NVIDIA GDDR7-based memory would power the next-gen GPU lineup under the GeForce RTX 50 “Blackwell” branding for laptops. The note has been shared by CLEVO, The roadmap highlights the next-generation AMD / Intel / NVIDIA CPU and GPU, which were leaked as part of a ransomware attack on CLEVO.
More About NVIDIA GeForce RTX 50 “Blackwell”
This data leak suggests that there are at least 6 models from the GeForce RTX 50 “Blackwell” lineup based on GDDR7 memory technology. The GN22-X11 and GN22-X9 are the top-tier SKUs, aiming to combat certain high-end laptops. These chips boast 16 GB of GDDR7 memory running on a 256-bit bus interface and are built on the GN22 Board 1 PCB, which is a new package design not compatible with existing GPUs like the GN21 “Ada” GeForce RTX 40 series.
Another top-of-the-line version, the GN22-X7, is understood to be using the GB205 GPU and has 12 GB of GDDR7 memory at its disposal over a 192-bit bus interface. The lowest-end of these GPUs are apparently 8GB SKUs falling in the GN22-X2, GN22-X4, and GN22-X6 categories for entry-level laptops, with just the X2 having leaked thus far. It appears that these are not planned as replacements to the existing entry-level SKUs like RTX 4050 6 GB, RTX 3050 6 GB, and RTX 2050 4 GB (64-bit) which NVIDIA is supposed to carry over into 2025.
All three variants of the GeForce RTX 50 “Blackwell” Laptop GPUs will utilize the GN22 Board 2 package, another new design not compatible with older PCBs. These GPUs are anticipated to debut around CES 2025, powering the latest gaming platforms alongside next-generation CPUs like AMD’s Fire Range and Intel’s Arrow Lake. Further details on these GPUs are expected following the launch of the desktop family, rumored for late 2024 or early 2025.
Apple, the leading technology company, has hit another milestone, returning to $3 trillion in market capitalization. That number, which is a measure of the combined value of every single share still out there, highlights the lasting power and position Apple continues to have in the markets. But even as the glow of that accomplishment blew out, all eyes turned to chipmaker Nvidia, which also had some headlines of its own.
All About the Apple And Nvidia
Meanwhile, Nvidia for the first time evaluated over $3 trillion and closed trading as the second most valuable company in the world behind Apple. A feat second in the company market value only to Microsoft. Nvidia’s soaring stock value tripled in a year and demonstrates how much Giants need Nvidia’s AI server chips – and how much more of it they must have – and by what cost established it as Silicon Valley royalty, and essentially, the backbone of the digital revolution.
The first time in 2022 that Apple has not come in second, it was the first $3 trillion company in January 2022 since 2019. Meanwhile, and as long as you put the above swings into perspective, the fact that Apple’s market cap now serves as a daily reminder about the company’s standing on the global pyramid is slowly but surely taking on an aluminum-size life of its own.
Apple can then unveil its AI play at WWDC, and consequently, we should expect a range of new iOS 18 releases as well. Some popular rumors include hyperbolic tales of a more capable Siri featuring a large language model (LLM), wild AI additions, and more. Rumors have also recently indicated that Apple is in talks with OpenAI to bring ChatGPT to the iPhone ecosystem, before integrating it into the AI suite of etiquette, indicating they are chipping away at the user experience using cutting-edge AI.
Well, moving away from globalization, the TIGR Index offers a perspective on how Apple is doing in the marketplace, while Apple keeps moving along through the ever-changing tech world. While fluctuations in stock prices are inevitable, Apple’s enduring legacy and innovative prowess position it as a key player in shaping the future of technology and driving transformative change across various sectors.
At COMPUTEX 2024, MediaTek made a groundbreaking announcement that it will integrate NVIDIA TAO with its NeuroPilot SDK, part of the company’s edge inference silicon roadmap. This integration aims to create a seamless experience for developers as they bring advanced edge AI capabilities, including Generative AI, to a wide array of IoT applications powered by MediaTek’s cutting-edge silicon.
MediaTek Integrates NVIDIA TAO ToolKit with NeuroPilot SDK for Accelerated Development of Edge AI Applications in IoT
MediaTek, which powers over two billion connected devices annually, is committed to maximizing the performance of edge AI applications while ensuring efficiency. The company’s edge silicon portfolio integrates industry-leading multimedia and connectivity technologies, catering to premium, mid-range, and entry-level markets. This allows brands to deliver incredible AI experiences across various price points.
