[W1zzard]

Introduction

Out with the old, in with the new. Let's welcome the newest kid on the block, the Radeon HD 7970, part of AMD's spanking new Southern Islands GPU family. This card is the industry's first with a few things, it uses the first ever high-performance GPU built on the 28 nanometer silicon fabrication process; Radeon HD 7970 is the industry's first card compliant with Microsoft's DirectX 11.1 API, which will ship with the next major version of Windows; and is the first card to use the PCI Express 3.0 x16 bus, that doubles system interface bandwidth to 32 Gb/s and is touted by motherboard manufacturers as the next big thing since PCI.

New generations of GPUs naturally bring with them performance increments, some times even 100% that of preceding generations, they also serve as launch-vehicles for new features that quickly go on to become industry standards, and help the technology grow. The Radeon HD 7970 has both of these responsibilities resting on its shoulders: to score performance wins, and pack some killer new features that matter to the end-user.

Product Positioning

The AMD Radeon HD 7970 is a unique card from a market-positioning standpoint. After Radeon HD 2900 series, and the completion of ATI's merger with AMD, the company took up a unique model of product development that ensured it could have competitive products out in the market targeting every segment, while not having to spend much on making large GPUs. Its goal with a new GPU architecture always involved making a killer high-performance (not high-end) GPU, and using it both ways: in dual-GPU cards as high-end products, and by disabling some components/features to carve out cheaper/cost-effective products.

The AMD Radeon HD 7970, particularly the GPU behind it, codenamed "Tahiti", is stretching that model a little towards the higher-end. Tahiti is bigger than what AMD's typical "high-performance" GPU is supposed to be. For one, it features a 384-bit wide memory interface. AMD was the first to market with GDDR5 memory standard, which it initially sought as a way to circumvent the need for a GDDR3/4 memory bus wider than 256-bits. With NVIDIA catching up with the memory standard, and implementing a 384-bit GDDR5 memory interface on its GeForce Fermi 100/110 GPUs, AMD felt the pinch for doing something to increase the memory bandwidth of the HD 7970, to keep up with the increasing compute performance of their GPUs. The company chose slightly faster GDDR5 memory chips with HD 6900 series, but it could only yield small bandwidth gains. The only option left without having to switch memory architecture to the lesser known XDR2, was to increase the memory bus width physically by 50%, hence 384-bit. Coupled with the faster 5.5 GT/s memory chips it used on the HD 6900, and appropriate clock speeds, it yields around 260 GB/s of memory bandwidth.



The unconventionally-wider memory bus of the Radeon HD 7970, combined with a brand new math-processing machinery contribute to the HD 7970's product placement, which is between the previous-generation single-GPU Radeon HD 6970, and previous-generation dual-GPU HD 6990, tilting closer towards the HD 6990.

AMD Radeon HD 7970 Market Segment Analysis

	GeForce GTX 560	GeForce GTX 560 Ti	GeForce GTX 560 Ti 448 C	Radeon HD 6950	GeForce GTX 570	Radeon HD 6970	GeForce GTX 580	Radeon HD 7970	Radeon HD 6990	GeForce GTX 590
Shader Units	336	384	448	1408	480	1536	512	2048	2x 1536	2x 512
ROPs	32	32	40	32	40	32	48	32	2x 32	2x 48
Graphics Processor	GF114	GF114	GF110	Cayman	GF110	Cayman	GF110	Tahiti	2x Cayman	2x GF110
Transistors	1950M	1950M	3000M	2640M	3000M	2640M	3000M	4310M	2x 2640M	2x 3000M
Memory Size	1024 MB	1024 MB	1280 MB	2048 MB	1280 MB	2048 MB	1536 MB	3072 MB	2x 2048 MB	2x 1536 MB
Memory Bus Width	256 bit	256 bit	320 bit	256 bit	320 bit	256 bit	384 bit	384 bit	2x 256 bit	2x 384 bit
Core Clock	810 MHz	823 MHz	732 MHz	800 MHz	732 MHz	880 MHz	772 MHz	925 MHz	830 MHz	607 MHz
Memory Clock	1002 MHz	1002 MHz	950 MHz	1250 MHz	950 MHz	1375 MHz	1002 MHz	1375 MHz	1250 MHz	855 MHz
Price	$180	$210	$290	$250	$330	$340	$500	$549	$700	$750

