TweakTown NewsRefine News by Category:
NVIDIA blew us all away with its reveal of the Tesla P100 during GTC 2016, but there was no word on the consumer side of things. I heard whispers during the show, but nothing concrete - what we do know, however, is that both NVIDIA and AMD will reveal their new video cards before Computex (which kicks off in the last few days of May). We chatted with Hardware Canucks during GTC 2016 about all things GPUs, something you can watch below.
It was only a few days ago that we reported that the next-gen NVIDIA GeForce video cards would feature GDDR5, but it looks like NVIDIA could tap the faster GDDR5X on its upcoming GeForce GTX 1080 and GTX 1070 (but I'm still adamant that they won't be called that).
There will be two GP104-based video cards, with the GTX 1080 featuring 8GB of GDDR5X on a 384-bit memory bus with 320GB of bandwidth. The GTX 1070 will use the same GP104 GPU, but drop down to GDDR5 with a 256-bit memory bus and 256GB/sec bandwidth. We should note that the current GM200-based GTX 980 Ti has 6GB of GDDR5 on a 384-bit memory bus with 334GB/sec - so the new GTX 1080 would have similar memory bandwidth, but with the new Pascal architecture, on the 16nm FinFET process.
GTC 2016 - During NVIDIA's GPU Technology Conference, I spent most of the week with their incredibly awesome Dmitry Novoselov. We collaborated during the show, chatting about the new Pascal-based Tesla P100, and more during Hardware Canucks' latest video. Check it out, below!
During the show, NVIDIA unveiled the new Pascal architecture in better detail - with the new Tesla P100 unveiled in full detail. During the video, Dimitry asked what I think the new GeForce cards would be called - where I said GeForce 10K. We have the GTX 900 series, so we'd expect the GeForce 1080 series - but 1080 is too close to 1080p, and that makes the new cards seem mid-range.
We will be working with HWC more in the future, so expect more content between us as we head towards Computex in late May.
AMD's new Radeon Pro Duo is a dual-GPU video card with two Fiji GPUs, 8GB of HBM1 (4GB per GPU), four DisplayPort outputs, and 16 TFlops of performance. Well, it hits retail on April 26 with a not-so-bad price tag of $1500.
This isn't a video card for most people, as it'll be the glory card for all things VR - gaming, production and engineering with VR, as well as the insane enthusiasts (like myself) who just want the most powerful video cards they can get.
NVIDIA unveiled its impressive Tesla P100 video card at GTC 2016, powered by the new Pascal architecture. P100 is built from the impressive 16nm FinFET process, as well as the hugely transformative HBM2.
Well, what about the consumer side? Well, the GP104 will reportedly be unveiled later this month or early May, but new reports are pointing towards NVIDIA using GDDR5 on the next-gen GeForce video cards. We should expect 8GB of newer, denser, faster GDDR5 RAM. Both the GP104- and GP106-based solutions should feature GDDR5.
The newer GDDR5 should be clocked at 8Gbps, which will provide a decent upgrade over the 7Gbps on the current GDDR5 used in the GTX 980 Ti and Titan X. We will bring you up to date information on all next-gen video cards, so expect a bigger wave of GPU related news in the coming months.
Multiple sources around the world, as well as our own, are saying that NVIDIA will unveil their next-gen video cards in the coming weeks. With Computex right around the corner, this makes perfect sense.
We've already seen the company unveil its new Tesla P100 video card, powered by the new Pascal architecture, rocking HBM2 and the spiffy new 16nm FinFET process. The P100 features 15 billion transistors on the GPU itself, with another 135 billion transistors for the rest of the card - thanks to the incredible HBM2.
Most sites are still reporting that NVIDIA will call the new cards the GeForce GTX 1080 and GTX 1070, but I don't think we'll see this happen. I think we'll see NVIDIA move towards something like GeForce 10K, which sounds much better.
GTC 2016 - On day two of NVIDIA's GPU Technology Conference, we walked around the exhibition halls with the super-awesome Dimitry from Hardware Canucks and ran into the SK Hynix booth. What did we see? HBM2. HBM1 was there side-by-side with a 1GB chip, while the HBM2 chip was a much larger 4GB chip.
Last year at GTC 2015, we did the same thing and walked into HBM1 - because at the time it was only becoming available, and was still months away from AMD using it on their Fiji-based Radeon R9 Fury X. Fast forward a few months, and HBM1 made its way onto the Fury X.
