AMD Radeon Pro Vega II Duo displaces the NVIDIA Titan RTX to become the "world's most powerful GPU", but don't expect to game on it anytime soon
When the NVIDIA Titan RTX launched late last year, many dubbed it the "T-Rex" as it was essentially the most powerful graphics card on earth with 16.3 TFLOPs of FP32 power. With the launch of Apple's new Mac Pro yesterday, AMD announced two new GPUs, the Radeon Pro Vega II and the Radeon Pro Vega II Duo. With a massive 28.3 TFLOPs of FP32 performance, the Vega II Duo takes the crown from the Titan RTX to become the world's most powerful GPU.
It's been quite sometime since we've seen a dual-GPU card from AMD. The last consumer dual-GPU was the AMD Radeon R9 295X2 in 2014 while workstation users had the option of choosing a Radeon Pro Duo in 2016. The latest Radeon Pro Vega II is based on the same 7nm architecture (Vega 20, GCN 5.1) as the Radeon VII but enables the four CUs that were disabled in the latter. So, we now have a total of 64 CUs and 4096 SPs on the Radeon Pro Vega II along with 32 GB of HBM2 VRAM that offers 1 TB/s memory bandwidth. The whole GPU runs at a peak engine clock of 1.7 GHz. Combine two such GPUs on a single PCB and you get the Radeon Pro Vega II Duo.
The Radeon Pro Vega II Duo will thus offer 128 CUs, 8192 SPs, and a total of 64 GB of HBM2 VRAM for a peak FP32 performance of 28.4 TFLOPs at the same 1.7 GHz clock. Both the Vega 20 dies communicate via an Infinity Fabric link that allows for 84 GBps bandwidth per direction. Apple Mac Pro users also get the option of using a dedicated Mac Pro Expansion Module (MPX) that allows for running up to two Radeon Pro Vega II Duos at once for a whopping 56.8 TFLOPs of FP32 performance. Power supply to the card is taken care of by a proprietary PCIe connector that can draw up to 475W apart from the standard PCIe power supply of 75W. Both the Radeon Pro Vega II and the Vega II Duo offer four Thunderbolt 3 ports and one HDMI 2.0 port for hooking up displays.
The AMD Radeon Pro Vega II GPUs are aimed at making demanding workloads such as 8K video editing, 3D modeling, etc. a stroll in the park, and as such will cost an arm and a leg. The Radeon Pro Vega II and Duo are exclusive to the Apple Mac Pro as of now and, in all likelihood, will not be available for PCs.
While computation software can scale well between multiple GPUs, games aren't really optimized to take advantage of such setups, which is why SLI and CrossFire have very few takers. That being said, Infinity Fabric presents an interesting paradigm for developers and gamers in the future if AMD somehow manages to obfuscate the rendering pipeline across multiple GPUs from the game engine.
Speaking to PCGamesN close to a year ago, AMD's SVP Engineering in RTG David Wang said,
To some extent you’re talking about doing CrossFire on a single package. The challenge is that unless we make it invisible to the ISVs [independent software vendors] you’re going to see the same sort of reluctance.
We’re going down that path on the CPU side, and I think on the GPU we’re always looking at new ideas. But the GPU has unique constraints with this type of NUMA [non-uniform memory access] architecture, and how you combine features… The multithreaded CPU is a bit easier to scale the workload. The NUMA is part of the OS support so it’s much easier to handle this multi-die thing relative to the graphics type of workload.”
But Wang did express optimism that "Anything's possible", when asked if it is really possible to make the MCM design invisible to the developer. With cloud gaming gaining all set to become popular thanks to the likes of Google Stadia that uses a custom Vega 56 implementation, multi-GPU gaming is technically here, although not in a form we immediately recognize.
High-Performance AMD Radeon GPUs to Power All-New Mac Pro
New supercharged 7nm AMD Radeon™ Pro Vega II GPUs with high-speed HBM2 memory and AMD Infinity Fabric Link GPU interconnect technology power demanding content creation applications.
Santa Clara, Calif. 06/03/2019
AMD (NASDAQ: AMD) today announced that its new AMD Radeon™ Pro Vega II GPUs will be included in the all-new Mac Pro. Built on industry-leading 7nm process technology, AMD Radeon™ Pro Vega II GPUs provide groundbreaking levels of graphics performance for computation-intensive tasks, including rendering, 8K video, video effects, and other high-end content creation workloads.