Architecture

Tahiti, named after the lovely Tahiti islands of French Polynesia in the South Pacific, is AMD's new high-performance GPU. It succeeds Cayman, the chip powering the Radeon HD 6900 series and the dual-GPU HD 6990. Tahiti is built on TSMC's brand new 28 nanometer silicon fabrication process. In a nutshell, it retains the tried and trusted component hierarchy of its predecessors, but with major changes in the math processing SIMD machinery, and raster operations processors (ROPs) de-linked from the memory bus width.


With Tahiti, AMD is introducing the biggest revamp in the way its GPUs crunch numbers, since the Radeon HD 2000. Back then AMD adopted the "Graphics Parallel Core" compute architecture which uses clumps of super-scalar processors that work out "very long instruction words" (VLIW). The older version of Graphics Parallel Core used sets of four simple stream processors with one complex stream processor (that has more capabilities), along with branch units and general purpose registers. This made up with VLIW5 design. With Radeon HD 6900 "Cayman" GPU, AMD implemented a newer design that used groups of four equipotent stream processors, making up the more advanced VLIW4 design.


As mentioned earlier, for the most part, the GPU's component hierarchy is the same, except that its number-crunching parts have undergone, in AMD's words, a revolutionary change. VLIW5 to VLIW4 was evolutionary, in comparison. With Tahiti, AMD replaced its VLIW stream processor clusters with GCN (Graphics Core Next) compute units. Each GCN compute unit is a super-scalar processor that combines scalar and vector elements that follow a new non-VLIW instruction-set architecture, and utilizes an improved layout of shared and dedicated components. To the end-user, this architecture translates into higher performance per millimeter square of GPU die area. Smaller the GPU die, cheaper it is to produce, and sometimes, lower its power draw.


Graphics CoreNext also brings with it what AMD refers to as its 9th Generation Tessellation unit. Tahiti packs two geometry engines, and with it, two independent tessellation units. These units take advantage of larger parameter caches, new off-chip buffering capabilities, and new vertex reuse instructions to deliver a whopping four times tessellation performance improvement over the previous generation, at least on paper.


In a nutshell, the Tahiti die measures 365 mm², holding 4.31 billion transistors. It is built on the 28 nm TSMC process. It has 32 GCN compute units and 2048 stream processors in all. There are 128 texture memory units (TMUs), and 32 raster operation processors (ROPs). Despite having a 384-bit wide memory interface, the raster operations processor (ROP) count didn't increase proportionately to 48, as expected. This is because AMD reconfigured the way ROPs interact with the rest of the GPU, and hence retained its conventional ROP count of 32. The 384-bit wide memory interface combined with 5.5 GT/s GDDR5 memory chips sums up to a 264 GB/s memory bandwidth. Since it uses twelve 2 Gbit memory chips, it ends up with 3 GB of total memory on board.

Packaging & Contents

We received a card only from AMD, without packaging but rest assured, the retail units will come with standard accessories like adapters and power cables.

With the Radeon HD 7970 AMD requires all partners to bundle an HDMI to (single link) DVI adapter, so that you can use the card with two DVI monitors.

The Card

AMD's new HD 7970 follows the company's traditional theme with black and red colors. The length of the card is 27.5 cm, which is the same as the HD 6970, for example.


The card requires two slots in your system.


Display connectivity options include one DVI port, one full size HDMI port and two mini-DisplayPorts. You may use all the outputs at the same time, thanks to AMD's superior display output architecture.

Please note that all the connectors are on the "first" slot now, which opens up the second one for a cooler grill through which hot air is blown out of the case. This also makes it possible to use the HD 7970 in a single slot configuration with a waterblock to cool the components.

