GTC 2017 - SK Hynix is displaying its next-gen GDDR6 technology at NVIDIA's GPU Technology Conference, their new memory standard that will power NVIDIA's next-gen Volta GPU architecture, with Volta-based graphics cards set to arrive in 2018.
NVIDIA is currently using various different RAM standards on their graphics cards, with their Tesla P100 graphics card using HBM2 (yes, before AMD's upcoming Radeon RX Vega), GDDR5X at 11Gbps on the TITAN Xp, Titan X(P), GeForce GTX 1080 Ti, and the new GTX 1080 11Gbps model, and then GDDR5 for the rest of its graphics cards.
SK Hynix displayed the differences between GDDR5 and GDDR6, with some huge increases in speeds and bandwidth - and up to 16Gb chips, while offering less power consumption.
AMD will be releasing its next-gen Radeon RX Vega family of graphics cards in the next few weeks, and now I've had an exclusive industry source who has told me that AMD will have only a handful of Radeon RX Vega graphics cards at launch.
I've been told that there will be less than 16,000 cards that will ship in the first few months after it launches, something that will come down to the HBM2 used on the card. HBM2 is in extremely limited supply, and is expensive to use - and since there's not enough, that scarcity is driving up the production costs of the card - and will see AMD only having 16,000 cards or so in the months post-launch.
If this is true, AMD could be in a very rough spot with Radeon RX Vega - especially if it was to deliver on performance. There are hundreds of thousands of thirsty Radeon fans that want a next-gen graphics card, and the hype train for Radeon RX Vega is simultaneously withering out - and burning hotter than the sun. Personally, I want AMD to hit a home run with Radeon RX Vega - but at the same time, NVIDIA has completely secured the high-end market with the GTX 1080 and GTX 1080 Ti.
We can expect a family of cards, with the latest rumors on performance of what should hopefully be the GTX 1070 competitor in RX Vega form above. It's a prototype card, and a rumored benchmark run - but, if it's true - how many of these cards would fill up that 16,000 quantity if there are other higher-end SKUs made available in the Radeon RX Vega family.
There have been a few leaks on AMD's upcoming Radeon RX Vega graphics card, and now we're seeing the 687F:C1 device ID once again, with some details on its HBM2 speeds and GPU clocks.
First off, I've been reporting for a while now that there will be a family of Radeon RX Vega graphics cards, with a new Vega 10 prototype being used and spotted on 3DMark's Fire Strike database. The card in question had its Vega 10 GPU clocked at 1200MHz, while its 8GB of HBM2 was clocked at 700MHz.
The Radeon RX Vega prototype was benchmarked with AMD's Ryzen 7 1800X processor, but how was the result of its run in Fire Strike?
AMD has said previously that it will be launching its next-gen Radeon RX Vega graphics cards in Q2 2017, with just over 6 weeks to go - we're cutting it close. AMD has said that it will be unveiling new information on its next-gen GPU architecture, and not just Vega - but Navi, which is destined for 2018/2019.
RTG boss Raja Koduri and Computing & Graphics boss Jim Anderson will join AMD CEO Lisa Su on stage at AMD's HQ in Sunnyvale to talk about the company's long-term vision, including Vega, Navi, and even Zen+.
As for what to expect on May 16, we should receive a few more details on Vega - but this isn't a launch, so we might just receive another few drips of info. We might see a revised GPU roadmap for Navi, with a tease of the 'next-gen' memory that AMD will be using on it, while Zen+ could be better detailed now that Ryzen is finally here.
MSI is continuing to add to its roster of custom GeForce GTX 1080 Ti graphics cards with the announcement of the GTX 1080 Ti Sea Hawk EK X, the first custom GTX 1080 Ti with a full-cover water block.
MSI's upcoming GTX 1080 Ti Sea Hawk EK X features a custom PCB, 8+2-phase power design, and 8+8-pin PCIe power connectors.
