, filming, and editing those interviews without any financial compensation from Nvidia. This framing implies that Nvidia perceives all media interactions as fundamentally transactional, regardless of financial exchanges.

    The Ripple Effect on Media

    This framing also casts a shadow over other media organizations. Once Nvidia establishes precedent for linking access to editorial direction, every interview, performance graph, or editorial decision becomes questionable. GN argues that this not only undermines their credibility but also that of any reviewer discussing MFG or DLSS without full transparency regarding the nature of those discussions.

    Gamers Nexus emphasizes that their frustration is not aimed at Nvidia’s engineers. On the contrary, they regard professionals like Malcolm and Gamm0 as knowledgeable and trustworthy, conveying insights born from experience rather than marketing agendas. These individuals have cultivated genuine connections with GN and its audience. However, GN now believes that Nvidia is intentionally using that rapport to enforce compliance.

    Emotional Pressure Tactics

    Burke notes that Nvidia often references these engineers when discussing coverage requirements. This tactic seems designed to apply emotional pressure, knowing that GN values these relationships. GN describes this as a form of manipulation, suggesting that if restricting access to GPUs fails, perhaps restricting access to respected engineers will succeed.

    Despite producing over an hour of independent content on MFG and DLSS, including detailed analyses, GN states that Nvidia was still dissatisfied because the coverage was not integrated into review articles. The demand was explicit: include MFG4X in the charts or lose access.

    Erosion of Trust

    GN contends that this insistence has tainted any MFG or DLSS coverage presented by any outlet. Even if another reviewer includes those metrics out of genuine curiosity, audiences may question whether they did so voluntarily or under undue pressure. This erosion of trust harms both independent media and Nvidia’s own credibility.

    Gamers Nexus has a history of engaging with major companies. They have previously confronted Intel and AMD over similar matters, emphasizing that this isn’t personal. However, the outlet feels that Nvidia’s tactics have crossed a new line that now encompasses not just hardware access, but access to personnel and potential internal reprisal.

    Concerns Over Corporate Culture

    Burke cites prior reports from former Nvidia employees who described a corporate culture where executives frequently shift blame downwards. GN worries that by going public with these claims, they might inadvertently harm the engineers and PR contacts who have no influence over high-level policies. Nonetheless, the outlet believes that transparency is essential for progress.

    The article also recalls Nvidia’s contentious GeForce Partner Program (GPP) from 2018, which effectively required board partners to align their gaming brands exclusively with GeForce to retain Nvidia support. This program, widely criticized for being anti-competitive, was eventually withdrawn under public pressure and investigative scrutiny.

    A Call for Integrity

    In GN’s perspective, the current situation mirrors GPP in both approach and execution. The company is allegedly attempting to shape perception and limit coverage by using access and relationships as pressure points rather than relying on product quality and transparency. The message seems to be: adhere to the script, or risk exclusion.

    GN concludes with a firm statement: they will not yield. Even if it means sacrificing future review samples, interviews, or relationships, they are committed to maintaining editorial independence. They also urge other outlets to take a stand. “If you give Nvidia an inch, they’ll take a mile,” warns Burke.

    The piece finishes not by calling for a boycott, but by acknowledging that Nvidia produces remarkable products and its engineers offer valuable insights. However, both consumers and reviewers must be able to trust that coverage is accurate, independent, and not subject to corporate pressure.

    At the heart of this conflict lies more than just performance metrics or technologies; it concerns the role of the press in an industry where access has become a bargaining tool and transparency comes with strings attached.

    We will keep tracking this situation and provide updates as new information comes to light.

    Source:
    Link

  • Nvidia CEO Confirms Next China Chip Won’t Be Hopper

    Nvidia CEO Confirms Next China Chip Won’t Be Hopper

    Key Takeaways

    1. The Trump administration’s new chip export restrictions to China significantly affect manufacturers like Nvidia.
    2. Nvidia launched the less powerful H20 chip to comply with earlier regulations, but tighter restrictions have now been introduced.
    3. CEO Jensen Huang stated there won’t be a new version of the Hopper processor, and further modifications to it are not possible.
    4. Nvidia expects a $5.5 billion negative impact on quarterly results due to the ban on H20 chip exports, with China representing 13% of its sales.
    5. Concerns over the use of processors for artificial intelligence have led to increased regulations from the Trump administration.


    The recent restrictions imposed by the Trump administration on chip exports to China have had a significant impact on various manufacturers, including Nvidia. This company had previously launched the H20 chip aimed at this market in response to earlier regulations. The H20 chip, however, was less powerful compared to other models, allowing it to be exported to China. Nevertheless, tighter and more stringent regulations on sales to this country have now been introduced.

    Nvidia’s Forward Plans

    Jensen Huang, the CEO of Nvidia, shared insights during a livestream on Taiwan’s Formosa TV News, stating that the company is exploring its next steps. However, he made it clear that there will not be a new version of the Hopper processor. Huang also mentioned that further modifications to this chip are not feasible. Nvidia is currently evaluating how to proceed in this challenging market.

