NVIDIA today unveiled the world's first cloud-native, multi-tenant AI supercomputer—the next-generation NVIDIA DGX SuperPOD featuring NVIDIA BlueField -2 DPUs. Fortifying the DGX SuperPOD with BlueField-2 DPUs—data processing units that offload, accelerate and isolate users' data—provides customers with secure connections to their AI infrastructure.

The company also announced NVIDIA Base Command, which enables multiple users and IT teams to securely access, share and operate their DGX SuperPOD infrastructure. Base Command coordinates AI training and operations on DGX SuperPOD infrastructure to enable the work of teams of data scientists and developers located around the globe.

Remember that time when NVIDIA CEO Jensen Huang pulled the company's HGX system (based on Ampere GA100) out of his own oven? Well, it seems we might be looking at something similar come GTC, which is set for April 12th. The event will again take place solely via online transmissions and announcements due to the still heavily-grassing COVID-19 pandemic, and NVIDIA teased a new kick-off presentation from its CEO - from his kitchen.

NVIDIA has usually taken GTC as an opportunity to showcase products not for the consumer segment, however, so temper your expectations. You should also likely temper your expectations regarding Huang taking enough RTX 30-series graphics cards form his oven to satisfy the incredible demand and stock issues we've been seeing with NVIDIA's latest and greatest. What we could be looking for (as an appetizer to this year's over 1500 presentations) are products focused on the professional computing markets, from quantum computing to AI, passing through RTX Ampere accelerators for professionals, NVIDIA Drive products, as well as Jetson announcements. NVIDIA did warn some surprises might be in store, but again, this doesn't mean these are surprises aimed at consumer gaming products. Catch up on NVIDIA's official announcement and a previously-released GTC retrospective video after the break.

NVIDIA today announced NVIDIA AI Enterprise, a comprehensive software suite of enterprise-grade AI tools and frameworks optimized, certified and supported by NVIDIA, exclusively with VMware vSphere 7 Update 2, separately announced today.

Through a first-of-its-kind industry collaboration to develop an AI-Ready Enterprise platform, NVIDIA teamed with VMware to virtualize AI workloads on VMware vSphere with NVIDIA AI Enterprise. The offering gives enterprises the software required to develop a broad range of AI solutions, such as advanced diagnostics in healthcare, smart factories for manufacturing, and fraud detection in financial services.

NVIDIA has launched its GeForce RTX 3000 series of graphics cards based on the Ampere architecture three months ago. However, we are already getting information about the next-generation that the company plans to introduce. In the past, the rumors made us believe that the architecture coming after Ampere is allegedly being called Hopper. Hopper architecture is supposed to bring multi-chip packaging technology and be introduced after Ampere. However, thanks to @kopite7kimi on Twitter, a reliable source of information, we have data that NVIDIA is reportedly working on a monolithic GPU architecture that the company internally refers to as "ADxxx" for its codenames.

The new monolithically-designed Lovelace architecture is going make a debut on the 5 nm semiconductor manufacturing process, a whole year earlier than Hopper. It is unknown which foundry will manufacture the GPUs, however, both of NVIDIA's partners, TSMC and Samsung, are capable of manufacturing it. The Hopper is expected to arrive sometime in 2023-2024 and utilize the MCM technology, while the Lovelace architecture will appear in 2021-2022. We are not sure if the Hopper architecture will be exclusive to data centers or extend to the gaming segment as well. The Ada Lovelace architecture is supposedly going to be a gaming GPU family. Ada Lovelace, a British mathematician, has appeared on NVIDIA's 2018 GTC t-shirt known as "Company of Heroes", so NVIDIA may have already been using the ADxxx codenames internally for a long time now.