EK Water Blocks designed the water block, which features a gorgeous MSI dragon and logo, as well as NVIDIA's now-required GeForce GTX logo on the front. There's also a nickel-plated base cover on the GPU, memory, and VRMs.
As for the clock speeds, we're looking at the same overclocks as the Gaming X and Sea Hawk X cards - 1569/1683MHz for base/boost, respectively - while the 11GB of GDDR5X is clocked up to 11120MHz, an increase of 1%.
Manli has just unveiled their first custom GeForce GTX 1080 Ti, with the new GTX 1080 Ti Gallardo graphics card - with Gallardo in Italian meaning "elegant" and "bravery".
From the get go, Manli provides an interesting box for their GTX 1080 Ti Gallardo with gold and silver elements that match the "elegant" and "bravery" side of the card. The card itself features a slight overclock on the GPU, with clocks of 1531/1645MHz for GPU base/boost, respectively.
Manli has used a triple-fan cooling solution on their GTX 1080 Ti Gallardo, and from the images, it looks quite long, too.
AMD is preparing Radeon gamers for Prey with the release of the new Radeon Software Crimson ReLive 17.5.1 beta drivers, which even include some performance optimizations for the just-released Radeon RX 580 graphics card.
The new drivers feature a decent 4.7% performance boost for the RX 580 in Prey, but they also provide multi-GPU support for CrossFire owners who want to see additional performance from their RX 580 CF rig. There are the usual bug fixes and improvements in these new drivers as well, and AMD has also patched their auto updater so it no longer fails.
You can grab the new 17.5.1 drivers right here.
Colorful has one of the most interesting, and what will surely be one of the best GeForce GTX 1080 Ti graphics cards on its hands with the iGame GTX1080Ti Vulcan X OC.
The iGame GTX 1080 Ti Vulcan X OC has GPU clocks of 1620/1733MHz, and the usual 11GB of GDDR5X at stock 11Gbps on the 352-bit memory bus. Colorful deploys its impressive ROG-like SWORIZER cooler, but it's the inclusion of the built-in iGame Status Monitor, the LCD display, that has us excited.
Owners of the Colorful iGame GTX 1080 Ti Vulcan X OC will be able to see their GPU clocks and temps, as well as the fan speed and memory used on the LCD display - as well as a load level indicator.
There was a new Direct Rendering Manager (DRM) update for Linux pushed out yesterday, with AMD sliding in some major Vega feature support into the open source OS.
We now know that AMD have some interesting things planned for Vega, including GPU sensors, partial resident textures, network visualization, non-contiguous VRAM mapping, and more. But it's the internal specs of Vega 10 that we're all here for, so without getting your excitement meter up too much - please, take some of this salt and throw it over your shoulder.
Vega 10 will supposedly rock 64 next-gen compute units, each with 64 GCN stream processors - with a total of 4096 next-gen GCN stream processors in 4 divisions, each with a single shader engine. Every 1024 stream processor shader engine has two Asynchronous Compute Units, one render back end and 4 texture blocks.
Inside of each texture block are 16 texture mapping units, providing a total of 256 TMUs - while Vega 10 has the ability of supporting 8 independent work threads simultaneously. With Vega 10 clocked at 1.5GHz, we could expect a monstrous 12.5 TFLOPs of FP32 compute performance, and the high-speed 8GB of HBM2 with what I think will be the start of the show in High Bandwidth Cache (HBC), AMD could have one of the fastest graphics cards on the market with its Radeon RX Vega.
We've been hearing more and more about AMD's first truly next-gen GPU architecture in years, Vega - but now the company is getting official in PR statements for the Q1 2017 results.
In AMD's recent Q1 2017 highlights release, the company said: "AMD's "Vega" GPU architecture is on track to launch in Q2, and has been designed from scratch to address the most data- and visually-intensive next-generation workloads with key architecture advancements including: a differentiated memory subsystem, next-generation geometry pipeline, new compute engine, and a new pixel engine".
Exciting stuff, that's for sure. I'm sure we're going to see something unveiled just before, or during Computex - with a physical launch in the weeks after, sometime in late June.