Nvidia Ampere has been announced: here's what that means for you

[Karlston]

Nvidia Ampere has been announced: here's what that means for you

It's not the GeForce RTX 3080

(Image credit: Nvidia)

 

The Nvidia GeForce RTX 3080 may be the most anticipated product in the computing world, and while the rumor mill was guessing (and probably still is) that the next consumer-facing graphics card would be based on Ampere, that hasn't quite happened yet. 

 

Instead, Nvidia announced Ampere much in the same way it did with Volta a few years ago – built primarily for Data Center with no mention of PC gaming or GeForce. Now that AI has become so important in the world, thanks in large part to the unprecedented rise in cloud computing, Nvidia has been hard at work developing the A100 GPU, which should deliver a whopping 20x improvement in raw compute power, compared to Nvidia Volta.  

 

We don't know if Nvidia Ampere is going to be behind the next GeForce cards yet – and we wouldn't exactly rule it out – but that doesn't mean that the GPU architecture won't impact the lives of us normal folks who can't afford to drop hundreds of thousands of dollars on server equipment. 

 

These days IoT devices are piling up in everyone's homes, while companies like Amazon are creating, for example, grocery stores that let you automatically purchase products as you toss them into your shopping cart, and cars are getting ready to drive by themselves. All of this requires a ton of compute power, which is why the new 7nm Ampere architecture, along with its 3rd-generation Tensor Cores, is such a big deal. 

 

The GPU should be available soon for any business that could benefit from such sheer computing power, along with systems like the DGX A100, which packs eight A100 GPUs into a single rack that will cost an eye-watering $1 million. That sounds like a lot, but plenty of buyers for both the DGX A100 and the A100 on its own have been lined up, with the likes of Microsoft, Google, Dell and Amazon getting in on the Ampere action. 

What about GeForce?

So, Nvidia didn't announce a new gaming graphics card. It's unfortunate, sure, but we're not really expecting to see a new GeForce product until fall anyways, whether or not it's going to be based on the Ampere architecture – which isn't exactly guaranteed. 

 

 

Back when we thought the Nvidia GeForce RTX 2080 was going to be the GeForce GTX 1180, word on the street was that it was going to be based on Volta, but that didn't end up being true. Instead, that architecture was dedicated to data center products like the Tesla V100 and the Titan V. Ampere could easily follow in those same footsteps. 

 

 

 

However, Volta played an important role in Turing's development either way, as Volta cards were the first products that used the Tensor Cores that proved to be so important in the Nvidia RTX 20-series cards. The Ampere GPUs revealed today are using the third-generation Tensor Cores, which are much more powerful than the second-generation ones found in Nvidia Turing graphics cards. 

 

 

Even if the Nvidia GeForce RTX 3080 isn't based on this Ampere architecture, it's very unlikely that it won't be built on the same 7nm manufacturing process, which means efficiency and power improvements will be similar. We probably won't see a graphics card that's literally 20 times more powerful than the RTX 2080 Ti – but those rumors we saw earlier this week that suggested the 3080 Ti will be 40% faster than the current flagship are looking much more reasonable now. 

 

 

We don't know anything about the next GeForce graphics cards at the end of the day, and we won't know anything until Nvidia is ready to share some information. However, with how much more advanced its data center GPUs have become in the three years since Volta came out, we can only imagine what PC gaming is going to look like in just a few months. 

 

 

Source: Nvidia Ampere has been announced: here's what that means for you (TechRadar)

[Karlston]

Nvidia’s first Ampere GPU is designed for data centers and AI, not your PC

No new gaming PC GPUs just yet

 

Nvidia is unveiling its next-generation Ampere GPU architecture today. The first GPU to use Ampere will be Nvidia’s new A100, built for scientific computing, cloud graphics, and data analytics. While there have been plenty of rumors around Nvidia’s Ampere plans for GeForce “RTX 3080” cards, the A100 will primarily be used in data centers.

 

Nvidia’s latest data center push comes amid a pandemic and a huge increase in demand for cloud computing. Describing the coronavirus situation as “terribly tragic,” Nvidia CEO Jensen Huang noted that “cloud usage of services are going to see a surge,” in a press briefing attended by The Verge. “Those dynamics are really quite good for our data center business ... My expectation is that Ampere is going to do remarkably well. It’s our best data center GPU ever made and it capitalizes on nearly a decade of our data center experience.”

