ATI Radeon HD 5970 Hemlock Review

ATI now dominates the graphics industry with their DirectX 11 Radeon HD 5000-series video cards, which allows them to enjoy current-generation gaming on Windows 7 and Vista. The ATI Radeon HD 5870 has already beat the worn and weathered GeForce GTX 285, just as the Radeon HD 5850 and HD5770 do at their respective price points. By combining two Cypress XT GPUs together on one PCB, the ATI Radeon HD 5970 video card will now compete against the very best NVIDIA can offer: the GeForce GTX 295. In this article, Benchmark Reviews tests the 'unlocked' Hemlock GPU against the top graphics products available and demonstrates just how much ground AMD has gained in a few short months.

After years of leading the industry, NVIDIA has finally succumbed defeat to ATI. The back and forth battle between chip makers has seen Radeon top GeForce, and then GeForce top Radeon. This time, though, it has a lasting affect. When ATI launched the Radeon HD 5000 series in September (2009) they released a level of graphics product that NVIDIA was unprepared for. On the same day that ATI claimed the top end, NVIDIA responded by redefining the bottom end, and launch the GeForce 200 (a discrete video card with graphics power so low it competes with integrated graphics). Even two months later as the top-end Radeon HD 5970 launches, NVIDIA fires back with another low end alternative: the GeForce GT 240. From the reports, it won't be until February 2010 before NVIDIA launches their 40nm Fermi architecture. Even then, DirectX 11 support is unconfirmed.

Long before the 'Hemlock' is tested on its first benchmark, gamers really want to know if the ATI Radeon HD 5970 is simply a convenient package for two Radeon HD 5850's in CrossFire. The specifications might make it appear that way, but AMD has designed the HD5970 to overclock up-to and beyond the HD5870 specifications. So for those willing to push the Hemlock GPU further, the HD5970 could perform better than two HD5870's in CrossFire. Power efficient improvements are meant to reduce electrical draw, intended to make the HD5970 a more energy-efficient solution.

AMD has launched the Radeon 5000-series as their first assault on the multi-monitor ATI Eyefinity Technology feature, using native HDMI 1.3 output paired with DisplayPort connectivity. The new Cypress GPU features the latest ATI Stream Technology, which is designed to utilize DirectCompute 5.0 and OpenCL code. These new features improve all graphical aspects of the end-user experience, such as faster multimedia transcode times and better GPGPU compute performance. AMD has already introduced a DirectCompute partnership with CyberLink, and the recent Open Physics Initiative with Pixelux promises to offer physics middleware built around OpenCL and Bullet Physics. This looks like ATI's recipe for success, since NVIDIA does not have a GPU to compete against the Radeon 5800 series or support DirectX 11. It doesn't help matters any that NVIDIA GPUs do not support OpenCL and DirectCompute 11 environments, leaving them out in the cold for the coming winter months.

From these developments ATI has distanced themselves ahead of NVIDIA by placing gamers first in their consideration, and have positioned the ATI 5000-series to introduce enthusiasts to a new world of DirectX 11 video games on the Microsoft Windows 7/Vista Operating Systems. While most hardware enthusiasts are familiar with the back-and-forth competition between these two leading GPU chip makers, it might come as a surprise that NVIDIA actually remarked that DirectX 11 video games won't fuel video card sales, and have instead decided to revolutionize the military with CUDA technology. Perhaps we're seeing the evolution of two companies: NVIDIA transitions to the industrial sector and departs the enthusiast gaming space, while ATI successfully answers retail consumer demand.

About AMD/ATI

ATI is the graphics brand from AMD, and designs top-to-bottom discrete graphics processors that deliver outstanding 3D gaming performance; an immersive multimedia experience; and break-through energy efficiency. Advanced Micro Devices (NYSE: AMD) is an innovative technology company dedicated to collaborating with customers and technology partners to ignite the next generation of computing and graphics solutions at work, home and play. For more information, visit https://www.amd.com/.