AMD Radeon™ Pro Vega II GPUs provide exceptional computational performance by harnessing 7nm AMD Radeon Vega family GPUs, ultra-fast High Bandwidth Memory (HBM2), and AMD Infinity Fabric Link GPU interconnect technology that dramatically enhances data-transfer speeds between GPUs. Providing up to 14 TFLOPS of single-precision floating-point (FP32) performance and up to 28 TFLOPS of half-precision floating-point (FP16) performance, AMD Radeon™ Pro Vega II GPUs are optimized for powering demanding professional applications.
“Today’s high-end professional content creation applications are driving an insatiable need for ever increasing levels of processing power and memory,” said Scott Herkelman, corporate vice president and general manager, Radeon Technologies Group at AMD. “Equipped with Radeon™ Pro Vega II GPUs, the new Mac Pro delivers the computational horsepower and memory bandwidth to power ultra-high screen resolutions and help dramatically accelerate compute and content creation workloads, enabling creative professionals to focus on unleashing their creativity and delivering amazing results.”
“AMD continues to develop amazing graphics technologies to accelerate workflows for professionals who rely on DaVinci Resolve for faster color correction and editing,” said Dan May, president, Blackmagic Design. “Radeon Pro Vega II delivers outstanding performance, especially when DaVinci Resolve is put to work with 8K video footage combined with various filters.”
Key capabilities and features of AMD Radeon™ Pro Vega II GPUs include:
- Leading-edge compute performance – The AMD Radeon™ Pro Vega II GPU delivers up to 14 TFLOPS of single-precision FP32 performance and up to 28 TFLOPS of half-precision FP16 performance.
- Support for Infinity Fabric Link GPU interconnect technology – With up to 84GB/s per direction low-latency peer-to-peer memory access1, the scalable GPU interconnect technology enables GPU-to-GPU communications up to 5X faster than PCIe® Gen 3 interconnect speeds2.
- Ultra-fast HBM2 memory – 32GB of high-speed HBM2 memory delivers 1TB/s memory bandwidth, providing the memory capacity and data transfer speeds required by today’s high-resolution, multi-display setups, 8K video, and other demanding content creation workloads.
For more information about the Mac Pro visit www.apple.com/mac-pro.
- Learn more about AMD Radeon™ Pro Vega II GPUs here
- Become a fan of AMD on Facebook
- Follow AMD on Twitter
- Follow Radeon™ Pro graphics on Twitter
For 50 years AMD has driven innovation in high-performance computing, graphics and visualization technologies ― the building blocks for gaming, immersive platforms and the datacenter. Hundreds of millions of consumers, leading Fortune 500 businesses and cutting-edge scientific research facilities around the world rely on AMD technology daily to improve how they live, work and play. AMD employees around the world are focused on building great products that push the boundaries of what is possible. For more information about how AMD is enabling today and inspiring tomorrow, visit the AMD (NASDAQ: AMD) website, blog, Facebook and Twitter pages.