An HDMI sound device is included in the GPU, too. It is HDMI 1.4a compatible which includes HD audio and support for Blu-ray 3D movies. The DisplayPort outputs are version 1.2 which enables the use of hubs and Multi-Stream transport.


You may combine up to four HD 7970 cards from any vendor in a multi-GPU CrossFire configuration for higher framerates or better image quality settings.


Pictured above are photos of the front and back, showing the disassembled board. High-res versions are also available (front, back). If you choose to use these images for voltmods etc, please include a link back to this site or let us post your article.

A Closer Look

AMD's fansink uses a large area vapor chamber cooler to keep the GPU cool. You can also see the thermal pads for twelve memory chips and the VRM circuitry.


The card requires one 8-pin and one 6-pin PCI-Express power cable for operation. This power configuration is good for up to 300 W of power draw.


For voltage control the card uses the CHiL CHL8228 chip, the same model found on other graphics card designs from AMD. This means that software support for overvolting will be easy to implement.


The GDDR5 memory chips are made by Hynix, and carry the model number H5GQ2H24MFR-R0C. They are specified to run at 1500 MHz (6000 MHz GDDR5 effective).


AMD's new Tahiti graphics processor introduces a new shader architecture, it is also the first GPU to be produced on a 28 nm process at TSMC. The transistor count is 4.31 billion.

Please note the shiny metal shim around the GPU core, which is part of the GPU and prevents the crushing or chipping of the die during assembly. The GPU's model number 215-0821060 is printed on the bottom right of the shim instead of on the die itself.

Test System

Test System - VGA Rev. 16
CPU:	Intel Core i7 920 @ 3.8 GHz (Bloomfield, 8192 KB Cache)
Motherboard:	Gigabyte X58 Extreme Intel X58 & ICH10R
Memory:	3x 2048 MB Mushkin Redline XP3-12800 DDR3 @ 1520 MHz 8-7-7-16
Harddisk:	WD Caviar Blue WD5000AAKS 500 GB
Power Supply:	Antec HCP-1200 1200W
Software:	Windows 7 64-bit Service Pack 1
Drivers:	NVIDIA: 285.62 ATI: Catalyst 11.12 HD 7970: Dec 16 Reviewer Driver
Display:	LG Flatron W3000H 30" 2560x1600

Benchmark scores in other reviews are only comparable when this exact same configuration is used.

• All video card results were obtained on this exact system with the exact same configuration.
• All games were set to their highest quality setting unless indicated otherwise.
• AA and AF are applied via in-game settings, not via the driver's control panel.

Each benchmark was tested at the following settings and resolution:

• 1024 x 768, No Anti-aliasing. This is a standard resolution without demanding display settings.
• 1280 x 1024, 2x Anti-aliasing. Common resolution for most smaller flatscreens today (17" - 19"). A bit of eye candy turned on in the drivers.
• 1680 x 1050, 4x Anti-aliasing. Most common widescreen resolution on larger displays (19" - 22"). Very good looking driver graphics settings.
• 1920 x 1200, 4x Anti-aliasing. Typical widescreen resolution for large displays (22" - 26"). Very good looking driver graphics settings.
• 2560 x 1600, 4x Anti-aliasing. Highest possible resolution for commonly available displays (30"). Very good looking driver graphics settings.

Aliens vs. Predator

Aliens vs. Predator is based on a merger of the Aliens and the Predators franchise: two legendary alien species that are in conflict with each other, fighting to the death with human marines caught in between. The first person shooter game was developed by Rebellion Studios, who also developed the first AVP PC title and released in February 2010. It is one of the first DirectX 11 games with support for new features like tesselation, which is why AMD heavily promoted it at the time of their DX 11 card launches. We use the AVP benchmark utility with tesselation and advanced DX11 shadows enabled.