ATI Radeon HD 5970 Features

• Microsoft DirectX 11 Support (only available on Windows 7 and Vista)
• ATI Eyefinity Technology
• ATI Stream Technology Designed for DirectCompute 5.0 and OpenCL
• Accelerate Video Transcoding
• 40 nm Process Hemlock GPU Technology
• Advanced GDDR5 Memory Technology
• 2nd Generation TeraScale Engine
• Microsoft Windows 7 Support
• ATI CrossFireX Technology
• Enhanced Anisotropic Filtering
• Accelerated Video Transcoding
• Display Flexibility, Supports DL-DVI, DP, HDMI and D-Sub
• On chip HDCP Support
• ATI CrossFireX multi-GPU support for highly scalable performance. (Use up to four discrete cards with an AMD 790FX based motherboard)
• ATI Avivo HD Support
  • ATI Unified Video Decoder 2 (UVD) for Blu-ray and HD Video.
  • Accelerated Video Transcoding (AVT)
  • DVD Upscaling
  • Dynamic Contrast
  • Built-in HDMI with 7.1 surround sound support
  • Dynamic power management with ATI PowerPlay technology including memory clocks
  • Dolby TrueHD and DTSHD Master Audio Support

ATI Hemlock Specifications

• AISC: ATI Radeon HD 5870
• Bus Interface: PCI-E x16 (PCI-E 2.0/2.1)
• Memory: 2048MB / 512-bit GDDR5 (256-bit per GPU)
• Clock Speed: 725 MHz Eclk / Effective 4000 MHz Mclk
• Cooling System: Dual slot Fan with auto fan control
• Bracket: Full Height
• Display Support: Dual DL-DVI-I, Display Port, HDMI, VGA (Via adaptor)
• HDCP: Supported
• Crossfire Support: Native Hardware Crossfire
• External Power: PCIe Graphic External 8+6 pin

Video Card Attributes

ATI Eyefinity Multi-Monitors

ATI Eyefinity advanced multiple-display technology launches a new era of panoramic computing, helping to boost productivity and multitasking with innovative graphics display capabilities supporting massive desktop workspaces, creating ultra-immersive computing environments with super-high resolution gaming and entertainment, and enabling easy configuration and supporting up to six independent display outputs simultaneously.

In the past, multi-display systems catered to professionals in specific industries. Financial, gas and oil, and medical are just some industries where multi-display systems are not only desirable, but a necessity. Today, even graphic designers, CAD engineers and programmers are attaching more than one displays to their workstation. A major benefit of a multi-display system is simple and universal - it enables increased productivity. This has been demonstrated in industry studies which have shown that attaching more than one display device to a PC can significantly increase user productivity.

The early multi-display solutions were non-ideal. The bulky CRT monitors claimed too much desk space, thinner LCD monitors were very expensive, and external multi-display hardware was inconvenient and also very expensive. These issues are much less of a concern today. LCD monitors are very affordable and current generation GPUs can drive multiple display devices independently and simultaneously, without the need for external hardware. Despite the advancements in multi-display technology, AMD engineers still felt there was room for improvement, especially regarding the display interfaces. VGA carries analog signals and needs a dedicated DAC per display output, which consumes power and ASIC space. Dual-Link DVI is digital, but requires a dedicated clock source per display output and uses too many IO pins from our GPU. If we were to overcome the dual display per GPU barrier, it was clear that we needed a superior display interface.

In 2004, a group of PC companies collaborated to define and develop DisplayPort, a powerful and robust digital display interface. At that time, engineers working for the former ATI Technologies Inc. were already thinking about a more elegant solution to drive more than two display devices per GPU, and it was clear that DisplayPort was the interface of choice for this task.

In contrast to other digital display interfaces, DisplayPort does not require a dedicated clock signal for each display output. In fact the data link is fixed at 1.62Gbps or 2.7Gbps per lane, irrespective of the timing of the attached display device. The benefit of this design is that one reference clock source can provide the clock signals needed to drive as many DisplayPort display devices as there are display pipelines in the GPU. In addition, with the same number of IO pins used for Single-Link DVI, a full speed DisplayPort link can be driven which provides more bandwidth and translates to higher resolutions, refresh rates and color depths. All these benefits perfectly complement ATI Eyefinity Multi-Display Technology.