    Financial Implications

    Following the ban on H20 chip exports, Nvidia announced that this restriction would lead to a negative impact of $5.5 billion on its quarterly results. The Chinese market plays a crucial role for Nvidia, constituting 13% of its sales for the fiscal year ending January 2025.

    Rising Concerns

    The Trump administration has heightened regulations due to worries regarding the use of these processors in powering artificial intelligence (AI), particularly after the emergence of the Chinese language model DeepSeek.

    Source:
    Link

  • Nvidia Expands AI Chip Market with New Shanghai R&D Hub

    Nvidia Expands AI Chip Market with New Shanghai R&D Hub

    Key Takeaways

    1. Nvidia is negotiating for office space in Shanghai to establish a research-and-development center focused on local customer preferences.
    2. The new team in China will verify chip designs and improve existing products but will not design new graphics-processing units due to U.S. export laws.
    3. U.S. export regulations have significantly reduced Nvidia’s revenue share from China, prompting the development of lower-specification chips.
    4. Nvidia plans to launch a modified H20 accelerator in July to comply with stricter U.S. regulations, featuring reduced memory and performance.
    5. Nvidia aims to maintain its presence in China to tap into the growing AI-chip market, projected to reach $50 billion in three years, while staying compliant with export controls.


    Nvidia is currently in talks with officials in Shanghai to secure office space for a new research-and-development center. This facility aims to monitor local customer preferences and relay that information back to the company’s main office. During a visit in April, CEO Jensen Huang gained preliminary backing from the city’s mayor, who promised tax incentives and less bureaucratic hurdles for the initiative.

    Focus on Verification and Optimization

    The new team will not be involved in designing or altering graphics-processing units while in China. Instead, the engineers will focus on verifying chip designs, improving current products, and engaging in specific projects like research in autonomous driving, as per sources familiar with the initiative. Key intellectual property development will still take place outside of China to comply with U.S. export control laws.

    Impact of U.S. Export Licenses

    Since 2022, Washington has mandated export licenses for Nvidia’s most advanced AI processors. These regulations have reduced the revenue share from China to 13% in the last fiscal year, a drop from 26% prior to the enforcement of these rules. In response, Nvidia has begun to develop lower-specification versions of several chips, which has faced criticism from some U.S. officials.

    Upcoming Product Adjustments

    In its latest development, Nvidia has informed major Chinese cloud service providers that it will launch a modified H20 accelerator in July. This new version will have reduced memory capacity and performance to comply with the stricter U.S. regulations. Additionally, a Blackwell-based component that adheres to these limits is under development, but any shipments will need approval from Washington.

    Nvidia has a workforce of around 4,000 in China, with nearly half located in Shanghai. The new office space is expected to support existing employees and future recruits, enhancing local R&D efforts without relocating sensitive chip design tasks abroad. Huang has estimated that the AI-chip market in China could grow to about $50 billion within three years; he emphasized that maintaining a presence is crucial to prevent losing ground to local competitors like Huawei.

    Source:
    Link

  • Nvidia RTX 5090 Giveaway for Doom: The Dark Ages Launch Celebration

    Nvidia RTX 5090 Giveaway for Doom: The Dark Ages Launch Celebration

    Key Takeaways

    1. Nvidia is giving away a themed GeForce RTX 5090 GPU to celebrate the release of Doom: The Dark Ages.
    2. The giveaway is open worldwide, excluding certain areas in the US, Canada, and Australia, and participants must be adults and legal residents of eligible regions.
    3. To enter, leave a comment with #RTXON on social media before 5:00 PM PST on May 22nd, 2025.
    4. Winners will be announced on a separate page, but the exact announcement date is not specified.
    5. Doom: The Dark Ages offers a unique gaming experience with significant gameplay changes and is powered by the impressive idTech 8 engine.


    Now that Doom: The Dark Ages has been officially released for everyone, Nvidia is excited to mark this occasion by giving away a brand new GeForce RTX 5090. Since it’s an Nvidia-promoted title, the GPU features a unique Doom theme with custom artwork that will surely enhance your PC setup.

    Terms and Conditions

    The rules for the giveaway are quite simple. Although Nvidia’s wording suggests the offer is only for Australia, it’s actually open worldwide, except for certain areas in the US, Canada, and Australia. To join in, you need to be an adult and a legal resident of the regions mentioned on the giveaway page.

    How to Enter

    To enter, just leave a comment saying #RTXON on this X post (or on any social media platform you prefer, like Facebook, Instagram, TikTok, Threads, or Twitch) before 5:00 PM PST on May 22nd, 2025. Similar to past giveaways, the winners will be revealed on a different page. Nvidia hasn’t mentioned when they will announce the chosen gamer, so be prepared to wait a little while.

    Game Review

    In our review of Doom: The Dark Ages, we discovered that the game offers a thoroughly enjoyable experience that distinguishes itself from earlier versions. The combat feels completely different from Doom Eternal, with the shield playing a crucial role in the gameplay. Technically, its idTech 8 engine is impressive, and while the RTX 5090 will allow you to enjoy the best graphics, the game is still playable on the Steam Deck.

