AMD formally launches the Radeon R9 290X Hawaii GPU

AMD officially launches today the fastest, most powerful graphics card they have ever built – the Radeon R9 290X. The AMD Radeon R9 290X was originally introduced at the ’14 GPU, but it’s only today that it’s available to consumers who want great performance at 4K resolutions and beyond, at a hugely affordable price of just $ 549.

AMD Radeon R9 290X – Straight from the volcanoes of Hawaii!

With the launch finally taking place, it’s time to briefly talk about the specifications of the latest Hawaii graphics card. There’s a lot to talk about, such as the performance metrics and architectural details we’ll be describing in this article, but we’d have to rely on the following technical websites for the ratings as we haven’t received our card from AMD yet. Check out the reviews of the AMD Radeon R9 290X at the links below:

AMD Hawaii Architecture – AMD tackles the GK110 with its own monster

The Hawaii GPU architecture is an advance on the GCN 1.0 architecture introduced in the Radeon HD 7970. The successor, also known as the Radeon R9 290X, has the latest GCN 2.0 architecture, which is massive compared to the Tahiti core, but 30% smaller than the GK110 core from NVIDIA which measures 551mm2. The AMD GPU measures 438 mm2 and has been optimized and organized by AMD in such a way that more performance is achieved and the efficiency rate is twice as high as that of the core of the last generation. The overall architectural design is somewhat similar to Tahiti, but with tons of improvements enough to produce radical improvements in performance.

While the GK110 has 7.1 billion transistors, AMD’s Hawaii consists of 6.2 billion transistors in a 28 nm chip process. It’s surprising that even with a billion fewer transistors, the Hawaii comes so close to the GeForce GTX Titan in terms of performance, which has a GK110 core with 1 SMX unit disabled.

Above is the Hawaii GPU block diagram, which shows that the GCN core is made up of eight ACE or Asynchronous Compute Engines modules. This is twice as much as in Tahiti. In the middle of these modules is the main graphics command processor, which allows global data sharing. The Hawaii GPU consists of four shader engines, each with its own parallel linked geometry processors and rasterizers. Each shader engine contains 11 processing units with 64 stream processors, the complete Radeon R9 290X Hawaii GPU with 44 processing units and a shader number of 2816 streaming processors, 176 texture mapping units and 64 render back-ends.

The Hawaii GCN 2.0 core consists of up to four shader engines per GPU, so the Hawaii XT chip fused in the R9 290X uses the flagship core that AMD has to offer. There is one geometry process per shader engine as described above that can balance the load with other shader engines and has one primitive per cycle.

All shader engines share the same pool of 16 64K L2 cache with an L2 / L1 bandwidth of up to 1 TB / s. The Hawaii GPU has eight 64-bit MCs or storage controllers that lead to a 512 -Bit bus interface with an abundant 4 GB GDDR5 memory. In addition to the most important GEs, multimedia processors are offered, including VCE, UVD, DSPs for TrueAudio, six of the new XDMA, Eyefinity controllers and the DMA engine. Finally, there is a small PCI-e 3.0 bus interface that enables support with the latest PCI-e 3.0-capable motherboards.

AMD has taken a step forward with its GCN 2.0 architecture and given it a faster 512-bit interface, which is a comeback for high-performance graphics cards. 512-bit has been missing from flagship graphics cards for a few years, and AMD has made a comeback with a 4GB 512-bit interface on the Hawaii GPU running at 5.0Gbps. This achieves a memory bandwidth of 320 GB / s, which is a phenomenal increase compared to Tahiti with 288 GB / s.

A slide on the efficiency of the GCN 2.0 architecture shows the key improvements between Tahiti and Hawaii. We can see that the processing rate of the geometry with 4 billion primitives per second on Hawaii has increased 1.9 times compared to 2.1 billion on the predecessor. Similarly, texture and pixel fill rates have been increased by 30 and 90 percent, respectively, while bandwidth is 320 GB / s compared to 264 GB / s for the HD 7970. Ultimately, users on the calculator would get 12.8 peak GLops / mm2 compared to 12.2 in Tahiti. The Hawaii Radeon R9 290 series has a larger chip size of 438 mm2, so we expect up to 5.6 TFlops of computing power.