ATI Eyefinity Technology from AMD provides advanced multiple monitor technology delivering an incredibly immersive graphics and computing experience with innovative display capabilities, supporting massive desktop workspaces and super-high resolution gaming environments. Legacy GPUs have supported up to two display outputs simultaneously and independently for more than a decade. Until now graphics solutions have supported more than two monitors by combining multiple GPUs on a single graphics card. With the introduction of AMD's next-generation graphics product series supporting DirectX 11, a single GPU now has the advanced capability of simultaneously supporting up to six independent display outputs.

ATI Eyefinity Technology is closely aligned with AMD's DisplayPort implementation providing the flexibility and upgradability modern user's demand. Up to two DVI, HDMI, or VGA display outputs can be combined with DisplayPort outputs for a total of up to six monitors, depending on the graphics card configuration. The initial AMD graphics products with ATI Eyefinity technology will support a maximum of three independent display outputs via a combination of two DVI, HDMI or VGA with one DisplayPort monitor. AMD has a future product planned to support up to six DisplayPort outputs. Wider display connectivity is possible by using display output adapters that support active translation from DisplayPort to DVI or VGA.

The DisplayPort 1.2 specification is currently being developed by the same group of companies who designed the original DisplayPort specification. This new spec will include exciting new features for our customers. Its feature set includes higher bandwidth, enhanced audio and multi-stream support. Multi-stream, commonly referred to as daisy-chaining, is the ability to address and drive multiple display devices through one connector. This technology, coupled with ATI Eyefinity Technology, will re-introduce multi-display technology and AMD will be at the forefront of this transition.

ATI Radeon HD 5970 Closer Look

During the 2009 business year we've seen dozens of great video games released on the PC platform, but very few of them demand any more graphical processing power than games demanded back in 2006. Video cards certainly got bigger and faster, but video games has lacked fresh development. DirectX 10 helped the industry, but every step forward received two steps back because of the Windows Vista O/S. Introduced with Windows 7 (and also now available on Windows Vista with an update), enthusiasts now have DirectX 11 detail and special effects in their video games.

DirectX 11 is that software development leap we've been waiting for. Screen Space Ambient Occlusion (SSAO) is given emphasis in DX11, allowing some of the most detailed computer textures gamers have ever seen. Realistic cracks in mud with definable depth and splintered tree bark make the game more realistic, but they also make new demands on the graphics hardware. ATI offers the Radeon HD 5970 Hemlock video card for enthusiast gamers wanting nothing but the best graphics experience.

The ATI Radeon HD 5970 is one of the longest graphics cards ever produced by AMD, measuring 12.1" long from end to end. This profile relegates the HD5970 to full-size ATX cases with plenty of breathing room. ATI produces and directly supplies every Radeon HD 5970 that operates at reference specification, with add-in card/add-in board partner artwork applied wherever necessary.

In terms of overall design, the Radeon HD5970 looks like a stretched version of the HD5870 or HD5850. The biggest difference between the Radeon 5970 and the 5800 series is the lacking native HDMI 1.3 port, and the replacement of a DisplayPort-A connection for the smaller DisplayPort-B input. Two DVI digital video outputs are connected to monitors for dual-view, or a third monitor can be added via DisplayPort to enable ATI Eyefinity technology.

Unlike the other cards in the Radeon HD 5000-series, namely the HD5850/5870, the Hemlock HD5970 exhausts the heated air outside of the computer case. This is critically important, considering that the reference-speed Radeon HD 5970 we received for testing often passed 90°C in a 20°C room. The heated air coming out of the graphics card was hot, but nothing compared to the backside of the Radeon HD 5970. The 'Hemlock' graphics card pairs two Cypress-XT GPUs, and under load the retention brackets were measured to exceed 80°C - enough to scorch the hand that touches it.

