Latest News - Page 568

AMD formally announced its next-generation Zen 5-based Ryzen 9000 series CPUs at Computex 2024 earlier this month, and the company is already optimizing and tweaking its 3D V-Cache driver.

This is getting us all excited about Ryzen 9000X3D processors, which will fight Intel directly with its upcoming Arrow Lake-S CPUs as the best gaming CPU on the market. In a post on X, leaker "harukaze5719" discovered AMD's new 3D V-Cache Performance Optimizer driver.

AMD last updated the 3D V-Cache driver when it launched the Ryzen 7000X3D processors, and there's not much information on what the 3D V-Cache driver actually does. It was first released as a Windows 10 exclusive driver for the Ryzen 7 5800X3D processor, but it's also available with AMD's bigger Ryzen 7000X3D family of processors and works in Windows 11 with the 7000X3D series chips.

Project G-Assist was a surprise announcement from NVIDIA at Computex 2024. It is a new AI-powered digital assistant designed for PC gaming with a GeForce RTX graphics card - or what the company now calls an RTX AI PC. Like DLSS, it's a general LLM or AI model trained on a broad range of titles, and Project G-Assist is ready to be plugged into any existing or upcoming release.

Project G-Assist's capabilities and features blew us away when we got to participate in a live demo of the technology integrated into the popular survival game Ark: Survival Ascended with help from its developer, Studio Wildcard. One of Project G-Assist's key features is that it can analyze and see what's happening on the screen. Plus, it has real-time access to game-specific data.

Ark is a game full of dinosaurs, so a few of the first questions we asked were simple things like, "What's that dinosaur on the left there?" "What's the best weapon to take out that enemy, and where do I find it?" As a general model, Project G-Assist, as seen in this example, has access to up-to-date lore and various Ark knowledgebases and Wikis.

Game tech demos are usually created to showcase new rendering techniques, advanced animation, or stuff like real-time ray tracing. Even though NVIDIA and Inworld AI's Covert Protocol tech demo features an impressive-looking and lavish hotel lobby with detailed environments, ray tracing, and highly detailed character models, the visual fidelity is simply here to showcase a groundbreaking new bit of tech - digital humans that you can interact with.

NVIDIA first showcased Covert Protocol at GTX a few months ago. At Computex 2024 in Taipei, the company presented the same tech demo with AI digital humans powered by NVIDIA Avatar Cloud Engine (ACE) technologies running locally on a high-end GeForce RTX 4090 powered rig.

At the show, I had the chance to go hands-on with NVIDIA's impressive RTX AI technologies, and when it came to Covert Protocol, I was quietly blown away by the depth and breadth of its digital humans.

Samsung Electronics has unveiled its latest foundry innovations and outlined its vision for the "AI era" during its Samsung Foundry Forum (SFF) event, which it holds annually at its Device Solutions America HQ in San Jose, California, USA.

The theme was "Empowering the AI Revolution" to which Samsung reinforced its process technology roadmap, including two new cutting-edge nodes -- SF2Z and SF4U -- as well as its integrated Samsung AI Solutions platform that holds the "unique strengths" of its Foundry, Memory, and Advanced Package (AVP) businesses.

Dr. Siyoung Choi, President and Head of Foundry Business at Samsung Electronics, said: "At a time when numerous technologies are evolving around AI, the key to its implementation lies in high-performance, low-power semiconductors. Alongside our proven GAA process optimized for AI chips, we plan to introduce integrated, co-packaged optics (CPO) technology for high-speed, low-power data processing, providing our customers with the one-stop AI solutions they need to thrive in this transformative era".

TSMC is absolutely dominating, and we all know it. The big AI chip players are eating up its 3nm production into 2026, and now upstream IC design companies are reporting price increases.

NVIDIA, AMD, Intel, Qualcomm, MediaTek, Apple, and Google are all working on getting chips made on TSMC's new 3nm process. Ctee reports that Qualcomm's new Snapdragon 8 Gen4 processor is made on TSMC's new N3E process, and its quotation has increased by 25% compared to the previous generation, which Ctee reports "does not rule out subsequent price increases."

Another issue is that semiconductor competitor Samsung led the mass production of 3nm chips using the GAA (Gate All Around) process on June 22, but the first-gen N3 node "SF3E" was not successful at all. The yield rates of Samsung's in-house Exynos 2500 chips dropped, and Google Tensor processors are all made by Samsung. The current fourth-generation Tensor SoC uses Samsung's new 4nm process, but the rumored fifth-generation Tensor chip will switch to TSMC's new 3nm process.