During its GPU Technology Conference (GTC) in China, Mr. Bill Dally—NVIDIA's chief scientist and SVP of research—has presented many interesting things about how the company plans to push the future of HPC, AI, graphics, healthcare, and edge computing. Mr. Dally has presented NVIDIA's research efforts and what is the future vision for its products. Among one of the most interesting things presented was a plan to ditch the standard electrical data transfer and use the speed of light to scale and advance node communication. The new technology utilizing optical data transfer is supposed to bring the power required to transfer by a significant amount.

The proposed plan by the company is to use an optical NVLink equivalent. While the current NVLink 2.0 chip uses eight pico Joules per bit (8 pJ/b) and can send signals only to 0.3 meters without any repeaters, the optical replacement is capable of sending data anywhere from 20 to 100 meters while consuming half the power (4 pJ/b). NVIDIA has conceptualized a system with four GPUs in a tray, all of which are connected by light. To power such a setup, there are lasers that produce 8-10 wavelengths. These wavelengths are modulated onto this at a speed of 25 Gbit/s per wavelength, using ring resonators. On the receiving side, ring photodetectors are used to pick up the wavelength and send it to the photodetector. This technique ensures fast data transfer capable of long distances.

NVIDIA today announced that it will be hosting another GTC keynote for the coming month of October. To be held between October 5th and October 9th, the now announced keynote will bring updates to NVIDIA's products and technologies, as well as provide an opportunity for numerous computer science companies and individuals to take center stage on discussing new and upcoming technologies. More than 500 sessions will form the backbone of GTC, with seven separate programming streams running across North America, Europe, Israel, India, Taiwan, Japan and Korea - each with access to live demos, specialized content, local startups and sponsors.

This GTC keynote follows the May 2020 keynote where the world was presented to NVIDIA's Ampere-based GA100 accelerator. A gaming and consumer-oriented event is also taking place on September 1st, with expectations being set high for NVIDIA's next-generation of consumer graphics products. Although if recent rumors of a $2,000 RTX 3090 graphics card are anything to go by, not only expectations will be soaring by then.

Thanks to the sources of VideoCardz, we now have the first picture of the next-generation NVIDIA Tesla A100 graphics card. Designed for computing oriented applications, the Tesla A100 is a socketed GPU designed for NVIDIA's proprietary SXM socket. In a post few days ago, we were suspecting that you might be able to fit the Tesla A100 GPU in the socket of the previous Volta V100 GPUs as it is a similar SXM socket. However, the mounting holes have been re-arranged and this one requires a new socket/motherboard. The Tesla A100 GPU is based on GA100 GPU die, which we don't know specifications of. From the picture, we can only see that there is one very big die attached to six HBM modules, most likely HBM2E. Besides that everything else is unknown. More details are expected to be announced today at the GTC 2020 digital keynote.

NVIDIA is rumored to introduce their next-generation Ampere architecture very soon, at its GTC event happening on May 14th. We're expecting to see an announcement for the successor to the company's DGX lineup of pre-built compute systems—using the upcoming Ampere architecture of course. At the heart of these machines, will be a new GA100 GPU, that's rumored to be very fast. A while ago, we've seen NVIDIA register a trademark for "DGX A100", which seems to be a credible name for these systems featuring the new Tesla A100 graphics cards.

Today, NVIDIA's CEO was spotted in an unlisted video that's published on the official NVIDIA YouTube channel. It shows him pulling out of the oven what he calls "world's largest GPU", that he has been cooking all the time. Featuring eight Tesla A100 GPUs, this DGX A100 system appears to be based on a similar platform design as previous DGX systems, where the GPU is a socketed SXM2 design. This looks like a viable upgrade path for owners of previous DGX systems—just swap out the GPUs and enjoy higher performance. It's been a while since we have seen Mr. Huang appear with his leather jacket, and in the video, he isn't wearing one, is this the real Jensen? Jokes aside, you can check out the video below, if it is not taken down soon.Update May 12th, 5 pm UTC: NVIDIA has listed the video and it is not unlisted anymore.