The exposed view below illustrates the compact layout of electronic components which build the Radeon HD 5970. ATI intentionally reduced the clock speeds of the two Cypress-XT GPUs and GDDR5 vRAM to reduce TDP to under 300 watts at full load. ATI's uses the revamped PLX ExpressLane CrossFire bridge (good for 96-lanes of PCI-Express 2.1 bandwidth), which enables a new feature they call 'CrossFire Ultra-low Power State'. This new feature shuts down one of the GPU's and places it into a suspend power state when the video card doesn't require the graphics processing power. The end-result is an extremely efficient graphics card at idle.

Although designed for under 300W, the 6+8-pin power connections are good for up to 400W on demand and the cooling unit also offers more thermal control than necessary. AMD uses digital programmable Volterra regulators, Japanese pure ceramic SuperCapacitors, and 5.0 Gbps rated Qimonda GDDR5 to allow additional overclocking headroom. Doing this has enabled ATI to offer a quasi-Black Edition video card so that overclockers can have their way with it. ATI in-house tests pushed a Radeon HD 5970 up to 970MHz on each GPU, but individual results will vary depending on manufacturer and chip binning.

Considering the Cypress-XT GPU die size is a rather large 334 mm2, and fits 2154 million transistors, the overall heat dissipation is spread over a suitable landscape. The double-height cooler does a very good job of cooling two GPUs, but there is still a tremendous amount of heat that builds up on backside of the PCB. If you plan to overclock, I recommend that special attention be paid to the alloy plate mounted to the backside of the PCB of this video card.

In next several sections, Benchmark Reviews details our methodology for testing video cards followed by a performance comparison against many of the most popular graphics accelerators available. The Sapphire ATI Radeon HD 5970 competes against the dual-GT200 GPU NVIDIA GeForce GTX 295, so of course we'll be keeping a close eye on comparative performance.

VGA Testing Methodology

As of November 2009 Benchmark Reviews has discontinued testing on the Windows XP (DirectX 9) Operating System, although it is recognized that 50% or more of the gaming world still use this O/S. DirectX 11 is native to the Microsoft Windows 7 Operating System, and will be the centerpiece of our test platform for the foreseeable future. DirectX 11 is also available as a Microsoft Update to the Vista O/S. Because not all graphics solutions were DX11 compatible at the time this article was published, DirectX 10 test settings are utilized on the Windows 7 platform to standardize results.

According to the Steam Hardware Survey published at the time of Windows 7 launch, the most popular gaming resolution is 1280x1024 (17-19" standard LCD monitors) closely followed by 1024x768 (15-17" standard LCD). However, because these resolutions are considered 'low' by most standards, our benchmark performance tests concentrate on the up-and-coming higher-demand resolutions: 1680x1050 (22-24" widescreen LCD) and 1920x1200 (24-28" widescreen LCD monitors). These resolutions are more likely to be used by high-end graphics solutions, such as those tested in this article.

In each benchmark test there is one 'cache run' that is conducted, followed by five recorded test runs. Results are collected at each setting with the highest and lowest results discarded. The remaining three results are averaged, and displayed in the performance charts.

Intel X58 Test System

• Motherboard: Gigabyte GA-EX58-UD4P (Intel X58/ICH10R Chipset) with F10h BIOS
• Processor: Intel Core i7-920 Nehalem 2.67 GHz (BX80601920)
• System Memory: 6GB (3x 2GB) OCZ Triple-Channel 1333 MHz DDR3 CL 7-7-7-20 OCZ3P1333LV6GK
• Primary Drive: OCZ Vertex EX SLC SSD OCZSSD2-1VTXEX120G
• Power Supply Unit: Antec TruePower Quattro 850W Modular PSU TPQ-850
• Monitor: 26-Inch Widescreen LCD (up to 1920x1200@60Hz)

