Tag: NVIDIA

  • XMG APEX 15 and APEX 17 gets new processors and GPU

    XMG APEX 15 and APEX 17 gets new processors and GPU

    XMG has provided a complete overhaul for its APEX series just before the COMPUTEX event. The XMG APEX 15 and APEX 17 not only have completely new chassis, but they are also XMG’s first gaming laptops that use AMD’s latest Ryzen 7 7735HS processor.

    Despite the APEX series’ outstanding connectivity, which includes two native monitor ports, XMG advertises it as the most attractive alternative in its own portfolio in terms of price performance. After the former entry-level XMG FOCUS series recently got much closer to the PRO series in terms of features and performance, XMG’s APEX series is now the entry point to its gaming laptop line.

    Both the XMG APEX 15 and the APEX 17 pair an NVIDIA GeForce RTX 4060 or an RTX 4050, each having a maximum GPU power of 140 watts, with AMD’s current Ryzen 7 7735HS, which has 8 cores and 16 threads.

    XMG APEX 15
    credit: wccftech

    A Full HD IPS monitor (1920 x 1080 pixels) with 144 Hz is ideal for these graphics cards in terms of resolution. The XMG APEX 15 and APEX 17 have two SO-DIMM sockets for up to 64 GB of DDR5-4800 RAM and two M.2 SSD ports with PCI Express 4.0 connections.

    As is customary with XMG devices, the devices can be customised with RAM and SSDs from the partner shop bestware.com. Even if the laptop is provided with only one SSD, the M.2 slots already have a heat conduction pad.

    Three USB-A (2x 3.2 Gen1, 1x 2.0) and one USB-C port (3.2 Gen2x1) are provided, as well as an SD card reader for full-size SD cards, Gigabit Ethernet, and separate audio ports for headphones (compatible with combo plug headsets) and microphone. Furthermore, all laptops provide two native graphics ports – HDMI 2.1 and Mini DisplayPort 1.4 – both of which are directly connected to the NVIDIA dGPU.

    XMG APEX 15
    credit: wccftech

    The M23 model generation is substantially lighter than its predecessor, measuring 360.2 x 243.5 x 26.7 mm and 2.2 kg (XMG APEX 15) and 392.9 x 260 x 24.9 mm and 2.47 kg (XMG APEX 17). The chassis, which has been completely redesigned, houses a 62 Wh battery and is made of finely sculpted plastic. The display lid of the 17.3-inch variant, which has a striking gaming appearance, is made of aluminium and adds stability.

    The XMG APEX 17 (M23), which is freely configurable on bestware.com, comes with a Full HD IPS display with 144 Hz, an AMD Ryzen 7 7735HS processor, an NVIDIA GeForce RTX 4050 graphics card, 16 (2×8) GB of DDR5-4800 RAM, a 500 GB Samsung 980 SSD, and other standard components. The beginning cost is € 1,299 with 19% VAT.

    For an extra cost, you can upgrade to the faster GeForce RTX 4060 (€ 158). The XMG APEX 15 (M23), which has the same hardware except from a GeForce RTX 4060, costs at € 1,399 on bestware.com.

    Also Read:

    Source

  • Nvidia becomes the first ever company to reach $1 trillion market cap

    Nvidia becomes the first ever company to reach $1 trillion market cap

    Nvidia became the first chipmaker in history to breach the $1 trillion market value barrier, reaching $1.01 trillion as its shares rose in response to the growing adoption of generative AI. At the time of writing, Nvidia shares were trading at $412.71, a 6% increase over the previous day and a stunning 188% increase for the year.

    As a result, Nvidia joins Apple, Microsoft, Saudi Arabia, Google, and Amazon as one of just six corporations with a market valuation of more than $1 trillion. Nvidia is also only the eighth firm in history to reach $1 trillion in revenue.

    Nvidia’s tremendous ascent over the course of the year coincided with the mainstreaming of generative AI and huge language models like ChatGPT, with blue-chip tech giants like Google, Meta, and Microsoft spending massively in AI infrastructure to beef up their AI capabilities.