NVIDIA has announced that their annual GTC (Graphics Technology Conference) keynote will be held on May 14th, via a fully online distribution through YouTube set for 6 a.m. PST. Originally canceled due to the COVID-19 pandemic, NVIDIA planned to make a digital-only event covering GTC on the same day of March 23rd, but elected to postpone it since "there were no announcements that were too critical not to be able to be told at a later time". The "Get AMPED" catchphrase does seem to point to some Ampere announcements, though, and the entire PC enthusiast community very much thinks that is one critical announcement.

The GTC 2020 note will be a recorded announcement led by NVIDIA CEO Jensen Huang. The leather jacket-toting CEO will highlight the company's latest innovations in AI, high performance computing, data science, autonomous machines, healthcare and graphics during the recorded keynote, which will be made available, on demand, on YouTube. Starting on May 19th, though, the graphics technology event will continue via live webinars, demos, and recorded talks discussing the very latest in AI, HPC, data science, graphics, and more.

NVIDIA was supposed to launch its next-generation "Ampere" lineup of server-class GPUs at this year's GTC event. However, NVIDIA's president and CEO, Jensen Huang, shared the news that NVIDIA will now be showcasing the important GTC news it has prepared, but instead will only provide online materials from developers and researchers from all over the world. In the blog update, Mr. Huang said the following: "We have exciting products and news to share with you. But this isn't the right time. We're going to hold off on sharing our GTC news for now. That way, our employees, partners, the media and analysts who follow us, and our customers around the world can focus on staying safe and reducing the spread of the virus."

"We will still stream tons of great content from researchers and developers who have prepared great talks. This is a time to focus on our family, our friends, our community. Our employees are working from home. Many hourly workers will not need to work but they'll all be fully paid. Stay safe everyone. We will get through this together.", he added. While we won't get any new GPUs judging by this announcement, we can expect to see them once this situation is resolved.

NVIDIA's ANZ (Australia and New Zealand) Twitter handle posted a curious-looking teaser of something to look forward to on March 19. The now-deleted Tweet contained an animation showing the shape of an eye tracking two points of light crossing its iris. This date falls just a couple of days ahead of the March 22 debut of GTC 2020, the NVIDIA-promoted trade-show covering all things graphics tech. As many on the tweet and online forums discussing have pointed out, the announcement could probably be related to a commercial release of NVIDIA's foveated rendering technology that was demonstrated way back at SIGGRAPH 2016, at a time when VR and related technologies were all the rage.

The concept of foveated rendering is straightforward: eye-tracking is leveraged to ensure that areas of the frame you're looking at are rendered in greater detail than the others, to conserve system resources and improve performance. This is achieved by tracking your foveal vision (the primary part of your vision focused on detail), while shedding resources on parts of the frame that fall within your peripheral vision. Compared to 2016, rendering technologies to facilitate foveated rendering have advanced. A related technology is variable shading rate (VRS), which allows an application to render different parts of the scene at different levels of detail, and improve performance.A video presentation detailing NVIDIA's foveated rendering tech from 2016 follows.

NVIDIA today announced that, in light of the spread of the coronavirus, it is deferring plans to deliver a webcast keynote as part of the digital version of its GPU Technology Conference later this month. The company will, instead, issue on Tuesday, March 24, news announcements that had been scheduled to be shared in the keynote. This will be followed by an investor call with NVIDIA founder and Chief Executive Officer Jensen Huang, which will be accessible to other listeners. The call will begin at 8am Pacific time and will be accessible at investor.nvidia.com.

NVIDIA had initially planned to host its 11th annual GTC on March 22-26 at the San Jose Convention Center, with an anticipated audience of 10,000. Amid the worsening coronavirus situation, the company said earlier this month that it was shifting much of the conference to its digital platform. The company believes that continuing public health uncertainties would challenge its ability to produce and deliver a digital keynote. Other components of GTC Digital will still take place, including live webinars, recorded talks and panels, research posters, trainings, and Connect with Experts sessions available starting Wednesday, March 25. More details will be provided in the coming days.

