AMD's long awaited competitor to Nvidia's GTX 1080 may be right around as the company teases the launch venue for its upcoming Vega GPU. The teaser came straight from the Senior Director, Global Marketing and Public Relations, Radeon Technologies Group at AMD Chris Hook. Who posted a picture of the launch venue for Vega on his facebook page.
Vega is the code name of AMD's next generation high performance graphics architecture featuring 14nm FinFET technology and second generation stacked High Bandwidth Memory. In many ways Vega is the successor to AMD's last flagship graphics chip Fiji. The world's very first GPU to feature high bandwidth memory and the heart and soul that powers AMD's Fury X, Fury & Nano graphics cards.
AMD Teases Vega GPU Launch, GTX 1080 Radeon Competitor May Be Coming Sooner Than Expected
The VEGA teaser is a teaser in every sense of the word. Chris Hook did not reveal too much but just enough to hook us in, pun definitely intended. The photo he posted on his facebook page shows what looks like a large industrial space that's in desperate need of "renovations" as Chris put it.
Unless Hook is attempting some kind of elaborate prank here, this will be where the venue will eventually be held once the place is set up and ready to go. So it's quite clear that while VEGA may very well be coming sooner than we had expected it's still several months away at the very least.
The VEGA GPU architecture was first announced by AMD's head of the Radeon Technologies Group Raja Koduri at the company's big "Capsaicin" media event held earlier this year. Out of Capsaicin came the very first long term GPU roadmap that we've seen from AMD since the introduction of the very first GCN graphics products in 2011. the company's Graphics Core Next architecture, or GCN for short, is now leveraged across the company's entire graphics portfolio. From discrete GPUs, to APUs and console semi-custom SOCs.
The just launched Polaris based RX 400 series products are based on the company's 4th generation GCN architecture. According to several filed patents and some very interesting leaks via AMD's engineering staff VEGA is based on a brand new iteration of GCN. At Capsaicin no specific launch dates were given for VEGA, not even Polaris for that matter. Despite the fact that it was it had been scheduled to launch just a few months after the event.
VEGA In October...?
With that being said, the company's official roadmap suggested that VEGA would launch in 2017. We've heard some whispers since VEGA's official announcement back in March which suggested that AMD was aggressively attempting to push the launch ahead to October of 2016, rather than early 2017.
VEGA's Development Reached A Major Milestone in June
If you've been keeping taps on WCCF and hardware news you'll remember that AMD's Chief Architect of the Radeon Technologies Group Raja Koduri had revealed back in late June that VEGA had passed a major development milestone. In fact Koduri tweeted several photos of the celebration right after he delivered the news.
Design team is thrilled with your response. Celebrated a milestone with the team. Long way to go before you see it pic.twitter.com/duQVBBGict
— Raja Koduri (@GFXChipTweeter) June 23, 2016
Koduri did not offer much detail about what the milestone specifically was. It could have been the very first working silicon out of the fab, or final tapeout of the product. Whatever it was, it was clearly a huge deal.
Incredibly proud of our gpu design team here in shanghai that delivered polaris family and next Vega pic.twitter.com/ejvFJW1DoY
— Raja Koduri (@GFXChipTweeter) June 23, 2016
We broke the news on the Vega 10 code name in an exclusive back in January. This is the infamous "Greenland" reincarnate that AMD has been working on for so many years. The VEGA architecture hadn't been announced until March, and the "Vega 10" chip hadn't been mentioned in a public setting until Raja broke the news about the major milestone a couple of months ago. Now we have Chris Hook's sneaky teaser. It definitely seems like the pace is picking up around VEGA. We'd be genuinely surprised if AMD did not at least announce one or both of its upcoming VEGA GPUs before the year's end even if no products are launched.