    Nvidia has been instrumental in the development of the AI ecosystem of hardware and, perhaps more crucially, software that is powering the AI revolution.

    Nvidia
    credit: tomshardware

    The cheap maker appears to have no clear rival for its top position in AI hardware processing horsepower and the effectiveness of its CUDA software programming architecture.

    The company’s strengthened its position at Computex 2023 this week, unveiling a broad portfolio of new solutions to address a wide range of segments, ranging from massive supercomputers like its DGX GH200 to a new line of reference MGX system designs that will bring AI to mainstream data centres and enterprises.

    Nvidia is also utilising the rewards of its strategic acquisition of Mellanox to construct its own AI-centric networking infrastructure that connects its solutions, giving it yet another competitive advantage: the capacity to design tightly connected systems at data centre scale.

    Also Read:

    Source

  • Revitalizing Virtual Characters: NVIDIA ACE for Games Unleashes the Power of Generative AI

    Revitalizing Virtual Characters: NVIDIA ACE for Games Unleashes the Power of Generative AI

    NVIDIA has announced the introduction of NVIDIA Avatar Cloud Engine (ACE) for Games, a specialized AI model foundry service that revolutionizes gaming by incorporating intelligence into non-playable characters (NPCs) through AI-powered natural language interactions.

    NVIDIA ACE for Games Sparks Life Into Virtual Characters With Generative AI

    Game developers, middleware providers, and tool creators can leverage ACE for Games to develop and deploy customized speech, conversation, and animation AI models within their software and games.

    Revitalizing Virtual Characters: NVIDIA ACE for Games Unleashes the Power of Generative AI

    John Spitzer, NVIDIA’s vice president of developer and performance technology, expressed excitement about the potential of generative AI to transform player interactions with game characters and enhance immersion. NVIDIA’s expertise in AI and extensive collaboration with game developers position them at the forefront of utilizing generative AI in games.

    ACE for Games, built on the NVIDIA Omniverse platform, provides optimized AI foundation models for speech, conversation, and character animation. This includes NVIDIA NeMo™ for building and customizing language models, NVIDIA Riva for automatic speech recognition and text-to-speech capabilities, and NVIDIA Omniverse Audio2Face™ for generating expressive facial animation that matches speech tracks. Developers have the flexibility to integrate the entire ACE for Games solution or choose specific components as needed.

    Revitalizing Virtual Characters: NVIDIA ACE for Games Unleashes the Power of Generative AI

    NVIDIA collaborated with Convai, an NVIDIA Inception startup focused on conversational AI for virtual game worlds, to demonstrate the potential of ACE for Games. In the demo called “Kairos,” players interact with Jin, an NPC who responds realistically to natural language queries in line with the game’s narrative backstory, thanks to generative AI.

    The ACE for Games foundry service allows developers to fine-tune models for their games and deploy them via NVIDIA DGX Cloud, GeForce RTX PCs, or on-premises for real-time inferencing. The models are optimized for low latency to ensure immersive and responsive interactions in games.

    Leading game developers and startups are already leveraging NVIDIA‘s generative AI technologies. For example, GSC Game World is adopting Audio2Face in the highly anticipated game S.T.A.L.K.E.R. 2 Heart of Chernobyl, while Fallen Leaf is using Audio2Face for character facial animation in their sci-fi thriller Fort Solis. Charisma.ai utilizes Audio2Face to power animation in its conversation engine, enabling virtual characters through AI.

    To explore further advancements in NVIDIA Omniverse, ACE, and other technology developments, attendees can learn more at COMPUTEX.

