AMD Teases Dual Fiji GPUs – Fury X2 On The Horizon?
AMD's fastest Radeon yet, the fabled "Fury X2" dual Fiji graphics card has potentially just been teased on twitter by the company's resident frame pacing guru. We're talking about Scott Wasson, former editor-in-chief at TechReport.com, who's been snagged by AMD late last year. Along with our friend Ryan Shrout from PCPer.com he has been a pioneer of the frame pacing revolution in dual GPU graphics card evaluation that's all about stutter free smooth gaming.
Frame-pacing gave us the ability to explain and analyze one of the most elusive characteristics of dual GPU boards. A characterstic that everyone had already agreed exists but did not fully understand and that's what used to be widely known as micro-stuttering. This previously unexplained behavior that plagued dual GPU boards with stutters and hitches for years but was poorly understood.
Today Scott is with AMD and he's charged with making sure that all of AMD's future dual GPU offering as well as CrossfireX dual graphics card setups provide the smoothest gameplay experiences there is. AMD has been preparing its fastest graphics card yet, a flagship board code named Gemini with two full fledged Fiji XT GPUs for some time. Back in E3 of last year AMD had announced that a dual Fiji board will arrive to market sometime in the fall, but the fall came and went and there was no dual Fiji board to be seen.
Late last year just as we entered winter the company announced that it had delayed the introduction of its dual Fiji board to align it with major VR product launches this year in late Q1 and early Q2. We're just about to enter late Q1 and as it happens AMD has begun teasing its latest and greatest creation. So the wait is very assuredly about to be over.
AMD Shows Off Dual Fiji Croes On Twitter - "Fury X2" Is On The Way?
The new ultra-enthusiast flagship Radeon graphics card is set to replace the current Radeon king of the hill & the holder off the world's fastest graphics card title, the R9 295X2. The official product name for the dual Fiji XT board code named "Gemini" is still unknown, so we'll be using the "Fury X2" designation to refer to the card in the time being.
Scott Wasson tweeted the photo you see above today in silence. It shows two Fiji XT GPUs, which are the heart and soul of AMD's R9 Fury X, R9 Fury and R9 Nano graphics cards. This is eerily similar to the company's teaser campaign for the R9 295X2 which the company launched a couple of years ago. The campaign included all sorts of humorous teasers. From sending reviewers' mirrored photos of themselves with the hashtag #2BETTERTHAN1 to sending them boxes of volcanic water and two cans of chips.
Some figures in the upper echelon of the game development world such as Johann Andersson, chief architect behind the Frostbite engine, have had a pre-release sample of AMD's upcoming dual Fiji card to play with in as early as November of last year, three months ago. So Scott Wasson's tease was certainly not the first one.
Dat feeling when your just arrived closed liquid cooling pre-release GPU turns out to not be so closed after all pic.twitter.com/umpVUtdVwc
— Johan Andersson (@repi) November 26, 2015
The specifications for the card have been finalized sometime ago. We shared them with you in an exclusive article back in October. The card will feature two full fledged Fiji XT GPUs with a total of 8192 GCN cores, 8GB of stacked High Bandwidth Memory and a total memory bandwidth of one Terabyte per second. That's a whole lot of performance. Which if we're talking about R9 Nano clock speeds, and we are, is equivalent to 16 Teraflops/s of single precision compute. That's equivalent to four and a half R9 380 graphics cards, in a board that's shorter than a reference design GTX 960.
[UPDATED] 08:41 AM. February 14, 2016 Eastern Time (ET)
XDMA, The Technology Behind Crossfire's Impressive Scaling
In 2013 AMD introduced a brand new Crossfire hardware implementation by adding a DMA engine to the Crossfire composting block on the GPU die. This was first introduced with the Hawaii GPUs in September of 2013, and each subsequent GPU that AMD introduced also had this technology which AMD dubbed XDMA, short for Crossfire DMA.
NEW AMD CROSSFIRE™ TECHNOLOGY -‐ XDMA
- XDMA is a modern approach to negoIaIng the communicaIon of mulIple GPUs for peak gaming performance with the ultra-‐high-‐resoluIon displays of tomorrow.
- Radeon™R9-‐290X includes a Hardware DMA engine in the AMD CrossFire™ composiIng block
- Designed for AMD Eyefinity and UltraHD resoluIons via DisplayPort™ ‒
- Allows for direct access between GPU display pipelines over PCI Express®
- No external connector required
- CompaIble on AMD Catalyst™ frame pacing technologies
- No performance penalty versus external bridge
The technology enables faster, lower overhead communication between multiple GPUs, this in turn translates to better Crossfire scaling. And that's what we're seeing take effect here. It also enables graphics cards in Crossfire setups to inter-communicate through the PCI Express interface which eliminates the need for a Crossfire bridge. But most importantly is that it enables more effective and efficient multi-GPU scaling. Allowing the performance of Crossfire, TriFire and QuadFire GPU setups to scale significantly better.
We've taken an in-depth look at Crossfire and SLI scaling in a previous article which you should check out if you want to learn more about how this technology can affect your gaming experience as well as reduce frametime variance, which is what causes the perceived micro suttering that gamers sometimes notice from multi-GPU setups.
In closing, you can find all the specs for the R9 Fury X2 that we've managed to confirm thus far as well as the R9 Fury X, R9 Fury, R9 Nano and R9 290X below arranged conveniently in a table for you to feast your eyes upon.
AMD Radeon Duo Pro Specifications
|WCCFTech||AMD Radeon Pro Duo||AMD Radeon R9 Fury X||AMD Radeon R9 Nano||AMD Radeon R9 Fury||AMD Radeon R9 290X|
|GPU||Fiji XT x 2||Fiji XT||Fiji XT||Fiji Pro||Hawaii XT|
|GCN Compute Units||128||64||64||56||44|
|Render Output Units||128||64||64||64||64|
|Texture Mapping Units||512||256||256||224||176|
|GPU Frequency||Up to 1000 Mhz||Up to 1050Mhz||Up to 1000 MHz||Up to 1000 MHz||1000Mhz|
|Memory||8GB HBM (4 GB Per GPU)||4GB HBM||4GB HBM||4GB HBM||4GB GDDR5|
|Memory Interface||8192bit (4096 Per GPU)||4096bit||4096bit||4096bit||512bit|
|Memory Frequency||500Mhz||500Mhz||500 MHz||500Mhz||1250Mhz|
|Effective Memory Speed||1Gbps||1Gbps||1Gbps||1Gbps||5Gbps|
|Memory Bandwidth||1 TB/s||512GB/s||512GB/s||512GB/s||320GB/s|
|Cooling||Liquid Cooling||Liquid Cooling||Air Cooling||Air Cooling||Air Cooling|
|Performance (FP32)||16.38 TFLOPs||8.6 TFLOPS||8.19 TFLOPS||7.2 TFLOPS||5.6 TFLOPS|
|Launch Date||26th April||24th June 2015||7th September 2015||10th July 2015||24th October 2013|
The ~ symbol next to a spec denotes that it's not yet confirmed.