(Update, March 4th: Another NVIDIA graphics card has been discovered in the Geekbench database, this one featuring a total of 124 CUs. This could amount to some 7,936 CUDA cores, should NVIDIA keep the same 64 CUDA cores per CU - though this has changed in the past, as when NVIDIA halved the number of CUDA cores per CU from Pascal to Turing. The 124 CU graphics card is clocked at 1.1 GHz and features 32 GB of HBM2e, delivering a score of 222,377 points in the Geekbench benchmark. We again stress that these can be just engineering samples, with conservative clocks, and that final performance could be even higher).

NVIDIA is expected to launch its next-generation Ampere lineup of GPUs during the GPU Technology Conference (GTC) event happening from March 22nd to March 26th. Just a few weeks before the release of these new GPUs, a Geekbench 5 compute score measuring OpenCL performance of the unknown GPUs, which we assume are a part of the Ampere lineup, has appeared. Thanks to the twitter user "_rogame" (@_rogame) who obtained a Geekbench database entry, we have some information about the CUDA core configuration, memory, and performance of the upcoming cards.

NVIDIA today announced that their yearly GPU Technology Conference (GTC) will be held online, instead of locally at the San Jose Convention Center as previously planned. The company cites Covid-19 (colloquially called Coronavirus) concerns as the root cause - the health of both its professionals and attendees is paramount for the company, and as such, an online event seems more inline with the state of the world.

In the press note, NVIDIA also said that NVIDIA founder and CEO Jensen Huang will still deliver a keynote address, which will be available exclusively by livestream. All those who registered for a GTC pass will receive a full refund. This is yet another event that has been impacted by the recent virus outbreak that has the world's economies teetering, but for the more cynical among us, it could be the beginning of a new way of doing this sorts of events: online, rather than locally, thus reducing the required investment and logistics usually associated with such.

NVIDIA will kick off the 2019 GPU Technology Conference later today, at 2 PM Pacific time. The company is expected to either tease or unveil a new graphics architecture succeeding "Volta" and "Turing." Not much is known about this architecture, but it's highly likely to be NVIDIA's first to be designed for the 7 nm silicon fabrication process. This unveiling could be the earliest stage of the architecture's launch cycle, would could see market availability only by late-2019 or mid-2020, if not later, given that the company's RTX 20-series and GTX 16-series have only been unveiled recently. NVIDIA could leverage 7 nm to increase transistor densities, and bring its RTX technology to even more affordable price-points.

NVIDIA has posted a short (literally short) teaser, treating users to the promise of the "Ultimate Gaming Experience". This might mean something, such as the new, expected NVIDIA 11** series of graphics cards... Or it may mean something much less exciting, and have something to do with the 4K, HDR gaming experience that is supposed to be reaching gamers in a couple of weeks, at an expected cost of more kidneys than the average human has.

Officially, though, GTC 2018 Taiwan will revolve around artificial intelligence tech (what doesn't these days, really?) Translated, the teaser image reads something along the lines of "Utilizing GPU computing to explore the world's infinite possibilities - witness the power of artificial intelligence and the ultimate gaming experience in GTC Taiwan and Computex 2018." Remember, however, that marketing almost always has a way of blowing things out of proportion - don't hold your breath for a new graphics card series announcement.

NVIDIA made big moves to bring a semblance of real-time ray-tracing to the masses, with the new RTX technology, as part of its efforts to replace rasterized rendering, which has dominated 3D graphics for the past three decades. Microsoft has come out with its own extension to DirectX 12, with the new DXR API. NVIDIA is now reportedly working with the Khronos Group to bring RTX to Vulkan.

A new Vulkan extension titled "VK_NV_raytracing" surfaced in tech-documents accessed by Phoronix, which is the company's contribution to a multi-vendor standard for ray-tracing, being developed by the Khronos Group. This extension could expose several NVIDIA RTX features and presets to Vulkan. It also has similar code-structures to DXR, to minimize duplication of effort, or skill-building. NVIDIA will detail its adaptation of RTX to Vulkan further at GTC.