Enhanced Vision AI Capabilities with NVIDIA TAO Integration
The integration of MediaTek’s NeuroPilot SDK with NVIDIA TAO will push the boundaries of Vision AI on all MediaTek devices, both large and small. Developers will benefit from an easy-to-use interface and performance optimization features. NVIDIA TAO offers over 100 ready-to-use pretrained models, unlocking customized vision AI capabilities for a wide range of solutions and use cases. It automates the tuning of AI models, reducing development time and complexity, enabling quicker product launches even for developers without deep AI expertise.
Democratizing Access to AI
“Integrating NVIDIA TAO with MediaTek NeuroPilot will further expand our vision of democratizing access to AI, helping drive a new wave of AI-powered devices and experiences,” said CK Wang, General Manager of the IoT business unit at MediaTek. “With these expanded resources and MediaTek’s Genio product line, it’s easier than ever for developers to design innovative Edge AI products that stand out from the competition.”
Deepu Talla, Vice President of Robotics and Edge Computing at NVIDIA, added, “Generative AI is turbocharging computer vision with higher accuracy for AIoT and edge applications. Combining the power of NVIDIA TAO and MediaTek NeuroPilot SDK brings the latest and greatest vision AI models to billions of IoT devices.”
Empowering Developers and Accelerating Innovation
With MediaTek’s support for NVIDIA TAO, companies of all scales can harness the potential of edge AI in diverse IoT verticals, including smart retail, manufacturing, healthcare, transportation, and smart cities. This collaboration creates opportunities for developers to innovate and deploy AI-powered solutions more efficiently and effectively.
By integrating NVIDIA TAO with MediaTek’s NeuroPilot SDK, MediaTek is paving the way for the next generation of AI-powered IoT devices. This collaboration promises to enhance vision AI capabilities, simplify development processes, and democratize access to cutting-edge AI technology, empowering developers to create innovative solutions that transform industries.
NVIDIA(NASDAQ: NVDA) dominated COMPUTEX 2024 with a flurry of announcements showcasing the future of AI-powered PCs. From innovative in-game assistants to powerful content creation tools, NVIDIA’s latest advancements empower gamers, creators, and developers alike.
NVIDIA Unveils Groundbreaking AI Advancements at COMPUTEX 2024
Project G-Assist: Your AI-Powered Gaming Guru
Introducing Project G-Assist, an RTX-powered AI assistant that revolutionizes the way you interact with your games. No more scouring the web for answers! Project G-Assist provides real-time, context-aware help within your game. Ask questions about characters, lore, gameplay mechanics, or even system performance, and Project G-Assist delivers instant answers.
Digital Humans Take Center Stage with NVIDIA ACE
The future of human-computer interaction is here! NVIDIA is expanding its ACE digital human platform with new partners and NVIDIA NIMs. These microservices simplify deploying AI-powered digital humans for natural language interaction, speech recognition, and animation in games and applications.
Windows Copilot Gets RTX Boost
Microsoft and NVIDIA are collaborating to bring next-level AI capabilities to Windows applications. Developers can leverage Windows Copilot Runtime with RTX acceleration for on-device large language models (LLMs) and reasoning abilities. This opens doors for content creation automation, summarization, and more.
New RTX AI Laptops Deliver Power and Performance
Get ready for the latest wave of RTX AI laptops from ASUS and MSI. Featuring powerful GeForce RTX 4070 Laptop GPUs and efficient processors, these laptops boast Windows 11 AI PC capabilities and will receive a future upgrade to Copilot+ PC experiences.
DLSS 3.5 Takes Gaming to New Heights
Experience stunning visuals with the announcement of DLSS 3.5 with Ray Reconstruction. This cutting-edge technology is coming to Star Wars™ Outlaws (launching August 30th) and Marvel Rivals, delivering unparalleled performance and image quality.
Small Form Factor PCs Made Easier with SFF-Ready Initiative
Building a compact powerhouse just got simpler! NVIDIA’s SFF-Ready Enthusiast GeForce Cards and Compatible Cases initiative takes the guesswork out of choosing components for Small Form Factor (SFF) PCs.