Benchmark Applications

• 3DMark Vantage v1.01 (Extreme Quality, 8x Multisample Anti-Aliasing, 16x Anisotropic Filtering, 1:2 Scale)
• BattleForge (Very High Quality, 8x Anti-Aliasing, Auto Multi-Thread)
• Crysis Warhead v1.1 with HOC Benchmark (DX10, Very High Quality, 4x Anti-Aliasing, 16x Anisotropic Filtering, Airfield Demo)
• Devil May Cry 4 Benchmark Demo (DX10, Super-High Quality, 8x MSAA)
• Far Cry 2 v1.02 (DX10, Very High Performance, Ultra-High Quality, 8x Anti-Aliasing, HDR + Bloom)
• Resident Evil 5 Benchmark Demo (DX10, Super-High Quality, 8x MSAA)
• S.T.A.L.K.E.R. Call of Pripyat Benchmark Demo (Ultra-Quality, Enhanced DX10 light, 4x MSAA, SSAO on and off)
• Unigine Heaven Benchmark Demo (DX11 and DX10, High-Quality Shaders, Tessellation, 16x AF, 4x AA)

Video Card Test Products

3DMark Vantage GPU Tests

3DMark Vantage is a PC benchmark suite designed to test the DirectX10 graphics card performance. FutureMark 3DMark Vantage is the latest addition the 3DMark benchmark series built by FutureMark corporation. Although 3DMark Vantage requires NVIDIA PhysX to be installed for program operation, only the CPU/Physics test relies on this technology.

3DMark Vantage offers benchmark tests focusing on GPU, CPU, and Physics performance. Benchmark Reviews uses the two GPU-specific tests for grading video card performance: Jane Nash and New Calico. These tests isolate graphical performance, and remove processor dependence from the benchmark results.

3DMark Vantage GPU Test: Jane Nash

Of the two GPU tests 3DMark Vantage offers, the Jane Nash performance benchmark is slightly less demanding. In a short video scene the special agent escapes a secret lair by water, nearly losing her shirt in the process. Benchmark Reviews tests this DirectX 10 scene at 1680x1050 and 1920x1200 resolutions, and uses Extreme quality settings with 8x anti-aliasing and 16x anisotropic filtering. The 1:2 scale is utilized, and is the highest this test allows. By maximizing the processing levels of this test, the scene creates the highest level of graphical demand possible and sorts the strong from the weak.

As far as 3dMark Vantage is concerned, the ATI Radeon HD 5770, NVIDIA GeForce GTX 260, and Radeon HD 4890 are all less than optimal graphics solutions for 'extreme' settings like 8x AA and 16x AF. With that being the case, we'll concentrate more on the higher-level video cards.

The ATI Radeon HD 5850 outperforms the overclocked ASUS GTX 285 TOP model in both resolutions, while the Radeon HD 5870 widens this lead as the most powerful single-GPU graphics card with a full 30% advantage at 1680x1050 and nearly 32% at 1920x1280. While the NVIDIA GeForce GTX 295 does manage to out-muscle the Radeon HD 5870, it's nowhere near the level demonstrated by the ATI Radeon HD 5970 Hemlock video card. The dual Cypress-XT GPUs in the HD5970 offer a 33% advantage over the dual GT200 GPUs in the GeForce GTX 295, when operating at 1920x1200.

3DMark Vantage GPU Test: New Calico

New Calico is the second GPU test in the 3DMark Vantage test suite. Of the two GPU tests, New Calico is the most demanding. In a short video scene featuring a galactic battleground, there is a massive display of busy objects across the screen. Benchmark Reviews tests this DirectX 10 scene at 1680x1050 and 1920x1200 resolutions, and uses Extreme quality settings with 8x anti-aliasing and 16x anisotropic filtering. The 1:2 scale is utilized, and is the highest this test allows. Using the highest graphics processing level available allows our test products to separate themselves and stand out (if possible).