Batman: Arkham City

Batman is back on the LCD screen with Arkham City, a sequel to Batman: Arkham Asylum, by Rocksteady Games and WB. It was released to the PC platform in November. Batman is imprisoned in Arkham City, an infamous district of the DC Universe that contains the scum of Gotham, most of which Batman helped get in there. In order to get out he must go through scores of baddies, and encounter many of the iconic super-villains along the way. He's not entirely alone.
Batman Arkham City uses the same Unreal Engine by Epic, as Arkham Asylum, but thanks to the engine's modularity, it has been overhauled, outfitted with the latest technologies, including a graphics engine that takes advantage of DirectX 11.

Battlefield 3

Arguably the most anticipated online shooter title among real gamers - PC gamers, Battlefield 3 is the latest addition to some of the most engaging online multi-player shooter franchises. It combines infantry combat with mechanized warfare including transport vehicles, armored personnel carriers, main battle tanks, attack helicopters, combat aircraft, pretty much everything that goes into today's battlefields. The infantry combat is coupled with role-playing elements, which makes the experience all the more engaging. It also has a single-player campaign which added a few gigabytes to its installer.
Behind all this is a spanking new game engine by EA-DICE, Frostbite 2. It makes use of every possible feature DirectX 11 has to offer, including hardware tessellation, and new lighting effects, to deliver some of the most captivating visuals gamers ever had access to. Not playing this game on PC is grave injustice to what's in store. Faster PCs are rewarded with better visuals.

BattleForge

BattleForge, a card based RTS, is developed by the German EA Phenomic Studio. A few months after launch the game was transformed into a Play 4 Free branded game. That move and the fact that it was included as game bundle with a large number of ATI cards made it one of the more well known RTS games of 2009. You as a player assemble your deck before game to select the units that will be available. Elemental force choices can be from forces of Fire, Frost, Nature and Shadow to complement each other.
The BattleForge engine has full support for DX 9, DX 10 and DX 10.1, we use the internal benchmark tool in DirectX 11 mode with highest settings to acquire our results.

Call of Duty 4

Call of Duty 4 is a first-person shooter that is built on the award winning Call of Duty Series. It is the first version to play in modern times. In a near-future conflict between the United States, Europe and Russia you get to play as a United States Marine and a British SAS operative. The engine is Infinity Ward's own creation and has true dynamic lighting, depth of field, dynamic shadows and HDR. Even though the game plot is scripted you will find yourself in intense battles, often working together with computer controlled team mates. Later installments of the Call of Duty Series use the same game engine, so this test is also representative of Call of Duty: Modern Warfare 3 performance.

Civilization 5

Sid Meier's Civilization V (or Civ 5 in common jargon), is the latest addition to the franchise of masterfully-crafted realtime strategy games that let you play God to a nascent civilization of your choice all the way up to the space-age. Civilization V uses large 3D worlds that are procedurally-generated, and takes advantage of hardware tessellation features offered by DirectX 11 to exponentially step up complexity of cities, models, terrains, and objects. It is also expected of this generation of GPUs to handle the larger texture loads that come with the eye-candy.

Crysis

After the tremendous success of Far Cry, the German game studio Crytek released their latest shooter Crysis in 2007. The game was by far the most hyped and anticipated game in 2007, and forums were full of "Can my system run Crysis?" threads because of the high hardware requirements of this game. Just like in Far Cry the plot evolves on a small island with a thick and richly detailed jungle world. A lot of attention has been given to small details like accurate physics. For example when you fire on a tree trunk, it will shatter and the tree will fall over leaving a stump behind. Enemies in a car can be stopped by shooting the tire of the car. The game graphics are top notch, even today, yet the game still runs well on most computers.

Crysis 2

Crysis 2 takes the player into an alien-infested New York City. The game adds a tactical options mode that allows several approaches to attack a heavily infested enemy location. The new Nanosuit 2.0 that the player uses offers more freedom in ability use, for example multiple abilities can be used at the same time. To better accomodate a given play style weapons can be customized with silencers, laser sights or even a sniping scope.
For rendering Crytek's CryEngine 3 is used which comes with reduced system requirements compared to the first Crysis game. Since Crysis 2 is a multi-platform game, with major development focus on console, the graphics on launch day were only DirectX 9. DirectX 11 functionality was added later in a patch. We use the DX11 version and the high-res texture pack for our benchmarking.