RTX Remix Goes Open Source: Empowering Modders
RTX Remix is now even more accessible for modders. NVIDIA is open-sourcing the RTX Remix Toolkit, allowing for streamlined asset replacement, scene relighting, and expanded file format support. This empowers modders to breathe new life into classic games with ray tracing, DLSS, and more.
AI for Content Creators: Supercharge Your Workflow
Get ready for faster rendering and enhanced creativity! NVIDIA is integrating RTX AI acceleration into popular apps like ComfyUI, delivering significant performance improvements for content creators. Additionally, NVIDIA RTX Video – the AI-powered super-resolution tool – is now available as an SDK for developers and will soon be integrated into editing software like DaVinci Resolve and Filmora.
RTX AI Toolkit: Building Custom AI Models for PCs
Developers, rejoice! RTX AI Toolkit provides an end-to-end workflow for customizing, optimizing, and deploying AI models on RTX AI PCs. This empowers the creation of application-specific AI tools that unlock new possibilities.
New NVIDIA App Beta Update: Enhanced Features and User Experience
The essential NVIDIA app for gamers and creators gets even better! The June 4th update introduces 120 FPS AV1 video capture, one-click GPU performance tuning, and an improved overlay.
Free PC Game Pass: Experience Hundreds of Games
Enjoy a world of gaming possibilities with three free months of PC Game Pass for GeForce gamers, courtesy of a partnership between NVIDIA and Xbox.
G-SYNC Compatible Displays: A Smoother Gaming Experience
Immerse yourself in smooth, tear-free visuals with the expansion of the G-SYNC Compatible display lineup. Acer unveiled new Predator gaming monitors with G-SYNC Compatible features, while additional models from ASUS, Lenovo, LG, and Philips are on the horizon.
This marks a monumental shift in the world of AI and computing. NVIDIA’s groundbreaking advancements empower users to create, play, and work smarter than ever before.
NVIDIA GeForce RTX 5090 Graphics Card rumored to feature Modified 448-bit Bus Interface for up to 28 GB GDDR7 memory. This speculation comes from Chiphell Forum member, Panzerlied, who had once teased a 512-bit bus and new PCB to fit 16 GDDR7 memory modules.
The GB202 GPU allegedly driving the newer NVIDIA GeForce RTX 5090 will see its configuration reduced slightly, but it is still a step up from what’s in the GeForce RTX 4090 in terms of memory spec too, the draftsman has warned.
More About NVIDIA GeForce RTX 5090
The NVIDIA GeForce RTX 5090 graphics card will, in simple terms, feature a cut-down version of the 192 SM chip, and is known as GB202 GPU. The GPU will also come with a cut-back memory configuration of a 448-bit bus interface (or 14 modules) that will have a max of up to 28 GB of GDDR7 memory. The full version of the GB202 GPU has, in comparison, a 512-bit interface while the new PCB is able to support up to 16 GDDR7 memory modules for a total capacity of 32 GB.
NVIDIA could launch 28 Gbps GDDR7 memory modules right out of the gate, hitting up to 1568 GB/s in the 448-bit configuration: 56% more bandwidth than the RTX 4090’s 384-bit @ 21 Gbps.
Among other things, a 512-bit interface would provide up to 1792 GB/s to the GPU (a 15% gain), so perhaps the 448 was chosen for thermal, power, and cost reasons. By going with a 256-bit bus and 16GB of memory, possibility a future model, including the workstation graphics card will use the full 512-bit interface and 32GB of memory, possibly more cores. Whether retail versions of the RTX 5090 stick with this specifying is up in the air.
The potential GeForce RTX 5090 will be part of the next-generation RTX 50 “Blackwell” gaming lineup, which NVIDIA is currently planning to launch sometime in the third quarter of the year. Alongside the RTX 5090, the lineup is rumored to include the RTX 5080, which may be the first to hit retail shelves, followed by the RTX 5090 a few weeks later.
A new flagship GPU offering from the folks over at NVIDIA could enter the market as GeForce RTX 5090 with a huge monolithic GB202 ‘Blackwell’ die, reports Kopite7kimi. It will power the GeForce RTX 5090 graphics card and is expected to feature as many as 192 SMs (24,567 CUDA cores assuming it sticks with the existing design of using 128 per SM for its AD102 “Ada” chips). Recently unveiled details also confirmed the GPU will be monolithic.