Last year's top-end is this year's middle market graphics solution, which is why a newly released low-end Radeon HD 5770 can match brawn with the GTX 260 and HD4890. The DirectX 11-compatible Cypress GPU clearly dominates the field in this test, allowing the ATI Radeon HD 5850 to easily overtake the GeForce GTX 285 by 11% at 1680x1050 and then an astonishing 27% at 1920x1200. The Radeon HD 5870 is well ahead of NVIDIA's GTX 285 by more than 40% at 16x10 and 46% at 19x12 resolutions. NVIDIA's GeForce GTX 295 is no match for the HD5870 in the New Calico benchmark, which allows the more-powerful ATI Radeon HD 5970 to literally step over it to establish a 61% performance advantage at 1920x1200.

Although Benchmark Reviews does not include CrossFire HD5850 test results in this article, past experience has shown us that two Radeon video cards placed in CrossFire do not amount to exactly twice the performance. Under optimal circumstances and the ideal game engine, an ATI Radeon CrossFire set may provide up to 90% improvement over a single video card. In terms of confronting the claim that the ATI Radeon HD 5970 is simply a portable CrossFire 5850 alternative, the HD5970 actually appears to match the 'best case' CrossFire performance. It should be noted however, that the HD5970 was built as an enthusiast video card and designed to be overclocked beyond the specs of a HD5870.

BattleForge Performance

BattleForge is free Massive Multiplayer Online Role Playing Game (MMORPG) developed by EA Phenomic with DirectX 11 graphics capability. Combining strategic cooperative battles, the community of MMO games, and trading card gameplay, BattleForge players are free to put their creatures, spells and buildings into combination's they see fit. These units are represented in the form of digital cards from which you build your own unique army. With minimal resources and a custom tech tree to manage, the gameplay is unbelievably accessible and action-packed.

Benchmark Reviews uses the built-in graphics benchmark to measure performance in BattleForge, using Very High quality settings (detail) and 8x anti-aliasing with auto multi-threading enabled. BattleForge is one of the first titles to take advantage of DirectX 11 in Windows 7, and offers a very robust color range throughout the busy battleground landscape. The first chart illustrates how performance measures-up between video cards when Screen Space Ambient Occlusion (SSAO) is disabled, which runs tests at DirectX 10 levels.

When SSAO is disabled, older GeForce and Radeon products are compared on a more even playing field (so long as you discredit the fact that we have a few DirectX 10 cards in the mix, and that BattleForge is a DirectX 11 game). Looking at performance using the 1920x1200 resolution, the ATI Radeon HD5770 Juniper GPU does extremely well and slightly outperforms the overclocked GTX 260 model with 29.1 FPS. The ASUS GeForce GTX 285 TOP comes in right behind them, scoring 34.0 FPS. ATI's Radeon HD 4890 still has muscle to flex, rendering 36.2 FPS and trailing just behind the Radeon HD 5850's score 38.8 FPS.

Even when Screen Space Ambient Occlusion is disabled to give older cards their last chance for high frame rates, the ATI Radeon HD 5870 reminds them that SSAO isn't a challenge it has to concern itself with. Scoring 45.3 FPS, the Radeon 5870 outperforms an overclocked GTX 285 by more than 33% when settings are uniform to accommodate all products. The surprise here is the dual-GPU performance from the NVIDIA GeForce GTX 295 is nearly even with the Radeon HD 5870 in BattleForge. Less surprising is that the dual-GPU Radeon HD 5970 pretends the GTX295 doesn't exist and produces frame rates almost 156% faster.

The next chart (below) illustrates how BattleForge reacts when SSAO is enabled, which forces multi-core optimizations that DirectX 11-compatible video cards are best suited to handle:

As expected, the DirectX 11-compatible ATI 5000 series runs rings around everything NVIDIA currently offers. SSAO isn't a technology that GeForce products handle very well, and thus far it seems that aside from CUDA technology there is little advantage to be gained over the Radeon alternative. If gaming is the primary purpose for a discrete graphics card, then you will want to consider that nearly all new video games will be developed with SSAO and other DirectX 11 extensions. These features make it difficult (and sometimes impossible) to enjoy the game on non-compliant graphics hardware.