    Source:
    Link

  • Nvidia RTX 5050 Mobile GPU with 8GB VRAM Found in Lenovo Leak

    Nvidia RTX 5050 Mobile GPU with 8GB VRAM Found in Lenovo Leak

    Key Takeaways

    1. Nvidia’s RTX 5050 mobile GPU is expected to launch soon, with Lenovo providing specs and pricing for upcoming laptops.
    2. The official release of the RTX 5050 may coincide with the availability of the desktop RTX 5060, but Nvidia has not confirmed this yet.
    3. Pricing for the RTX 5050 laptops in the UK includes import taxes, with the LOQ model at £1,150 (~$1,430) and the Legion 5 at £1,350 (~$1,680).
    4. In the US, RTX 5050 laptops are expected to retail for around $1,100 or less, based on current RTX 5060 prices.
    5. The RTX 5050 will feature 8 GB of VRAM, an increase from the 6 GB found in RTX 4050 models, and is likely based on the GB207 model with 20 Streaming Multiprocessors.


    With the ongoing delays and issues around the availability of Nvidia’s RTX 5000 mobile series, the debut of the entry-level RTX 5050 dGPU has remained somewhat unclear since its unofficial announcement at CES 2025 earlier this year. Reports from Laptops Direct, as noted by Videocardz, hint that laptops featuring the RTX 5050 might be launching soon, as Lenovo has already shared specifications and pricing for the LOQ and Legion 5 models, which are expected to begin shipping in about three weeks.

    Potential Launch Timing

    Videocardz indicates that the official release of the RTX 5050 could happen around the same time the desktop RTX 5060 cards become available. However, Nvidia has yet to confirm the existence of this SKU. In the meantime, the UK retailer has taken down the two models from its search results.

    Pricing Details

    The listings from Laptops Direct show UK pricing that likely includes European import taxes, so these figures might not directly translate to the US market. The LOQ model, which also features an Intel i5-13450HX CPU, is priced at £1,150 (~$1,430), while the Legion 5 with an i7-13650HX can be pre-ordered for £1,350 (~$1,680). In the US, it’s expected that retailers will offer the RTX 5050 laptops for around $1,100 or less, especially since the RTX 5060 laptops currently retail for no more than $1,300.

    Specifications Overview

    According to the specifications shared by Lenovo, the RTX 5050 is confirmed to come with 8 GB of VRAM, which is an increase of 2 GB compared to the earlier RTX 4050 models. While no additional specifications have been confirmed at this time, Videocardz suggests that this SKU is likely based on the GB207 model, featuring 20 Streaming Multiprocessors, and it might be limited to GDDR6 128-bit memory.

    Source:
    Link


  • Nvidia Increases Prices by 5-15% Due to Export Restrictions

    Nvidia Increases Prices by 5-15% Due to Export Restrictions

    Key Takeaways

    1. Nvidia faced a $5.5 billion loss due to U.S. government restrictions on chip shipments to China.
    2. Rising costs from material, logistics, and tariffs have led Nvidia to increase prices for nearly all its products.
    3. The GeForce RTX 5090 price rose by over 10%, with other RTX 50-series cards increasing by 5-10%.
    4. AI silicon demand remains strong despite export restrictions, supporting Nvidia’s financial stability.
    5. Nvidia projects significant revenue growth, estimating around $43 billion for the upcoming quarter, a 65% increase year-on-year.


    Nvidia is changing its pricing approach due to a mix of outside challenges. As per supply-chain reports from Digitimes Taiwan, the firm faced a hefty loss of $5.5 billion after the U.S. government stopped shipments of its H20 chips to China. At the same time, the transfer of Blackwell production to TSMC’s facility in Arizona, along with increased costs for materials, logistics, and tariffs, has caused expenses to rise significantly.

    Price Increases for Products

    To safeguard its profit margins, Nvidia has raised the official prices “for nearly all of its products” and has permitted board partners to do the same. The channel price for the prominent GeForce RTX 5090 has surged from about NT$90,000 to NT$100,000 ($2,966 to $3,295)—an over 10 percent rise since its launch. Other cards in the RTX 50-series have also seen price hikes of 5-10 percent. Datacenter equipment hasn’t been immune either: the H200 and B200 modules now carry an increase of around 10-15 percent, a cost that server vendors are starting to transfer to their customers.

    Continued Demand for AI Silicon

    Even with the export restrictions and rising tariffs, the need for AI silicon from both U.S. and global cloud service providers remains robust. Therefore, supply-chain insiders anticipate that Nvidia’s financial results for the quarter ending in late May will align with their previous guidance and exhibit “excellent profit performance.” The company reported a revenue of $39.3 billion in its last quarter and has projected around $43 billion for the upcoming period, suggesting a year-on-year growth of approximately 65 percent.

    Source:
    Link