  • NVIDIA becomes the only second company in history to increase its market cap by $200 billion

    NVIDIA becomes the only second company in history to increase its market cap by $200 billion

    NVIDIA appears to have fully adapted itself as the face of the ongoing AI revolution based on the company’s blazing-hot stock price growth. For those who may not be aware, NVIDIA recently released financial results for the first quarter of its fiscal year 2024. The company’s $7.19 billion in quarterly revenue topped average projections by 13%. Similar to this, the green team announced an optimistic adjusted EPS of $1.09, beating the $0.92 consensus forecast.

    The company’s wildly optimistic forecast that it will achieve an adjusted gross margin in the range of 69.5 percent and 70.5 percent in its fiscal Q2 2024 was the real stunner for investors. For reference, analysts predicted that this statistic would be close to 66.9 percent. Additionally, the company anticipates revenue of $11 billion (plus or minus 2%) for the second quarter of 2024. This outlandish forecast was a staggering 55 percent more than the current average projection of $7.1 billion.

    NVIDIA
    credit: businesstoday

    NVIDIA shares finished at $305.38, giving the company a market value of $755.24 billion.

    This results in a $979.43 billion market cap. This implies that NVIDIA has raised its market capitalization by a staggering $224.19 billion in the course of just a few short hours. Only the $597 billion increase in market value that PetroChina had registered on the 5th of November 2007 is bigger than this record in the history of the market.

    According to Nomura, NVIDIA’s prognosis for the July quarter indicates that the recovery in chip demand from US cloud server providers started sooner than anticipated. The brokerage believes NVIDIA extremely high overall sales estimate implies ongoing quarter-over-quarter growth in the July quarter. NVIDIA’s PC-related (gaming) sales increased quarter over quarter in the April quarter.

    Also Read:

    Source

  • NVIDIA shares rise in value after an amazing Q1 2023

    NVIDIA shares rise in value after an amazing Q1 2023

    NVIDIA Corporation announced earnings for the first quarter of fiscal year 2023 recently. In comparison to NVIDIA’s Q4 2022, the previous quarter was relatively stable, and significantly, it demonstrated sequential growth for all of the company’s key business areas.

    The quarter was a huge success for NVIDIA, whose revenue was $7.1 billion for the three-month period, and earnings per share came out at $1.08 – a tremendous beat above the consensus estimates of $0.92.

    The show’s star, as usual, was NVIDIA’s data centre portion. Both NVIDIA and AMD have seen their income statements shift away from gaming and consumer items and towards data centres, particularly in the last year when rising inflation had a greater impact on consumer purchasing. This has resulted in their data centre divisions becoming the largest business sector, and NVIDIA was no exception in Q1 2024.

    NVIDIA
    credit: nvidia

    The data centre sector was NVIDIA’s sole business division that experienced strong annual growth and the only one with more than a billion dollars in revenue.

    It generated $4.2 billion in revenue for the chip maker in the quarter, representing a 14% annual increase over the $3.7 billion in sales generated in the previous quarter. The yearly sales rise for data centres also outperformed NVIDIA’s annual revenue reduction of 13%. This expansion was fueled by several large business wins, including those from Google and Dell.

    However, in terms of consecutive growth, gaming takes the top spot. During the quarter, the company’s long-standing bread and butter business division generated $2.2 billion in revenue. While this was still down 38% year on year, revenue increased by 22% during the quarter, indicating that a potential recovery is just around the corner.

    NVIDIA

    Ms. Collette Kress, NVIDIA’s chief financial officer, credits a substantial amount of the increase in data centre income to the ramp-up in demand for generative artificial intelligence technologies.

    The CEO noted in her statement that the green teams Hopper and Ampere solutions are leading the way in this rapidly growing industry category. Ms. Kress said that while macroeconomic conditions caused an annual slowdown, the RTX 40 graphics processing units contributed to revenue growth in the first quarter.

    During the quarter, the company’s inventory decreased while its accounts receivable increased, indicating that the company is easing through some of the channel difficulties that have afflicted it in recent quarters. Higher sales also resulted in increased cash flow, although Ms. Kress refused to provide any specifics concerning changes in inventory or receivables.