More About the NVIDIA GeForce RTX 5090 GPU
NVIDIA has gone with a chiplet design for its HPC/AI chips like the B100 and B200, but so far the company does not appear to be willing-sized consumer GPUs leaving monolithically. The “Blackwell” GB202 GPU, on the other hand, is expected to be physically monolithic, possibly doubling up the number of SMs and cores in comparison with a cut-down GB203 die that fuels the GeForce RTX 5080. This difference is likely to be very important in the performance between RTX 5090 and 5080.
NVIDIA could use two GB203 die designs inside of one monolithic package, with better inter-die communication than traditional chiplet designs. The hardware solutions that NVIDIA offers, such as the NVLINK for dealing with communication bottlenecks are feasible, but they require capital, and costs might be reduced.
NVIDIA already connects two halves of a smaller die using an interconnect plus shared L2 cache approach in the GA100 and GH100 GPUs, which has scaled up well. The GB202 “Blackwell” GPU is also expected to utilize TSMC’s 4NP process node (5nm), offering a 30% increase in transistor density, which should enhance performance alongside architectural improvements.
There are also reports of a 512-bit interface for the RTX 5090 and a new cooling and PCB solution being developed. With rumors suggesting that AMD might withdraw from the ultra-high-end graphics segment with its RDNA 4 lineup, NVIDIA’s Blackwell GPUs, including the RTX 5090, could further solidify its dominance in the gaming market. The RTX 5090 is expected to launch shortly after the RTX 5080, which may be the first Blackwell gaming GPU available.
New rumors appeared around NVIDIA’s GeForce RTX 5090 “Blackwell” GPU including a brand new PCB design and a 512-bit GDDR7 memory interface (allegedly). As usual, Chiphell Forum member Panzerlied is to thank for bringing up the rumor and discussing it in relation to NVIDIA’s GeForce RTX 50 “Blackwell” GPUs. He has traditionally been on the money with feisty tech tidbits like this one. Panzerlied also previously leaked information on the Blackwell GPU SKUs featuring under the GB2** family and how they fare against the performance of RTX 4090.
More About NVIDIA GeForce RTX 5090
The rumors add that the NVIDIA GeForce RTX 5090 GPU will also feature a memory layout that is densely packed and differs from the rest of high-end modern GPUs with their 384-bit memory interfaces. Instead, the RTX 5090 is said to bring back that old 512-bit bash instead.
Multiple reliable leakers have so far confirmed that the NVIDIA GeForce RTX 5090 GPU will be using a 512-bit memory interface, but this is where we get our first whisper on its configuration. The GPU could have up to 16 DRAM modules, which are built on the GDDR7 standard and will run at an initial speed of 28 Gbps. which will consist of four memory modules located on the top, five on each side, and two more at the bottom, requiring a new PCB design.
NVIDIA’s response is working on a new PCB solution for the GeForce RTX 5090 Founders Edition card that could see up to three different PCB designs. It is not yet clear whether this is three individual PCBs in one card or even three distinct designs. The core and memory are to be cooled using a dual-fan system, while graphics card designs with double-sided memory modules have also been quoted in the related discussions. The basic configuration and initial expectations probably top out at 32 GB, with workstation variants potentially offering up to 64 GB.
That said, it’s not uncommon for NVIDIA to develop prototypes and custom PCBs during the early stages of experimentation on new GPUs. A good example is the prototype of an NVIDIA GeForce RTX 4090 GPU that boasts a unique PCB and cooling approach, but never went into production: Unfortunately, that also doesn’t help to prove whether the NVIDIA RTX 5090 GPU mentioned is a retail unit or actually another prototype.
NVIDIA has explored various PCB designs, including V-shaped solutions exclusive to Founders Edition variants. Additionally, the company has tested new cooling solutions for its GeForce RTX 50 “Blackwell” graphics cards. The potential adoption of these designs in final retail graphics cards remains uncertain. Currently, the NVIDIA GeForce RTX 5080 and RTX 5090 are expected to launch in Q4 2024.
Nvidia, the renowned chipmaker, made headlines in its fiscal first-quarter earnings report by announcing a 10-for-1 forward stock split. According to the company’s release, the shares will begin trading on a split-adjusted basis starting June 10, 2024.