The world of AI chips hasn't stopped, and won't stop, with advanced packaging production capacity in short supply.

TSMC's new CoWoS plant in Nanke Chiayi Park is entering the "environmental review stage," reports UDN, with TSMC now purchasing equipment to help speed up the construction of the new CoWoS advanced packaging facility. At the same time, the two new CoWoS advanced packaging factories were planned to be built in Nanke Chiayi Park, but they're "not enough," reports UDN.

TSMC reportedly sent personnel down to southern Taiwan to scout for land for a third CoWoS advanced packaging facility. UDN reports that the Chiayi County Government previously announced that TSMC's new advanced packaging plant will be located in Nanke Chiayi Park, covering an area of around 20 hectares. The first CoWoS advanced packaging plant was planned to take up around 12 hectares of space, will be completed by 2026, and will create 3000 jobs.

Reuters reports that Taiwanese chip design giant MediaTek is developing an Arm-based PC processor that will run Windows. " Three people familiar with the matter" say this.

We have been hearing whispers of MediaTek teaming with NVIDIA for a new Arm-based AI PC processor and comments from Dell CEO Michael Dell that we will see an NVIDIA-based processor released for a next-gen AI PC in 2025.

Microsoft also revealed its next-generation laptops with chips designed by Arm last month, its huge new Copilot+ PC push, and the huge relationship with Qualcomm and its Arm-based Snapdragon X series of processors that are inside Copilot+ PC systems running Windows. It seems Qualcomm has the exclusivity here, but that will run out in 2025 and we're going to see an absolute flood of Arm-based AI PC processors hitting the market.

ASRock has launched its new DeskMeet X600 Mini-PC that has a very cool trick up its sleeve: it features a PCIe extension cable that lets you use an external graphics card... which is something we don't see in a world dominated by Thunderbolt 4 and OCulink connectivity.

It's impossible to cram in a monster GeForce RTX 4090 graphics card inside a Mini-PC system, but with a full PCIe interface through an external PCIe extension cable, it just makes sense. ASRock has placed the PCIe connector hidden under a cover (because some users won't use an external discrete graphics card) that can be taken off and used at any time.

The company does miss a huge opportunity, though: ASRock doesn't include a high-quality PCIe extension cable with its new DeskMeet X600 Mini-PC. This means you'll need to find one yourself, and if you have problems, it's not ASRock's issue. ASRock, you should've included the PCIe extension cable with the system.

AMD has one of the best gaming processors on the planet with its Ryzen 7 7800X3D and its second-gen 3D V-Cache technology. However, it looks like the 7800X3D will be a faster gaming CPU than the next-gen Zen 5-based Ryzen 9 9700X, and that's coming directly from AMD.

In an interview with Tom's Hardware, AMD's Senior Technical Marketing Manager, Don Woligroski, said: "Is it the fastest in gaming? It's faster than the competition in our tests. X3D is still the king of the hill, but by a much smaller margin than typically between X3D and non-X3D. So a 7800X3D would, yes, be faster than 9700X, but maybe not by as much as you would expect".

In an interview with PCGamer earlier this week, Donny said: "The X3D stuff, we have a lot to say about it. The best part about it is we're not just resting on laurels. We're improving what we can do with X3D, it's really exciting and I'm super looking forward to talking to people about that. It's not like, hey, we've also added X3D to a chip. We are working actively on really cool differentiators to make it even better. We're working on X3D, we're improving it".

NVIDIA's next-generation flagship GeForce RTX 5090 has some new leaked specs, with the GB202 GPU configuration being leaked onto X by kopite7kimi:

GB202 is going to be an absolute freaking beast, offering 33% more CUDA cores than the AD102 GPU inside of the GeForce RTX 4090, with GB202 rumored to pack in an incredible 24,576 CUDA cores. AD102 has just 18,423 CUDA cores in comparison. GB202 will power the new ultra-enthusiast flagship GeForce RTX 5090 graphics card, which is rumored to have a monster 512-bit memory bus with ultra-fast, next-generation GDDR7 memory.