Despite not being backed by an official statement, NVIDIA's next-generation crypto-mining Turing graphics cards are expected to be revealed at GTC 2018 between March 26 to 29. According to DigiTimes's latest report, NVIDIA is expecting a drop in demand for graphics card later this year. In an effort to prolong the lifecycle of their current graphics cards, mass production for Turing won't start until the third quarter of 2018. Sources around the industry have also revealed that NVIDIA had a sit-down with AIB partners to address the current situation. In short, NVIDIA partners are now forbidden to promote activities related to cryptocurrency mining and to sell bulks of consumer graphics cards to cryptominers.

NVIDIA will host thousands of the world's leading AI experts at its ninth annual GPU Technology Conference (GTC) on March 26-29 at the San Jose McEnery Convention Center. NVIDIA founder and CEO Jensen Huang will deliver a keynote address on Tuesday, March 27, at 9 a.m. Pacific time to an expected 8,000 attendees representing the diverse, rapidly expanding AI and GPU computing community.

"GTC is where the world's leading researchers and business leaders learn how to harness the power of AI," said Greg Estes, vice president of Developer Programs at NVIDIA. "As GPU computing continues to drive the AI revolution, GTC is where you'll see the future take shape."

It's being reported by Tom's Hardware, citing industry sources, that NVIDIA isn't looking to expand upon its graphics cards lineup at this years' GDC (Game Developers Conference) or GTC (GPU Technology Conference). Even as reports have been hitting the streets that pointed towards NVIDIA announcing (if not launching) their two new product architectures as early as next month, it now seems that won't be the case after all. As a reminder, the architectures we're writing about here are Turing, reportedly for crypto-mining applications, and Ampere, the expected GeForce architecture leapfrogging the current top of the line - and absent from regular consumer shores - Volta.

There's really not much that can be gleaned as of now from industry sources, though. It's clear no one has received any kind of information from NVIDIA when it comes to either of their expected architectures, which means an impending announcement isn't likely. At the same time, NVIDIA really has no interest in pulling the trigger on new products - demand is fine, and competition from AMD is low. As such, reports of a June or later announcement/release are outstandingly credible, as are reports that NVIDIA would put the brakes on a consumer version of Ampere, use it to replace Volta on the professional and server segment, and instead launch Volta - finally - on the consumer segment. This would allow the company to cache in on their Volta architecture, this time on consumer products, for a full generation longer, while innovating the market - of sorts. All scenarios are open right now; but one thing that seems clear is that there will be no announcements next month.

NVIDIA has been on a roll lately with their market capitalization and share valuation, which could very well send the company soaring past the $100 billion dollar mark today. Whether or not that happens (and if it does, it will be a historical milestone for the company), NVIDIA's growth of almost $25 billion dollars since May 13th is nothing short of jaw-dropping.

The "sudden" market valuation on NVIDIA comes on the heels of the company's strong graphics execution and increasingly entrenched position in the high performance GPU computing market for machine learning. The company's Volta architecture, which was showcased by Jensen Huang at their GTC keynote on May 10th, boosted confidence in the company significantly. Since then, the company's market cap has increased from the $75 billion dollar it was at shortly after GTC, towards its $96.31 billion rated market cap today. More recently, with the recent rise of the crypto wave craze, NVIDIA's GPUs have been talked about as real alternatives to AMD's previously (and perhaps hurtful for the company) grasp on this kind of workloads.

Further planting its roots on the VR SDK and development field, NVIDIA has just announced availability of two more SDK packages, for their VRWorks Audio and 360 Video suites. Now a part of NVIDIA's VRWorks suite of VR solutions, the VRWorks Audio SDK provides real-time ray tracing of audio in virtual environments, and is supported in Epic's Unreal Engine 4 (here's hoping this solution, or other solutions similar to it, address the problem of today's game audio.) The VRWorks 360 Video SDK, on the other hand, may be less interesting for graphics enthusiasts, in that it addresses the complex challenge of real-time video stitching.