This press release contains forward-looking statements concerning Advanced Micro Devices, Inc. (AMD) including AMD’s expected benefits associated with utilizing AMD Radeon™ Pro Vega II GPUs in Apple’s next-generation Mac Pro, which are made pursuant to the Safe Harbor provisions of the Private Securities Litigation Reform Act of 1995. Forward-looking statements are commonly identified by words such as "would," "intends," "believes," "expects," "may," "will," "should," "seeks," "intends," "plans," "pro forma," "estimates," "anticipates," or the negative of these words and phrases, other variations of these words and phrases or comparable terminology. Investors are cautioned that the forward-looking statements in this document are based on current beliefs, assumptions and expectations, speak only as of the date of this document and involve risks and uncertainties that could cause actual results to differ materially from current expectations. Such statements are subject to certain known and unknown risks and uncertainties, many of which are difficult to predict and generally beyond AMD's control, that could cause actual results and other future events to differ materially from those expressed in, or implied or projected by, the forward-looking information and statements. Material factors that could cause actual results to differ materially from current expectations include, without limitation, the following: Intel Corporation’s dominance of the microprocessor market and its aggressive business practices may limit AMD’s ability to compete effectively; AMD has a wafer supply agreement with GLOBALFOUNDRIES Inc. (GF) with obligations to purchase all of its microprocessor and APU product requirements, and a certain portion of its GPU product requirements, manufactured at process nodes larger than 7 nanometer from GF with limited exceptions. If GF is not able to satisfy AMD’s manufacturing requirements, AMD’s business could be adversely impacted; AMD relies on third parties to manufacture its products, and if they are unable to do so on a timely basis in sufficient quantities and using competitive technologies, AMD’s business could be materially adversely affected; failure to achieve expected manufacturing yields for AMD’s products could negatively impact its financial results; the success of AMD’s business is dependent upon its ability to introduce products on a timely basis with features and performance levels that provide value to its customers while supporting and coinciding with significant industry transitions; if AMD cannot generate sufficient revenue and operating cash flow or obtain external financing, it may face a cash shortfall and be unable to make all of its planned investments in research and development or other strategic investments; the loss of a significant customer may have a material adverse effect on AMD; AMD’s receipt of revenue from its semi-custom SoC products is dependent upon its technology being designed into third-party products and the success of those products; global economic and market uncertainty may adversely impact AMD’s business and operating results; AMD’s products may be subject to security vulnerabilities that could have a material adverse effect on AMD; IT outages, data loss, data breaches and cyber-attacks could compromise AMD’s intellectual property or other sensitive information, be costly to remediate and cause significant damage to its business, reputation and operations; AMD’s operating results are subject to quarterly and seasonal sales patterns; AMD may not be able to generate sufficient cash to service its debt obligations or meet its working capital requirements; AMD has a large amount of indebtedness which could adversely affect its financial position and prevent it from implementing its strategy or fulfilling its contractual obligations; the agreements governing AMD’s notes and the Secured Revolving Line of Credit impose restrictions on AMD that may adversely affect AMD’s ability to operate its business; the markets in which AMD’s products are sold are highly competitive; AMD’s worldwide operations are subject to political, legal and economic risks and natural disasters, which could have a material adverse effect on it; the conversion of the 2.125% Convertible Senior Notes due 2026 may dilute the ownership interest of AMD’s existing stockholders, or may otherwise depress the price of its common stock; uncertainties involving the ordering and shipment of AMD’s products could materially adversely affect it; the demand for AMD’s products depends in part on the market conditions in the industries into which they are sold. Fluctuations in demand for AMD’s products or a market decline in any of these industries could have a material adverse effect on its results of operations; AMD’s ability to design and introduce new products in a timely manner is dependent upon third-party intellectual property; AMD depends on third-party companies for the design, manufacture and supply of motherboards, software and other computer platform components to support its business; if AMD loses Microsoft Corporation’s support for its products or other software vendors do not design and develop software to run on AMD’s products, its ability to sell its products could be materially adversely affected; and AMD’s reliance on third-party distributors and add-in-board partners subjects it to certain risks. Investors are urged to review in detail the risks and uncertainties in AMD's Securities and Exchange Commission filings, including but not limited to AMD's Quarterly Report on Form 10-Q for the quarter ended March 30, 2019.
©2019 Advanced Micro Devices, Inc. All rights reserved. AMD, the AMD Arrow logo, Radeon, and combinations thereof are trademarks of Advanced Micro Devices, Inc. “Vega” is an internal AMD codename and is not a product name. Apple, Mac, and the Apple Logo are trademarks of Apple Inc., registered in the U.S. and other countries. Other product names used in this publication are for identification purposes only and may be trademarks of their respective companies. Other product names used in this publication are for identification purposes only and may be trademarks of their respective companies.
- Infinity Fabric Link Bandwidth with 2 bi-directional x16 links at up to 21Gbps = 2 x 16 x 21Gbps / 8 = up to 84GB/s bandwidth per direction
- Gen 3.0 PCI Express x16 slot capable of 16 GB/s of bandwidth per direction and Radeon Pro Vega II with Infinity Fabric Link capable of 84GB/s bandwidth per direction. 84 / 16 = 5.25X faster than Gen 3.0 PCI Express x16.