NVIDIA announced its new GP100-based Tesla P100 video card yesterday, which is the first video card to feature the Pascal architecture, and HBM2. Seeing HBM2 at GTC 2016 was great, as we could see how small it was in person, but we weren't allowed to get close to it - which is why the image isn't exactly up close. If we get closer later in the day, we'll take another photo and update the article.
Remember that HBM2 has bandwidth of up to 1TB/sec, with the Tesla P100 hitting 720GB/sec - an insane jump on the GeForce GTX Titan X with its 334GB/sec (with GDDR5) and even the HBM1-based R9 Fury X and its 512GB/sec.
GTC 2016 - NVIDIA just unveiled its next-generation Pascal architecture at GTC 2016, showcasing its new Tesla P100 video card - with an insane amount of power. The Tesla P100 has 16GB of HBM2, with an insane amount of horsepower behind it.
Well, NVIDIA also announced its new DGX-1 - the world's first Deep Learning supercomputer. DGX-1 packs 8x Tesla P100 video cards, each with 16GB of HBM2. DGX-1 has 7TB of SSDs, 2 x Intel Xeon processors, 170 TFlops of performance, Quad IP 100Gbps and dual 10GbE connectivity. It uses just 3200W of power (3.2kW) which might sound like a lot to you and I, but this is a supercomputer... not a traditional rackmount PC.
NVIDIA's new Pascal-based DGX-1 is priced at $129,000.
GTC 2016 - NVIDIA's GPU Technology Conference kicked off today, where the company talked about its work with AI and deep learning were a large focus, as well as VR being a large part of NVIDIA's opening keynote at GTC 2016. But the star of the show was the new Tesla P100.
We have the new Tesla P100, which is "the most advanced hyper scale datacenter GPU ever built". The Pascal chip itself is 600mm2 and built on the incredible 16nm FinFET process with the Tesla P100 packing a how-did-they-do-it 150 billion transistors (150,000,000,000), with the super-fast HBM2 memory included. The new Pascal-based Tesla P100 is in volume production right now, which is incredibly exciting news as it means HBM2 is ready right now - and NVIDIA has it.
Now, the 150 billion transistors sounds absolutely mind-blowing - as its 1775% more than the 8 billion that the GM200 includes (the GPU that powers the GTX 980 Ti and Titan X). We chatted with Rob and Fudo from Fudzilla, who say that the Tesla P100's 150 billion transistors, is made up from 17 billion on the GPU - with the rest of the transistors coming from the HBM2. This makes much more sense.
The GP100 itself packs 3584 stream processors, with a base clock of 1328MHz and 1480MHz boost. The 4096-bit memory bus provides a claimed 720GB/sec bandwidth, with a 300W TDP.
NVIDIA said they'll be shipping it "soon", but didn't provide a date on Tesla P100.
AMD has today released its Radeon Software Crimson Edition 16.4.1 drivers.
With it comes up to 35% better performance in Quantum Break with the R9 Fury X, a fix for framerate capping issues in some DirectX 12 apps, and a fix for flickering in Hitman when using DirectX 11 mode and high shadow settings; support for HTC Vive and Oculus Rift is listed again for some reason (it was introduced last week in the 16.3.2 drivers).
GTC 2016 - World Exclusive - We've just had some breaking news delivered to us, on NVIDIA's next-generation video card. We don't have specifics to share with you right now, but we do know that it's going to be what we expected: a freakin' monster.
I expected NVIDIA's next-gen video card to be something quite special, thanks to it being based on the Holy Trinity of GPU upgrades - the shift from the 28nm to 16nm FinFET process, the new exciting Pascal architecture and the insanely fast HBM2 memory. Well, according to our exclusive source - NVIDIA's new video card was right on the expectations that I was able to summarise about in the last few months.
We should expect the unannounced video cards to be a beast, with an option between 8/16GB of VRAM or 16/32GB of VRAM. The performance numbers have not been unveiled, and neither has the new naming or branding for the card - but we do know it's going to be insanely powerful. I'm expecting it to be ~1.5x the power of the GeForce GTX Titan X, and the AMD Radeon R9 Fury X (which runs HBM1 @ 512GB/sec).
The opening keynote by NVIDIA co-founder and CEO Jen-Hsun Huang will kick off in less than 8 hours from now at 9AM PST, where we will be there covering it all live for you. I'm beyond excited.