RTX 5090 performance expectations: I'm expecting to see at least a 50% improvement in 4K gaming performance, with that coming from a mix of Blackwell GPU architecture improvements over Ada Lovelace, and the wicked new, ultra-fast GDDR7 memory on that deliciously huge 512-bit memory bus. We should expect at least 50-70% more 4K gaming performance with Blackwell + GDDR7 + 512-bit memory bus, and that's without DLSS 4 in the mix.

SK hynix is far ahead of its HBM competitors, but they're simultaneously behind their GDDR7 memory competitors, where the South Korean memory giant won't have GDDR7 memory ready until Q1 2024.

The last time we heard about SK hynix's GDDR7 memory was back at NVIDIA's GTC 2024 event (GPU Technology Conference) where we saw GDDR7 modules at 16-24Gb, up to 40Gbps speeds, and up to 160GB/sec memory bandwidth per module. The baseline of GDDR7 memory is 32Gbps and 128GB/sec bandwidth per module, so SK hynix has a rather nice 25% bandwidth improvement for future variations of GDDR7 memory.

Last week at Computex 2024, SK hynix said that its GDDR7 memory will enter mass production in Q1 2025, which should be after NVIDIA launches its GeForce RTX 5090 and GeForce RTX 5080 graphics cards. If mass production is coming in Q1 2025, then we should expect that SK hynix is already sampling customers (namely NVIDIA) right now.

The Paint app in Windows 11 is going to offer smart new powers for AI-powered image generation, and we've just learned more about the requirements to benefit from this.

The Cocreator feature originally debuted in Paint as a typical text-based prompt, as seen in many incarnations across various AI offerings, late on last year as you may recall.

However, the functionality is being pepped up with the ability to sketch in a side-by-side panel to further refine the AI's creation (with a slider to allow Cocreator a freer rein if required, departing more from the sketch into its own ideas).

Up until now X users could see what other content users have Liked on the platform, but that is all about to change according to a recent announcement from the company and it's owner Elon Musk.

In the past, public personalities and politicians have been caught with their pants down, Liking content that was considered by some to be inappropriate or, at the very least, unsavory. This outing of some X users resulted in the individuals in question being attacked online, and according to X, that is going to stop as the company has now launched the ability to hide user Likes. However, this new feature is locked behind the X Premium paywall.

When the feature was announced, X took to its official account to write, "Keep spicy likes private by hiding your likes tab," and provided an image of the new profile customization feature. The social media platform further explained that users will still be able to see posts they have liked, but others won't be able to. Additionally, Like count and other metrics for your own posts will still appear under notifications. Furthermore, the author of a post will be able to see who Liked their post.

Computex 2024 featured many announcements from AMD regarding its incoming next-generation Ryzen 9000 series CPUs, and with many excited about the performance uplift in specific tasks, gamers shouldn't be, or at least if you have a specific previous-gen CPU.

Toms Hardware interviewed AMD's Senior Technical Marketing Manager for Consumer Processors, Donny Woligroski. In the interview, he told the publication that the difference between AMD's new Ryzen 9000 series CPUs and the Ryzen 7000X3D chips in gaming is very little.

Woligroski stated that while its coming Ryzen 9000 series chips won't be the Ryzen 7000X3D chips in gaming, they will still be top-of-the-line CPUs for gaming, but the difference between chips will be extremely small. However, the Ryzen 9000 series CPUs will still beat the 7000X3D chips in non-gaming tasks, meaning the Ryzen 9000 series will be the overall better CPU in terms of overall performance.

Last week amongst all of the Computex 2024 news Starship conducted the fourth test flight for the world's largest and most powerful rocket, Starship.

The company blasted Starship off from its launch facilities on June 6, with the main objectives of the test flight being flying Starship further than any test previously, and demonstrating capabilities to return and reuse Starship and its Super Heavy booster. SpaceX achieved the desired reusability demonstration by calmly splashing down Super Heavy in the Gulf of Mexico. More specifically, SpaceX was attempting to simulate a "virtual tower" with the splashdown of Super Heavy, as the next step will be landing the megalithic booster on land.

Additionally, SpaceX just wanted the Starship vehicle to survive reentry into Earth's dense atmosphere, which, during its descent, gets extremely hot from generated friction. SpaceX achieved more than it set out to do as a minimum, as Starship also splashdowned within a designated target area. Now, Elon Musk is already talking about Flight 5, and during a live stream on X, the SpaceX CEO said that Starship will be taking to the skies once again in "about a month".