    Also Read:

    Source

  • Nvidia orders has raised TSMC’s 5nm production capacity to its absolute max

    Nvidia orders has raised TSMC’s 5nm production capacity to its absolute max

    According to reports, Nvidia is making rush orders with TSMC for its H100 and A100, as well as the H800 and A800 tailored exclusively for China. TSMC’s SHR for Nvidia’s AI GPUs will be available throughout the year.

    The orders have already increased TSMC’s 7/6nm process family utilisation rates, which had previously been falling fast due to diminishing demand, while the foundry house’s utilisation rates for the 5/4 processes are also climbing quickly, reaching nearly full capacity.

    Nvidia CEO Jensen Huang remarked, while announcing the company’s results for the first quarter of fiscal 2024 (ending April 30, 2023), that the computer industry is undergoing two transformations at the same time, namely accelerated computing and generative AI.

    Nvidia
    credit: digitimes

    Nvidia is riding the AI wave that was sparked by the introduction of ChatGPT. The company commands a commanding position in the AI GPU industry. Following the US chip restrictions against China, Nvidia rapidly created scaled-down versions of its H100 and A100 processors for the country, the H800 and A800, to avoid the ban.

    According to reports, demand for the reduced versions has been significant from large Chinese corporations such as Baidu.

    Nvidia posted strong financial results for the first quarter of fiscal 2024, with revenue increasing 19% sequentially to $7.19 billion. It has also issued a bullish forecast for the second quarter.

    Even Intel CEO Pat Gelsinger admits that Nvidia is well-positioned to capitalise on AI potential.

    Nvidia’s good performance benefits TSMC, its chipmaking partner. TSMC had expected big orders from Apple to support the vendor’s introduction of new products, but not the surge of the companys urgent orders.

    TSMC stated during its April 2023 investor conference that its first-quarter 2023 business was hit by weaker macroeconomic conditions and declining end market demand, which caused customers to alter their demand accordingly. It stated that the company expects clients’ inventory adjustments to continue to have an impact on business in the second quarter.

    Also Read:

    Source

  • NVIDIA Announces The Lord of the Rings: Gollum with DLSS 3 and More

    NVIDIA Announces The Lord of the Rings: Gollum with DLSS 3 and More

    Today, NVIDIA announced that The Lord of the Rings: Gollum™ will launch on May 25th with support for DLSS 3 and many other NVIDIA technologies. The game will also be available with DLSS 2 support on Firmament and Togges.

    The Definitive The Lord of the Rings: Gollum Experience

    The Lord of the Rings: Gollum™ is a story-driven stealth adventure game based on the books by J. R. R. Tolkien. Players take on the role of Gollum, a twisted and deformed creature who is on a perilous journey to reclaim the One Ring. The game will feature a unique narrative that explores Gollum’s thoughts and motivations, as well as a variety of gameplay mechanics that allow players to sneak, climb, and solve puzzles.

    NVIDIA Announces The Lord of the Rings: Gollum with DLSS 3 and More

    Multiply Performance with DLSS 3

    DLSS 3 is a new version of NVIDIA’s AI-powered upscaling technology that delivers stunning visuals and improved performance. In The Lord of the Rings: Gollum™, DLSS 3 can deliver up to 3.5x the performance at 4K resolution with all settings maxed out and ray tracing enabled.

    Enhance Image Quality with DLAA

    DLAA is an AI-based anti-aliasing mode that uses the same technology as DLSS to deliver smoother, more detailed images. In The Lord of the Rings: Gollum™, DLAA can be combined with DLSS Frame Generation for the best possible image quality and performance.

    Responsive Gameplay with NVIDIA Reflex

    NVIDIA Reflex is a technology that reduces system latency, making games feel more responsive and competitive. In The Lord of the Rings: Gollum™, NVIDIA Reflex can reduce system latency by up to 56%.

    More DLSS Games

    DLSS is quickly becoming the standard for high-performance gaming. In addition to The Lord of the Rings: Gollum™, DLSS 2 is also available on Firmament and Togges. More DLSS games are expected to be released in the coming months.