Vega 10 Specs
| WCCFTech | Polaris | Vega | Pascal | Pascal | Pascal |
|---|---|---|---|---|---|
| Year | 2016 | 2017 | 2016 | 2016 | 2017 |
| GPU | Polaris 10 | Vega 10 | GP102 | GP104 | GP102 |
| Graphics Card | RX 480 | TBA | GTX TItan X Pascal | GTX 1080 | GTX 1080 Ti |
| Process | 14nm | 14nm | 16nm | 16nm | 16nm |
| Transistors In Billions | 5.7 | TBA | 12 | 7.2 | 12 |
| Stream Processors | 2304 | 4096 | 3584 | 2560 | 3328-3840 |
| Performance | 5.8 TFLOPS | 12 TFLOPS | 11 TFLOPS | 9 TFLOPS | 10 TFLOPS |
| Memory | 8GB GDDR5 | 16GB HBM2 | 12GB GDDR5X | 8GB GDDR5X | 12GB GDDR5X |
| Memory Bus | 256bit | 4096bit | 382bit | 256bit | 382bit |
| Bandwidth | 256 GB/s | 512 GB/s | 480 GB/s | 320 GB/s | 480 GB/s |
Leaked & rumored specs.
VEGA Built On 14nm FinFET - Features Second Generation - HBM2 - Stacked High Bandwidth Memory
The Fiji GPU powering the R9 Fury series is AMD's largest ever graphics processing unit. It's also the world's first to feature the 3D structured, 2.5D stacked High Bandwidth Memory technology. This standard was co-invented by AMD and SK Hynix, one of the world's largest memory makers. The advantages of vertically stacking dies are many. One of which is the enablement of significantly greater densities compared to GDDR5. HBM cubes also occupy a lot less space than GDDR5, allowing engineers to achieve tremendous area savings on the printed circuit board of the graphics card. This is what enabled AMD to make the popular R9 Nano, a tiny six inch graphics cards with 8 TFLOPS of compute. A figure that greatly overshadows similarly sized cards.
There other advantages to HBM, namely how it's packaged. Because the memory chips sit next to the GPU on an interposer, a very short distance away from the GPU die, it allows engineers to create immensely wider memory busses with reduced latency and very low power requirements. The end result translates to a memory standard that's fast, scalable, very power efficient and very area efficient.
VEGA Is All About Bringing HBM To The Masses
The Radeon Technologies Group's plan is to introduce HBM2 technology with the upcoming VEGA graphics architecture across more market segments compared to where the technology currently exists. Which is confined only to the highest end segment of the market. If everything goes according to plan VEGA will be the first architecture to bring HBM to all segments of the graphics market, up and down the stack.
HBM2 is the second generation of the vertically stacked High Bandwidth Memory standard that AMD introduced last year with the Radeon R9 Fury X graphics card, the R9 Fury and the R9 Nano. Second generation High Bandwidth Memory is not only faster than the first generation but it also scales to capacities 8 times larger than HBM1. HBM2 feature 1GB per die and up to 8 Hi stacks for 8GB per cube. Which is eight times more than the highest capacity currently available per cube on HBM1. Additionally, HBM2 operates at twice the speed of HBM1 for double the memory bandwidth.
This means that VEGA 10 can be configured with up to 32GB of HBM2 memory for a total of one terabyte/s of memory bandwidth. Paired with over four thousand next generation stream processors VEGA 10 will be the most potent graphics chip from AMD yet. That is until the next big thing comes out. As is the nature of the beast.
AMD Radeon Graphics Architectures
| WCCFTech | Year | Consumer Product | GPU | Process | Transistors In Billions | Performance | Memory | Bandwidth |
|---|---|---|---|---|---|---|---|---|
| Southern Islands | 2012 | HD 7970 Ghz | Tahiti | 28nm | 4.3 | 4.1 TFLOPS | 3GB GDDR5 | 264GB/s |
| Volcanic Islands | 2013 | R9 290X | Hawaii | 28nm | 6.2 | 5.6 TFLOPS | 4GB GDDR5 | 320GB/s |
| Caribbean Islands | 2015 | R9 Fury X | Fiji | 28nm | 8.9 | 8.6 TFLOPS | 4GB HBM1 | 512GB/s |
| POLARIS | 2016 | RX 480 | Polaris 10 | 14nm | 5.7 | 5.2 TFLOPS | 8GB GDDR5 | 256 GB/s |
| VEGA | 2017 | RX Vega 64 | Vega 10 | 14nm | 12.5 | 13 TFLOPS | 8GB HBM2 | 480GB/s |
| VEGA Refresh VEGA 2 | 2018 2018 | TBA TBA | TBA Vega 20 | 12nm 7nm | 12.5 TBA | TBA TBA | TBA 32GB HBM2 | TBA 1 TB/s |
| NAVI | 2019 | TBA | Navi 10 | 7nm | TBA | TBA | Nextgen Memory | TBA |