NVIDIA Share Price: AI Leader’s Stock Soars After 600% Profit Surge and 10-for-1 Stock Split Announcement
Understanding the Stock Split
While a stock split doesn’t alter the financial fundamentals of a company, it does make each share more affordable. This often has a positive psychological impact on retail investors, making the stock appear more accessible. Nvidia’s statement highlighted that the split aims to make ownership “more accessible to employees and investors.”
Current and Post-Split Share Prices
Nvidia shares closed at $949.50 on Wednesday. With the 10-for-1 stock split, each share will be priced at approximately $94.95. Although an investor would need to purchase ten shares to maintain the same stake in the company, the lower price per share may attract a broader range of investors. Nvidia joins the ranks of other tech giants like Alphabet, Amazon, and Tesla, which all executed stock splits in 2022.
Market Reaction
Following the announcement, Nvidia shares surged nearly 4% in extended trading on Wednesday. The company also announced that each holder of Nvidia’s common stock will receive nine additional shares, distributed after market close on Friday, June 7. Trading on a split-adjusted basis will commence the following Monday.
A Historic Rally
Nvidia investors have enjoyed a remarkable rally over the past five years, with the stock price increasing 25-fold. Originally recognized as a leading manufacturer of advanced graphics processing units (GPUs) for video games, Nvidia has recently emerged as a key player in the artificial intelligence industry. Revenue in Nvidia’s fiscal first quarter skyrocketed by 262% compared to the same period last year, marking the third consecutive quarter with growth exceeding 200%.
Increased Dividend
In conjunction with the stock split, Nvidia announced an increase in its quarterly cash dividend from 4 cents per share to 10 cents on a pre-split basis. Post-split, this translates to a dividend of 1 cent per share.
Prepare for an unparalleled gaming experience as NVIDIA’s GeForce RTX technology takes center stage in the latest game releases. With day-one support for DLSS 3 and Reflex, Senua’s Saga: Hellblade II ensures gamers enjoy a definitive and immersive experience with reduced system latency.
DLSS 3 accelerating performance in Senua’s Saga: Hellblade II and Serum, XDefiant launching with latency reducing Reflex, and Star Citizen and Ships at Sea launch with DLSS Support
Senua’s Saga: Hellblade II
The highly anticipated release of Senua’s Saga: Hellblade II comes with DLSS 3 and Reflex support, providing GeForce RTX gamers with exceptional visuals and performance. Dive into this gripping narrative-driven adventure with the assurance of smooth gameplay and enhanced responsiveness.
Serum: Adrenaline-Fueled Adventure
Launching on May 23, Serum promises an exhilarating journey through a contaminated world filled with conspiracy and danger. GeForce RTX users can look forward to top-tier performance with DLSS 3, DLAA, and Reflex enhancements, allowing them to uncover the game’s secrets with stunning visual fidelity.
Star Citizen: Enhanced Super Resolution
The latest update for Star Citizen introduces DLSS 2 Super Resolution, significantly boosting frame rates in GPU-limited scenarios. This update enriches the gameplay experience, making space exploration more immersive than ever.
Ships At Sea: Next-Gen Maritime Simulation
Set to launch on May 23, Ships At Sea offers a next-gen ship simulation experience supported by DLSS 2 for optimized performance. Get ready to set sail and embark on thrilling maritime adventures with improved graphics and smoother gameplay.
EA SPORTS™ F1® 24: Racing Excellence
Racing enthusiasts can mark their calendars for May 28, when EA SPORTS™ F1® 24 hits the shelves. Equipped with DLSS 3, DLAA, Reflex, and Ray Tracing, this title promises an edge over the competition with extra frames and responsiveness. Experience the beauty of the race as you speed towards victory.
Wuthering Waves: Desolate World Exploration
Launching on May 22, Wuthering Waves invites players to explore the desolate world of Solaris-3. Featuring fast-paced combat and immersive gameplay, this highly anticipated title is further enhanced by DLSS 2 support for GeForce RTX users, ensuring a captivating experience.
XDefiant: Arena Shooter Action
XDefiant, Ubisoft’s long-awaited arena shooter, launches today with Reflex support, drastically reducing system latency by 58%. Dive into the action with improved responsiveness and precision.