Traditional VR audio ( and gaming audio, for that matter) provide an accurate 3D position of the audio source within a virtual environment. However, as it is handled today, sound is processed with little regard to anything else but the location of the source. With VRWorks Audio, NVIDIA brings to the table considerations for the dimensions and material properties of the physical environment, helping to create a truly immersive environment by modeling sound propagation phenomena such as reflection, refraction and diffraction. This is to be done in real time, at a GPU level. This work leverages NVIDIA's OptiX ray-tracing technology, which allows VRWorks Audio to trace the path of sound in real time, delivering physically accurate audio that reflects the size, shape and material properties of the virtual environment.

At its inaugural European edition of the Graphics Technology Conference (GTC), NVIDIA announced Xavier, an "AI supercomputer for the future of autonomous transportation." An evolution of its Drive PX2 board that leverages a pair of "Maxwell" GPUs with some custom logic and an ARM CPU, to provide cars with the compute power necessary to deep-learn the surroundings and self-drive, or assist-drive; Xavier is a refinement over Drive PX2 in that it merges three chips - two GPUs and one control logic into an SoC.

You'd think that NVIDIA refined its deep-learning tech enough to not need a pair of "Maxwell" SoCs, but Xavier is more than that. The 7 billion-transistor chip built on 16 nm FinFET process, offers more raw compute performance thanks to leveraging NVIDIA's next-generation "Volta" architecture, one more advanced than even its current "Pascal" architecture. The chip features a "Volta" GPU with 512 CUDA cores. The CVA makes up the vehicle I/O, while an image processor that's capable of 8K HDR video streams feeds the chip with visual inputs from various cameras around the vehicle. An 8-core ARM CPU performs general-purpose compute. NVIDIA hopes to get the first engineering samples of Xavier out to interested car-makers by Q4-2017.

NVIDIA's flagship dual-GPU graphics card, the GeForce GTX TITAN-Z, was expected to go on sale later today. That launch is now delayed, according to a SweClockers report. The three thousand Dollar question is why. According to some sources, NVIDIA is effecting a last minute design change that sees a meatier cooler on the card, than the one Jen-Hsun Huang rafikied to the press at GTC 2014.

There may have been a last-minute realization at Santa Clara, that the card - as presented at GTC - may not cut it in the ring against AMD's Radeon R9 295X2, or at least it won't be able to warrant its vulgar $3000 price tag, against the R9 295X2's $1500; despite AMD's rather messy three-piece approach to its liquid-cooled product (the card itself, a radiator, and coolant tubing), and so NVIDIA could be redesigning the GTX TITAN-Z with an even bigger cooler, to facilitate higher clock speeds.

NVIDIA updated its GPU micro-architecture roadmap at the 2013 GPU Technology Conference (GTC). Currently spearheaded by the "Kepler" micro-architecture, which drives its GeForce, Quadro, and Tesla product lines, and which will drive Tegra mobile SoCs in 2014; NVIDIA's next-generation "Maxwell" could make its debut some time in 2014. Going by NVIDIA's graph that puts performance-per-Watt against time, "Maxwell" should nearly double performance. Maxwell GPUs feature unified virtual memory, which lets CPUs treat graphics card memory as system memory, for faster general-purpose performance.

Although not backed by a tentative launch year, and assuming "Maxwell" has its run for another two years, 2016 could see the launch of NVIDIA's "Volta" GPU micro-architecture. In addition to advancements by its predecessors, "Volta" could introduce stacked DRAM technology. It would enable GPU memory bandwidths as high as 1 TB/s. Current high-end graphics cards such as GeForce Titan and Radeon HD 7970 GHz Edition, are capable of breaching the 300 GB/s mark, so NVIDIA's claims don't sound far-fetched.