    For more information, visit the NVIDIA website.

  • Microsoft and NVIDIA sign a 10-year GeForce NOW partnership

    Microsoft and NVIDIA sign a 10-year GeForce NOW partnership

    Xbox Corporate Vice President Sarah Bond announced a 10-year agreement with NIVIDIA, a computer parts maker recognised for their Graphics Processing Units. It grants Xbox users and other Microsoft devices access to GeForce NOW.

    Microsoft and NIVIDIA have collaborated to provide cloud gaming to Xbox devices, as well as iOS, Android, MAC, and SHIELD TV compatibility. The collaboration allows Xbox players to broadcast games using the native GeForce software. Xbox announced ambitions to introduce Cloud Gaming to linked devices last year.

    This statement follows Amazon’s announcement of an open world MMO set on the Lord of the Rings setting. The years 2023 and 2024 appear to be interesting for gaming.

    Should Microsoft’s plan to acquire the World of Warcraft publisher be allowed by regulators, Activision Blizzard games such as Call of Duty will be released on the cloud gaming service.

    Microsoft
    credit: tomshardware

    Streaming qualifying Xbox and Bethesda PC titles purchased through Steam or Epic Games Store from GeForce Now to PCs, macOS, Chromebooks, cellphones, and other devices is now possible.

    In the future months, support for games purchased from the Microsoft Store is expected. Microsoft recently secured a slew of game distribution deals in an effort to ease regulatory worries over its proposed acquisition of Activision Blizzard.

    Nvidia claims that the GeForce Now library contains over 1,600 games, with new titles added every Thursday. Tin Hearts and The Outlast Trials, in addition to Gears 5, were released today.

    Also Read:

    Source

  • AMD leads in German GPU sales over rival NVIDIA

    AMD leads in German GPU sales over rival NVIDIA

    The latest sales data from German retailer Mindfactory reveals an interesting trend among gamers in Germany – they are increasingly favoring AMD Radeon graphics cards over NVIDIA GeForce RTX GPUs.

    This pattern has persisted for ten consecutive weeks, indicating a steady rise in AMD’s market share. TechEpiphany conducted the analysis using GPU sales statistics from Mindfactory, while 3DCenter compiled the information into a convenient table.

    During the most recent sales period, there was a noticeable decline in GPU sales at the German retailer, with a significant 38% drop compared to sales figures from four weeks earlier. The total number of graphics cards sold stood at 4,510, which is a sharp decrease compared to the 2022 average of 22%.

    The comprehensive table prepared by 3DCenter.org provides insights into the sales figures of the three leading graphics card manufacturers: AMD, Intel, and NVIDIA. Although NVIDIA GeForce had a strong start to the year, Radeon began edging out NVIDIA in sales approximately ten months ago.

    While initially a slight lead, AMD’s advantage has steadily grown. Meanwhile, Intel continues to face challenges in keeping up with its rivals, capturing only 1% of the overall GPU sales at Mindfactory. However, there may be potential for change as Intel introduces new deals, promotions, and a fresh program that could alter the dynamics in their favor.

    AMD
    credit: wccftech

    The recent sales success of AMD on Mindfactory can be attributed to significant discounts and price drops on the previous-generation Radeon 6000 series graphics cards.

    These more affordable prices have made AMD GPUs more enticing to consumers compared to the higher-priced GeForce RTX 30 and RTX 40 offerings from NVIDIA. Considering the cautious approach of consumers due to the current economic situation and their concerns about hardware costs, it is understandable how AMD, with its competitive pricing strategy, has managed to maintain its lead.

    https://twitter.com/TechEpiphany/status/1657362688148643842?ref_src=twsrc%5Etfw%7Ctwcamp%5Etweetembed%7Ctwterm%5E1657362688148643842%7Ctwgr%5Ecfb95eabf7a33d32a2c0cca8a986be3438413faf%7Ctwcon%5Es1_&ref_url=https%3A%2F%2Fwccftech.com%2Famd-radeon-maintains-ten-week-long-gpu-sales-lead-over-nvidia-geforce-in-germany%2F

    TechEpiphany highlights the XFX Radeon RX 6950 XT Merc Black graphics card as the top-selling GPU, with custom NVIDIA graphics cards from MSI (including the RTX 4070 Ti variant and RTX 4070 Gaming X Trio) following closely in overall sales on Mindfactory.