Game Ready Driver and G-SYNC Compatible Displays
NVIDIA’s latest Game Ready Driver adds support for 11 new G-SYNC Compatible displays from partners including ASUS, Dell, Philips, IODATA, and LG. These monitors offer a superior Variable Refresh Rate experience, making gaming smoother and more enjoyable. For a complete list of all G-SYNC, G-SYNC ULTIMATE, and G-SYNC Compatible gaming monitors, visit the updated G-SYNC page.
Conclusion
GeForce RTX continues to push the boundaries of gaming technology, delivering unmatched performance and visual fidelity across a range of new titles. Explore the latest releases and experience the future of gaming with NVIDIA’s cutting-edge technology.
Get ready to experience the latest games at their absolute best. NVIDIA has released a brand new Game Ready Driver (version 555.85) optimized for a wave of exciting titles, including the highly anticipated F1 24 and Senua’s Saga: Hellblade II. This driver is a must-have for NVIDIA GeForce RTX gamers seeking to maximize performance and unlock stunning visuals.
F1 24 Roars to Life with Cutting-Edge Technologies
F1 fans, rejoice! The upcoming F1 24 is poised to deliver an unparalleled racing experience thanks to NVIDIA’s suite of cutting-edge technologies. The Game Ready Driver ensures seamless support for DLSS Super Resolution, Frame Generation, DLAA, Reflex, and the revolutionary new Ray Traced Dynamic Diffuse Global Illumination.
This translates to sharper visuals, smoother game play, and breathtaking lighting effects that realistically capture the atmosphere of high-octane racing.
Hellblade II: Dive Deeper with Unparalleled Performance
Senua’s Saga: Hellblade II is another major title benefiting from the new driver. This narrative-driven masterpiece boasts support for DLSS Super Resolution, Frame Generation, DLAA, and Reflex.
NVIDIA claims that players can experience a staggering threefold performance increase in Hellblade II by enabling DLSS Super Resolution (Performance mode) and Frame Generation. This means butter-smooth frame rates even at high resolutions, allowing you to fully immerse yourself in Senua’s harrowing journey.
Beyond F1 24 and Hellblade II: A Wave of DLSS-Powered Games
The excitement doesn’t stop there. The new Game Ready Driver brings NVIDIA’s performance-enhancing technologies to a plethora of other games. Here are just a few highlights:
Star Citizen: Alpha 3.23 update now features DLSS Super Resolution for a smoother space faring experience.
Wuthering Waves: This free-to-play open-world adventure joins the DLSS party starting tomorrow
Ships at Sea: Take to the high seas with smoother visuals thanks to DLSS Super Resolution support in this upcoming Early Access title.
Enhanced Reflex Support for Competitive Gamers
Tom Clancy’s XDefiant, the free-to-play competitive shooter, is another title receiving a boost with the new driver. The inclusion of NVIDIA Reflex promises a significant reduction in system latency, giving players a crucial edge in fast-paced online battles.
G-SYNC Compatibility Expands with New Displays
For those seeking the smoothest possible game play experience, the Game Ready Driver adds support for 11 new G-SYNC Compatible displays from leading manufacturers like ASUS, Dell, Philips, IODATA, and LG. This expanded range offers gamers a wider variety of options to find the perfect display for their setup.
Automatic Game Optimization and Bug Fixes
The new driver doesn’t just focus on the latest releases. It also includes automatic game optimization profiles for several popular titles through GeForce Experience or the new NVIDIA app. These profiles ensure optimal settings for Bellwright, Ghost of Tsushima Director’s Cut, Homeworld 3, and Infection Free Zone.
Finally, the driver addresses two reported bugs:
Random crashes in Tekken 8 on GeForce GTX 10-series graphics cards.
Reduced GPU utilization when Reflex is enabled (“On + Boost”) in Horizon Forbidden West Complete Edition.
Upgrade Your Experience Today: NVIDIA Game Ready Driver
The NVIDIA Game Ready Driver (version 555.85) is a comprehensive package for gamers seeking to unlock the full potential of their GeForce RTX graphics cards.
With support for the latest games, performance enhancements, and expanded G-SYNC compatibility, this driver is a must-have for anyone looking to elevate their gaming experience. Download the driver today from GeForce Experience, the new NVIDIA app, or the official NVIDIA website and prepare to be amazed!