 

The A100 sports more than 54 billion transistors, making it the world’s largest 7nm processor. “That is basically at nearly the theoretical limits of what’s possible in semiconductor manufacturing today,” explains Huang. “The largest die the world’s ever made, and the largest number of transistors in a compute engine the world’s ever made.”

 

 

Nvidia is boosting its Tensor cores to make them easier to use for developers, and the A100 will also include 19.5 teraflops of FP32 performance, 6,912 CUDA cores, 40GB of memory, and 1.6TB/s of memory bandwidth. All of this performance isn’t going into powering the latest version of Assassin’s Creed, though.

 

Instead, Nvidia is combining these GPUs into a stacked AI system that will power its supercomputers in data centers around the world. Much like how Nvidia used its previous Volta architecture to create the Tesla V100 and DGX systems, a new DGX A100 AI system combines eight of these A100 GPUs into a single giant GPU.

 

The DGX A100 system promises 5 petaflops of performance, thanks to these eight A100s, and they’re being combined using Nvidia’s third-generation version of NVLink. Combining these eight GPUs means there’s 320GB of GPU memory with 12.4TB/s of memory bandwidth. Nvidia is also including 15TB of Gen4 NVMe internal storage to power AI training tasks. Researchers and scientists using the DGX A100 systems will even be able to split workloads into up to 56 instances, spreading smaller tasks across the powerful GPUs.

 

Nvidia’s recent $6.9 billion acquisition of Mellanox, a server networking supplier, is also coming into play, as the DGX A100 includes nine 200Gb/s network interfaces for a total of 3.6Tb/s per second of bidirectional bandwidth. As modern data centers adapt to increasingly diverse workloads, Mellanox’s technology will prove ever more important for Nvidia. Huang describes Mellanox as the all-important “connecting tissue” in the next generation of data centers.

 

“If you take a look at the way modern data centers are architected, the workloads they have to do are more diverse than ever,” explains Huang. “Our approach going forward is not to just focus on the server itself but to think about the entire data center as a computing unit. Going forward I believe the world is going to think about data centers as a computing unit and we’re going to be thinking about data center-scale computing. No longer just personal computers or servers, but we’re going to be operating on the data center scale.”

 

Inside Nvidia’s DGX A100 system.

Image: Nvidia

Nvidia’s DGX A100 systems have already begun shipping, with some of the first applications including research into COVID-19 conducted at the US Argonne National Laboratory.

 

“We’re using America’s most powerful supercomputers in the fight against COVID-19, running AI models and simulations on the latest technology available, like the NVIDIA DGX A100,” says Rick Stevens, associate laboratory director for Computing, Environment and Life Sciences at Argonne. “The compute power of the new DGX A100 systems coming to Argonne will help researchers explore treatments and vaccines and study the spread of the virus, enabling scientists to do years’ worth of AI-accelerated work in months or days.”

 

Nvidia says that Microsoft, Amazon, Google, Dell, Alibaba, and many other big cloud service providers are also planning to incorporate the single A100 GPUs into their own offerings. “The adoption and the enthusiasm for Ampere from all of the hyperscalers and computer makers around the world is really unprecedented,” says Huang. “This is the fastest launch of a new data center architecture we’ve ever had, and it’s understandable.”

 

Much like the larger DGX A100 cluster system, Nvidia also allows each individual A100 GPU to be partitioned into up to seven independent instances for smaller compute tasks. These systems won’t come cheap, though. Nvidia’s DGX A100 comes with big performance promises, but systems start at $199,000 for a combination of eight of these A100 chips.

 

Nvidia’s GeForce RTX 2080 graphics card.

Photo by Stefan Etienne / The Verge

It’s not clear how Nvidia will now progress Ampere directly into consumer-grade GPUs just yet. Nvidia introduced its Volta architecture, with dedicated artificial intelligence processors (tensor cores) in much the same way as today’s Ampere unveiling. But Volta didn’t go on to power Nvidia’s line of GeForce consumer products. Instead, Nvidia launched a Volta-powered $2,999 Titan V (which it called “the most powerful PC GPU ever created”) focused on AI and scientific simulation processing, not gaming or creative tasks.