DiRT 3

The latest addition to the Collin McRae Rally franchise, DiRT 3, of multi-format rally motorsport. DiRT 3 introduced more of the same great racing experience Collin McRae DiRT 2 gave you, but with better gameplay, and the new Gymkhana freestyle motor-acrobatics mode, which you'll more likely love than hate. It uses a more polished, performance-optimized version of the EGO engine, version 2.0, which takes advantage of more DirectX 11 features than version 1.0 used on Collin McRae DiRT 2, did.

Dragon Age II

Dragon Age II is the second game in BioWare's Dragon Age franchise and was released in March 2011. As player, named Hawke, you will be able to pick your hero from several classes and grow him over the course of the adventure. Gameplay takes you through a linear narrated story of Hawke's rise to become the legendary "Champion of Kirkwall".
BioWare's Lycium Engine has support for DirectX 11, using tesselation, advanced dynamic lighting and camera effects like depth of field. We benchmark the DX11 version with details set to highest.

Hard Reset

Developed by Flying Wild Hog, a studio that prides itself with the fact that its creation is PC-exclusive (bless them), Hard Reset is a first person shooter that's set in a future cyberpunk setting of a dystopian world. It reintroduces many of the gameplay mechanics that made classics such as Quake wicked fun, which today's tactical military shooters eroded, creating a 'void' for.
The game uses the studio's in-house Road Hog Engine, which isn't particularly heavy on new-generation DirectX features, but can still get taxing with some GPUs.

Metro 2033

Metro 2033 is a first-person shooter game that is set in a post apocalyptic Moscow - as the name suggests inside the metro system. You will fight mutants or other humans who like to take away your shelter. The game has many gameplay elements similar to STALKER, also the engine has similar features. This is because two STALKER engine programmers left GSC Game World and started their own company which is now making Metro 2033.
The engine has support for all the latest eye candy like DirectX 11 and Tesselation. Unfortunately it leaves a less than optimized impression, making it a candidate to surpass Crysis for the highest hardware requirements. We test in DirectX 11 mode with details set to "Very High".

STALKER: Call of Pripyat

STALKER: Call of Pripyat takes places shortly after the events of the previous game STALKER: Shadow of Chernobyl. The player is one of many stalkers who are attracted by the Zone in hope of finding fame, wealth and artifacts. Over the course of the game you meet Strelok, the protagonist of the first STALKER game and team up with him to progress through the Zone.
An updated X-Ray Engine 1.6 powers the game with support for DirectX 11 using Compute Shaders for improved shadow rendering and tesselation to improve model quality.

StarCraft II

StarCraft II, released in July 2010, is a sequel to Blizzard's award-winning strategy game StarCraft. In the 26th century three species Terrans, Protoss and Zerg are at war. The campaign takes you through many missions on different planets where you have to face the various enemy factions, sometimes several of them. StarCraft II features a similar number of units as the original game, some of them new. Due to the massive success of the first game, Blizzard chose to focus large aspects of the game on multiplayer combat through Battle.net. The campaign serves as a good introduction to units and concepts and competitive multiplayer is where the action is at.
The StarCraft 2 engine supports only DirectX 9, but several patches have improved rendering quality and available options considerably. We test using a recorded 1 vs. 1 multiplayer replay in the late game phase. Please note that Star Craft II is very CPU limited on high-end cards, especially on lower resolutions, so you may not see much scaling between some cards.

Total War: Shogun 2

Set in 16th century feudal Japan, Total War: Shogun 2 takes the player on a quest for domination to conquer and unite the warlords of Japan. Moving away from the European setting of previous Total War games, the game is now designed around principles of the brilliant Chinese general Sun Tzu and his book "The Art of War". Gameplay is switched between real-time battles during which units on the battlefield are controlled and turn-based strategy which enable diplomacy, economy and production management. Taking control of a castles is comprised of several different stages which adds more complexity to warfare.
We benchmark using the highest settings in DirectX 11 mode, which was added via patch after release.