    Looking ahead, both NVIDIA and AMD are expected to release new graphics cards soon. AMD is anticipated to introduce the Radeon RX 7600 GPU, while NVIDIA is preparing to launch the GeForce RTX 4060 series graphics cards.

    Although Intel’s GPU series offers comparable performance to previous-generation mainstream NVIDIA GPUs and boasts “day one” support for new game releases, the company currently has a relatively smaller presence in the market.

    Also Read:

    Source

  • AMD RX 7900 XTX manages to match NVIDIA’s 4090 GPU in performance only with a massive 700W power supply

    AMD RX 7900 XTX manages to match NVIDIA’s 4090 GPU in performance only with a massive 700W power supply

    A Reddit user known as Jedi95 recently performed a modification on the AMD RX 7900 XTX, one of the top-performing GPUs, in order to eliminate the power limit and test its maximum potential. By utilizing a high-end custom water-cooling setup with extremely low coolant temperatures of around 10 degrees Celsius, Jedi95 achieved impressive results.

    The power-modded AMD RX 7900 XTX reached an astonishing clock speed of 3.33GHz and demonstrated performance comparable to that of the RTX 4090 in 3DMark benchmarks. However, achieving such performance required the card to consume nearly 700W of power.

    The performance achieved through the modified AMD RX 7900 XTX exceeded expectations and even surpassed some RTX 4090 cards in the 3DMark Time Spy Extreme test, scoring 18,335 points compared to the RTX 4090’s GPU index score of 16,654 points, as reported in PCGamer’s review.

    This suggests that AMD RX 7900 XTX has the potential to compete with Nvidia’s flagship RTX 4090, surpassing the performance of the RTX 4080, albeit at the cost of significantly higher power consumption. It’s important to note that 3DMark benchmarks may not necessarily reflect real-world gaming performance.

    Power throttling is a common occurrence with AMD’s RX 7900-series GPUs, as most games can push these GPUs to their maximum power rating, resulting in throttling or reduced clock speeds. This differs from Nvidia’s RTX 40-series GPUs, which often operate below their maximum power consumption during gaming.

    Jedi95’s modified AMD RX 7900 XTX also matched the best benchmark results of the GPU in the 3DMark browser, even when compared to GPUs likely cooled with liquid nitrogen.

    AMD RX 7900 XTX

    The modification performed by Jedi95 involved a “shunt-like” mod that manipulated the VRM controller to report false power consumption figures to the GPU, effectively bypassing AMD’s power limitations.

    This allowed the AMD RX 7900 XTX to operate without any power restrictions, as the GPU remained unaware of its actual power consumption. Jedi95 stated that this mod was superior to a traditional shunt mod, as power reporting could be controlled using software.

    It’s not surprising that pushing voltage and frequency beyond the chip’s designed limits significantly increases power consumption. Most chips, including CPUs, GPUs, and other system-on-chips (SOCs), are optimized for specific voltage and frequency ranges.

    These findings offer insights into the potential performance capabilities of AMD’s RDNA 3 GPUs when power limits are removed. The Navi 31 chip used in the AMD RX 7900 XTX can achieve incredibly high clock speeds well above 3.3GHz, resulting in significantly improved performance.

    Based on Jedi95’s results, it appears that AMD has significant headroom to create a theoretical RX 7950 XTX that could rival the RTX 4090. However, achieving such performance would require AMD to significantly increase the chip’s power consumption.

    Also Read:

    Source