AMD is also introducing CrossFireX bridgeless technology for the first time, which does not require an external connection for CrossFire to work. AMD mentions that the new CrossFire technology for the AMD Radeon R9 290X is designed for AMD Eyefinity and UltraHD (4K) resolution setups. As already mentioned, the new technology enables direct access between GPU display pipelines via PCI-Express, so that no external connections are required. The technology is fully compatible with AMD’s Catalyst frame pacing technologies, which can be used to reduce frame pacing issues that occur with multi-GPU setups that are a major problem with the Radeon 7000 series. Finally, the slide mentions that users with Sideport would not experience any performance degradation compared to traditionally used external bridges for CrossFireX.

The dual DMA engines and several new Eyefinity controllers could saturate the PCI-e 3.0 x16 bus bandwidth with bi-directional access of 16 GB / s, which would enable a seamless experience of CrossFire gaming performance.

AMD Radeon R9 290X – 4K Games Now Affordable

The Radeon R9 290X is the flagship with the full-blown Hawaii chip with 2816 stream processors, which are divided into 44 CUs (Compute Units). It has a clock rate of up to 1 GHz. Essentially, the graphics card has two clock rates at which it works. These are configured via the card’s dual BIOS switch, which is located directly next to where the CrossFire bridge is located on Radeon cards. Most reference cards do not ship with dual BIOS support. This is likely the first flagship GPU to ship with such an option. The high stream processor and the number of cores enable the Radeon R9 290X to deliver a processing power of 5.6 TFlops. It has 64 raster operators, twice as many as the Radeon HD 7970.

At the standard frequency, the card works at 800 MHz and the boost clock pushes it to 1000 MHz. More OC can be achieved with the new PowerTune enhancements, easily accessible via the latest AMD Overdrive software available in the new Catalyst Control Center.

In addition to the core specifications, the Radeon R9 290X has a generous 4 GB of GDDR5 memory that runs over a 512-bit interface. The clock rate for the memory is effectively kept at 5.0 GHz, which offers a bandwidth of over 320 GB / s. The Radeon R9 290 series is primarily aimed at the ultra high resolution audience who want to play the upcoming titles at 4K resolutions and beyond, as AMD showed demos of Eyefinity 4K monitors running over a single Radeon R9 290X GPU . The Battlefield 4 Angry Sea gameplay was also demonstrated on a single Radeon R9 290X in Eyefinity. The Radeon R9 290X has dual DVI, HDMI and display ports for viewing and a single 6-pin and 8-pin power connector for refreshing.

More information is available at http://www.amd.com/US/PRODUCTS/DESKTOP/GRAPHICS/R9/Pages/amd-radeon-hd-r9-series.aspx

AMD Radeon R9 290 Series Specifications:

AMD Radeon R9 290X AMD Radeon R9 290 GeForce GTX 780 GeForce GTX Titan
GPU code name Hawaii Hawaii GK110 GK110
GPU process 28nm 28nm 28nm 28nm
Stream processors 2816 2560 2304 2668
Base clock 800 MHz 900 MHz 863 MHz 837 MHz
Turbo clock 1000 MHz 947 MHz 902 MHz 876 MHz
VRAM 4GB 4GB 3 GB 6 GB
Memory bus 512-bit 512-bit 384-bit 384 bits
Memory clock 5 GHz (effective) 5 GHz
(Effective)
6 GHz (effective) 6 GHz (effective)
Energy configuration 8 + 6 pin 8 + 6 pin 8 + 6 pin 8 + 6 pin
PCB VRM 5 + 1 + 1 5 + 1 + 1 6 + 2 6 + 2
The size 438 mm2 438 mm2 551 mm2 551 mm2
Publication date October 24, 2013 October 31, 2013 June 23, 2013 February 21, 2013
Starting price $ 549 $ 449 $ 649 $ 999

AMD Radeon R9 290X Performance Charts / Slides:

AMD Radeon R9 290X Additional presentation slides:

Comments are closed.