The Elder Scrolls: Skyrim

This isn't just a game, it's a masterpiece. A very large sandbox game that rejects the quality-quantity inverse-proportionality. By genre a role-playing game, TES: Skyrim combines some of the best elements of older titles in the franchise, with some new sandbox elements to churn out an extremely engaging, and addictive game. It makes use of Bethesda's Creation Engine, which isn't visually-intensive in that it doesn't use taxing graphics features, but the game's presentation itself, with large open worlds, end up taxing your hardware. Faster GPUs result in smoother gameplay with most eyecandy turned on.

World of Warcraft: Cataclysm

World of Warcraft is the most successful massively multiplayer online game in the world with far over 12 milion monthly subscribers. The game is centered around the epic battle between the Horde and Alliance factions with many other races getting involved in a long and complex story line. Even though it has been released in 2004, Blizzard has always added incremental improvements to the graphics, especially with new expansions. One key success of World of Warcraft is that it will run on a large number of slower systems, but also delivers a decent graphics experience on high-end systems. We test in DirectX 11 mode with details set to "Ultra".

3DMark 11

3DMark 11 is the very latest from the house of Futuremark, which has given out some of the most comprehensive benchmark applications for PC enthusiasts and gamers. 3DMark 11, as the name might probably suggest, makes use of Microsoft DirectX 11 API, and puts every feature at its disposal to use, creating astonishingly-realistic visuals. In the process, it evaluates DirectX 11 compliant GPUs, and lets gamers know what to expect from games from the near future that make use of the API, in terms of visual realism. The tessellation and depth of field tests are particularly of interest here.

Unigine Heaven 2.0

Unigine Heaven was one of the first demos that supported DirectX 11. Heaven is a technology demonstration for Unigine engine which supports DirectX 9 through 11 and OpenGL too. Version 2.0 adds more scenes and optionally more complex tesselation features. While there is some controversy surrounding the benchmark whether it is an accurate representation of what to expect from future games in regards to DirectX 11 we still chose it as test to get an insight into potential future gaming.

Power Consumption

Cooling modern video cards is becoming more and more difficult, especially when users are asking for quiet cooling solutions. That's why the engineers are now paying much more attention to power consumption of new video card designs. An optimized fan profile is also one of the few things that board vendors can do to impress with reference designs where they are prohibited to make changes to the thermal solution or components on the card.

For this test we measure power consumption of only the graphics card, via PCI-Express power connector(s) and PCI-Express bus slot. A Keithley Integra 2700 with 6.5 digits is used for all measurements. Again, the values here reflect card only power consumption measured at DC VGA card inputs, not the whole system.

We chose Crysis 2 as a standard test representing typical 3D gaming usage because it offers: - very high power draw - high repeatability - is a current game that is supported on all cards due to its DirectX 9 nature - drivers are actively tested and optimized for it - supports all multi-GPU configurations - test runs a relatively short time and renders a non-static scene with variable complexity.

Our results are based on the following tests:

• Idle: Windows 7 Aero sitting at the desktop (1280x1024 32-bit) all windows closed, drivers installed. Card left to warm up in idle until power draw is stable.
• Multi-Monitor: Two monitors connected to the tested card, which use different display timings. Windows 7 Aero sitting at the desktop (1280x1024 32-bit) all windows closed, drivers installed. Card left to warm up in idle until power draw is stable.
• Average: Crysis 2 at 1920x1200, Extreme profile, representing a typical gaming power draw. Average of all readings (12 per second) while the benchmark was rendering (no title/loading screen).
• Peak: Crysis 2 at 1920x1200, Extreme profile, representing a typical gaming power draw. Highest single reading during the test.
• Maximum: Furmark Stability Test at 1280x1024, 0xAA. This results in a very high non-game power consumption that can typically be reached only with stress testing applications. Card left running stress test until power draw converged to a stable value. On cards with power limiting systems we will disable the power limiting system or configure it to the highest available setting - if possible. We will also use the highest single reading from a Furmark run which is obtained by measuring faster than when the power limit can kick in.
• Blu-ray Playback: Power DVD 9 Ultra is used at a resolution of 1920x1200 to play back the Batman: The Dark Knight disc with GPU acceleration turned on. Playback starts around timecode 1:19 which has the highest data rates on the BD with up to 40 Mb/s. Playback left running until power draw converged to a stable value.