Apple isn't necessarily known for making products that are great at gaming, but with the rise of the company's own in-house chips that dream is becoming more of a reality.

At WWDC Apple showed its intentions to drive deeper into the gaming market, especially considering the popularity of the App Store and all of the games it offers. However, Apple isn't just looking at mobile, as the company explained it wants to expand the possibilities of its Mac product line to gamers that want to experience full AAA titles.

Apple stated during its keynote that its MacOS named Sequoia was going to usher in some considerable upgrades for gamers, such as the inclusion of Personalized Spatial Audio, which will arrive for games that support Game Mode. Additionally, this new audio feature will be launching on iPhone. According to Apple, Personalized Spatial Audio "significantly reduces audio latency with AirPods Pro (2nd generation).

The astronauts who were part of the first all-civilian space mission to orbit have been studied by scientists, and across three studies, it was found that they've become younger.

SpaceX was behind the Inspiration4 mission that was conducted in 2021, where four civilian astronauts were transported to orbit via a SpaceX Crew Dragon capsule. The team of four only spent a few days in space and, upon return, committed their bodies to science, intending to understand how the environment in space affects human physiology. Now, three studies have been conducted, and it was found that astronauts got genetically younger during their stay in space, but the effects were only shortlived.

Scientists looked at the astronauts' DNA and found aging markers called telomeres, which are caps that protect chromosomes known to shorten with age after being exposed to certain environmental factors and stress. However, during the time spent in space, the astronauts' telomeres elongated, indicating the aging of DNA decreased. Researchers believe the telomeres getting longer is a response to the higher radiation environment in space.

According to NVIDIA and its CEO Jensen Huang, the AI PC era began with GeForce RTX and the 20 Series of desktop GPUs in 2018. For PC gamers, this meant using AI to make real-time ray-tracing a possibility with the arrival of RTX hardware and AI Tensor Cores. DLSS, or Deep Learning Super Sampling, is the not-so-secret ingredient.

Fast-forward to 2024, and AI-powered DLSS upscaling and DLSS 3 Frame Generation (which uses AI to generate entire frames) have pushed PC gaming performance to new heights across a wide range of hardware. With games like Cyberpunk 2077 and Alan Wake 2, we're now at a point where full Path Traced lighting is a reality.

At Computex 2024, we saw two examples of how NVIDIA plans to empower game developers with new AI technologies. NVIDIA ACE helps create digital humans with whom you can interact. Project G-Assist uses AI and a general model trained by NVIDIA to provide in-game assistance, gameplay guidance, and even performance optimization, all through natural dialogue or text inputs.

The US government is considering further restrictions on China's access to chip technology required for AI, aiming at new hardware that's hitting the market: GAA (Gate-All-Around) technology and HBM, two key parts of AI chips.

China would have its access to Gate-All-Around (GAA) transistor technology blocked, which is used to manufacture bleeding-edge chips, as well as access to High Bandwidth Memory (HBM), which comes in many forms: HBM3 has been powering NVIDIA's current-gen Hopper H100, while ultra-fast HBM3E is inside Blackwell B200, and HBM4 is coming shortly.

GAA nanosheet technology improves density while providing power and performance improvements, but it's only used on the most cutting-edge process nodes. Samsung has been using GAA technology with its 3nm node, Intel will use GAA in its upcoming Intel 20A node, and TSMC will use GAA with its upcoming A16 process node.

Dragon Age: The Veilguard (formerly known as Dragon Age: Dreadwolf) is the long-awaited next entry in the fantasy RPG series from BioWare, and after a tasty reveal trailer during the recent Xbox Games Showcase, we now have close to 20 minutes of gameplay to chew through ahead of the game's 'Fall 2024' release for PC, PlayStation 5, and Xbox Series X|S.

BioWare has been working on this for several years now, and even though what we see in the extended gameplay video is from the game's introduction, it all looks polished, promising, and authentic to the series (if a little cartoony). Like the original Dragon Age: Origins, you can choose your race (Human, Elf, Dwarf, Qunari), faction, and history (Grey Wardens!), which will play into relationships, conversations, and quests.

In the video, we see a familiar face in the form of Varric and the villain of the game - Solas, a pivotal character from Dragon Age: Inquisition. For fans of the series, there's some additional good news: you can import certain choices and decisions from Inquisition that will impact how the story begins.