 

Despite rumors of Volta powering future GeForce cards, Nvidia instead introduced its Turing architecture in 2018, which combined its dedicated tensor cores with new ray-tracing capabilities. Turing went on to power cards like the RTX 2080 instead of Volta, just weeks after Huang said the next line of graphics cards wouldn’t be launching for “a long time.” Nvidia even stripped out the RT and Tensor cores for Turing-powered cards like the GTX 1660 Ti.

 

New “RTX 3080” cards could be just months away then, but we still don’t know for sure if they’ll be using this new Ampere architecture. “There’s great overlap in the architecture, that’s without a doubt,” hinted Huang. “The configuration, the sizing of the different elements of the chip is very different.”

 

Nvidia uses HBM memory for its data center GPUs, and that’s not something the company uses for consumer PC gaming GPUs. The data center GPUs are also focused much more heavily on AI tasks and compute, than graphics. “We’ll be much more heavily biased towards graphics and less towards double-precision floating point,” adds Huang.

 

Speculation around Nvidia’s Ampere plans has intensified recently, and with the PlayStation 5 and Xbox Series X set to launch with AMD-powered GPU solutions later this year, Nvidia will surely need to have something new to offer PC gamers later this year.

 

 

Source: Nvidia’s first Ampere GPU is designed for data centers and AI, not your PC (The Verge)

[Karlston]

Don't worry, it looks like Nvidia Ampere may actually be coming to GeForce cards

It’s good news for those eyeing up the incoming GeForce RTX 3080

(Image credit: Nvidia)

 

Nvidia has seemingly confirmed that its newly-announced Ampere GPU architecture will be coming to both GeForce and data center graphics cards.

 

When it unveiled its 7nm Ampere microarchitecture earlier today, Nvidia focused entirely on the data center, with no mention of PC gaming or the consumer-facing GeForce RTX 3080. Instead, the 7nm architecture will debut with the Nvidia data center-focused A100 GPU, which should deliver an impressive 20x improvement in raw compute power compared to Nvidia Volta.

 

The company also announced that it has bundled eight of these GPUs inside the $1 million DGX A100, which can handle up to 56 tasks at once and reach up to 5 petaflops of AI performance.

 

While none of this means much to the average PC gamer, Nvidia CEO Jensen Huang said in a media briefing ahead of its virtual GTC conference that the 7nm Ampere microarchitecture will be used for all of its next-generation graphics cards, according to Market Watch – this presumably includes the next GeForce cards.

 

This doesn't reveal much about what these cards will actually look like, but we're sure Nvidia will reveal more when it shows off the cards – hopefully later this year. And we've reached out to Nvidia for clarification, and we'll update this article if we hear anything.   

All in the architecture

 

While it’s yet to release any information about consumer-facing GeForce GPUs using Ampere, when asked by a reporter in the briefing about the difference between enterprise and consumer approaches to Ampere, Huang said “there’s great overlap in the architecture, but not in the configuration.”

 

 

This now-confirmed move will mark the first time that Nvidia will unify all its products under a single microarchitecture. It’s previous generation of GPUs, for example, used Turing for its consumer-facing GeForce and Quadro CPUs, and Volta for Tesla GPUs destined for the data center.

 

 

 

The long-rumored Nvidia GeForce RTX 3080 is widely expected to be Team Green’s first consumer GPU to be based on the 7nm Ampere platform. According to a recent leak, the graphics card will boast 3,840 CUDA cores, compared to 2,944 in the RTX 2080, along with a 320-bit memory bus and probably 10GB of video RAM.

 

 

The GeForce RTX 3080, naturally, is expected to launch alongside the 3080 Ti, and a recent rumor suggests this GPU be some 40% faster than the current GeForce 2080 Ti.

 

Nvidia’s next-gen Ampere graphics cards are expected to arrive in the third quarter of 2020, starting with the higher-spec flagship models.

 

 

Source: Don't worry, it looks like Nvidia Ampere may actually be coming to GeForce cards (TechRadar)