Power consumption of the AMD Radeon HD 7970 is impressive.

The move from 40 nm to 28 nm itself promises reduced power requirements. Then AMD also worked on technologies that improve power consumption through various methods.

Idle power consumption, which matters for normal desktop work, has seen a substantial improvement, going down to 12 Watts, which is about where you would see last-generation's lower-end cards. Unfortunately multi-monitor power consumption has not received such big improvements. 45 W when running two or more screens, seems like a lot, considering single monitor power draw is down to 12 Watts now.

For gaming we see spectacular values, especially with performance taken into account. The HD 7970 consumes 20 W less power, yet is over 30% faster than the last generation HD 6970. This ensures that the card claims a leading spot in our performance per Watt ranking.

Blu-ray media playback power is a bit high at 53 W. During testing I noticed that the card went into a lower power state of around 40 W every few seconds, but jumped out of it almost immediately. With some additional tweaking I'm sure AMD can get this fixed.

A new feature of the HD 7900 Series is AMD ZeroCore Power, which will power off the card as soon as the monitor output is blanked, during screen saver for example. For additional power and noise reduction the fan will stop in this state, too. We measured a power consumption of 1.11 Watts for the whole graphics card during ZeroCore Power. As soon as you move the mouse the PC is back immediately, there is no lag or any delay.

Please note that ZeroCore Power seems to engage only when the screen is completely static. If you have an application running that draws to the screen, the monitor will go black, but the card will not enter the low power state or return from it quickly. To avoid this, minimize all applications and let Windows sit at the desktop.

Fan Noise

In the past years users would accept everything just to get more performance. Nowadays this has changed and people have become more aware of the fan noise and power consumption of their graphic cards.
In order to properly test the fan noise a card emits we are using a Bruel & Kjaer 2236 sound level meter (~$4,000) which has the measurement range and accuracy we are looking for.


The tested graphics card is installed in a system that is completely passively cooled. That is passive PSU, passive CPU cooler, passive cooling on the motherboard and a solid state drive.
This setup allows us to eliminate secondary noise sources and test only the video card. To be more compliant with standards like DIN 45635 (we are not claiming to be fully DIN 45635 certified) the measurement is conducted at 100 cm distance and 160 cm over the floor. The ambient background noise level in the room is well below 20 dbA for all measurements. Please note that the dbA scale is not linear, it is logarithmic. 40 dbA is not twice as loud as 20 dbA. A 3 dbA increase results in double the sound pressure. The human hearing is a bit different and it is generally accepted that a 10 dbA increase doubles the perceived sound level. The 3D load noise levels are tested with a stressful game, not Furmark.

Idle fan noise is acceptable, even though we have seen much quieter cards.

Fan noise under load is simply too high. 46 dbA is far from a quiet experience, and not far from the 48 dbA of the NVIDIA GTX 590, which is about 20% faster and produces almost twice the heat.

Performance Summary

The graphs on this page show a combined performance summary of all tests and resolutions from previous pages. Each graph shows the tested card as 100% and all other cards' performance relative to it. A sixth graph summarizes all tests in all resolutions to calculate the total relative performance of the review sample.

Performance per Watt

Using the relative performance scores from the previous page and the typical gaming power consumption result, the following graphs show efficiency of the cards in our test group.

Performance per Dollar

If you are looking for the best bang for the buck, then you will love this graph. We looked up the current USD price of each card on the popular online shop Newegg and used it and the relative performance numbers to calculate the Performance per Dollar Index.

Overclocking

The overclocks listed in this section were achieved with the default fan and voltage settings as defined in the VGA BIOS. Please note that every single sample overclocks differently, that's why our results here can only serve as a guideline for what you can expect from your card.



Maximum stable clocks of our card are 1075 MHz core (16% overclock) and 1715 MHz Memory (25% overclock).

Both overclocks are spectacular. It's amazing to see over 15% overclocking potential on a brand-new GPU design, using a brand new production process. Over 1700 MHz memory clock is actually the highest memory speed we have ever seen!

Overclocked Performance

Using these clock frequencies we ran a quick test of Call of Duty 4 to evaluate the gains from overclocking.


Actual 3D performance gained from overclocking is 14.1%.

Temperatures

Idle temperatures are comfortably low. Under load the card reaches a very acceptable maximum temperature that still has plenty of headroom for overclocking.

Clock Profiles

Modern graphics cards have several clock profiles that are selected to balance power draw and performance requirements.
The following table lists the clock settings for important performance scenarios and the GPU voltage that we measured. We measure on the pins of a coil or capacitor near the GPU voltage regulator.

	Core Clock	Memory Clock	GPU Voltage (measured)
Desktop	300 MHz	150 MHz	0.85 V
Multi-Monitor	350 MHz	1375 MHz	0.96 V
Blu-ray Playback	925 MHz	1375 MHz	1.18 V
3D Load	925 MHz	1375 MHz	1.16 V

CCC Overdrive Limits
Core	1125 MHz
Memory	1575 MHz

Value and Conclusion

	According to AMD, the HD 7970 will cost $549 (€ 499) and will be available starting January 9.
	Solid performance increase Low power consumption Excellent overclocking potential Native full-size HDMI & DisplayPort output Dual BIOS 3 GB of memory Adds support for PCI-Express 3.0 and DirectX 11.1 Support for multiple independent audio streams
	Noisy in 3D High price CCC Overdrive limits too low PowerTune and ZeroCore may complicate advanced overclocking
9.0	AMD's new Radeon HD 7970 successfully introduces the company's new Graphics Core Next GPU architecture to the market. The move away from the VLIW shaders promises increased flexibility and higher performance, while still retaining compatibility with driver optimizations done for the VLIW architecture. In our testing we see AMD's new flagship cruise past NVIDIA's GeForce GTX 580 with a healthy 15% performance margin averaged over all our benchmarks - at 2560x1600, making it the fastest single GPU graphics card in the world. 2560x1600, at least 1920x1200, should be your gaming resolution, otherwise the card won't be able to play out its full potential. With EyeFinity support for several monitors, the card could also power a triple monitor high-end gaming rig. Moving to a 28 nanometer GPU process brings a significant reduction in power consumption by itself. AMD has improved on this with several new technologies. Combined, this results in the highest performance per Watt result we have seen from any high-end card so far. With well below 200 W during typical gaming, the card's heat output is reduced as well. Even though AMD claims improvements in their heatsink design, the card is still fairly noisy during 3D gaming, not much quieter than NVIDIA's GTX 590, which is a dual GPU design that sucks twice as much power. In terms of overclocking we saw spectacular results for a new GPU architecture based on a new production process. We could increase the GPU frequency by over 15% and memory by 25% which results in the highest memory clock we have ever seen on a graphics card. Overclockers will also like the Dual BIOS feature which acts as a safety net in case something goes wrong during a BIOS flash. AMD's suggested pricing for the HD 7970 is $549, which feels a bit expensive, I would consider $499 reasonable and $449 a bargain deal. However, even at its $549 price point, the Radeon HD 7970 is an ok deal when looking at other options in the high-end graphics card market. And so begins another year, another generation of GPUs, and another round of fierce competition between AMD and NVIDIA, once the latter catches up with its GeForce Kepler GPU family. Watch this space.