In respect to EA's BattleForge, a reference-design mainstream ATI Radeon HD 5770 is able to outperform the overclocked ASUS GeForce GTX 285 TOP by 29% at 1920x1200, thus proving that Windows 7 will re-center the definition of 'mainstream' products. What was top shelf in Windows XP or Vista will soon become the low end with DirectX 11 in Windows 7. For gamers who plan to use Windows 7, and especially those who play BattleForge, the Radeon HD 5850 offered excellent performance and topped our overclocked GTX 285 by 91%. If that wasn't proof evident that NVIDIA should be worried, the ATI Radeon HD 5870 DirectX 11 video card was able to easily outperform NVIDIA's GTX 285 by more than 125% at the same price point when SSAO is called into action. NVIDIA's GTX 295 does much better than the GTX 285, but still falls just ahead of the HD5770 and below the HD4890. This allows the the Radeon HD 5970 to clinch a 266% FPS advantage of the GTX295, and clearing demonstrating how unprepared for SSAO the NVIDIA products are.

In terms of CrossFire HD5850 vs single HD5970, the ATI Radeon HD 5970 does appear to match the ideal CrossFire output. As previously mentioned, the HD5970 is intended for overclockers and is given the headroom to exceed HD5870 specifications. Doing so will boost performance well beyond a CrossFire set of HD5850's.

Crysis Warhead Test

Crysis Warhead is an expansion pack based on the original Crysis video game. Crysis Warhead is based in the future, where an ancient alien spacecraft has been discovered beneath the Earth on an island east of the Philippines. Crysis Warhead uses a refined version of the CryENGINE2 graphics engine. Like Crysis, Warhead uses the Microsoft Direct3D 10 (DirectX 10) API for graphics rendering.

Benchmark Reviews uses the HOC Crysis Warhead benchmark tool to test and measure graphic performance using the Airfield 1 demo scene. This short test places a high amount of stress on a graphics card because of detailed terrain and textures, but also for the test settings used. Using the DirectX 10 test with Very High Quality settings, the Airfield 1 demo scene receives 4x anti-aliasing and 16x anisotropic filtering to create maximum graphic load and separate the products according to their performance.

Using the highest quality DirectX 10 settings with 4x AA and 16x AF, only the most powerful graphics cards are expected to perform well in our Crysis Warhead benchmark tests. DirectX 11 extensions are not supported in Crysis: Warhead, and SSAO is not an available option.

The ATI Radeon HD 5770 does well, especially considering it's only meant as a lower mid-level graphics solution, yet still falls short of the GeForce GTX 260 and Radeon HD 4890 which share the same 21 frames per second performance. There are a few frames between the Radeon HD 5850 and the overclocked ASUS GeForce GTX 285 TOP, but at 1920x1280 those differences disappear. The Sapphire ATI Radeon HD 5870 model 21161-00-50R video card doesn't flex its muscle in this NVIDIA-optimized game the way it does in DirectX 11, but it still outperforms the GTX 285 within the same price point.

For the dual-GPU products, the NVIDIA GTX 295 muscles its way to the #2 position ahead of the Radeon HD 5870, but it falls very short of matching performance with the ATI Radeon HD 5970. At 1920x1200 resolution the GTX 295 produces only 27 FPS, and while this is still better than the HD5870's 23 FPS or the overclocked GTX285's 21 FPS, it's 44% behind the Radeon HD5970's 39 FPS. The single HD5970 actually comes close to doubling the performance of a single HD5850, but also consumes far less power in it's factory 'underclocked' state.

Devil May Cry 4 Benchmark

Devil May Cry 4 was released on PC in early 2007 as the fourth installment to the Devil May Cry video game series. DMC4 is a direct port from the PC platform to console versions, which operate at the native 720P game resolution with no other platform restrictions. Devil May Cry 4 uses the refined MT Framework game engine, which has been used for many popular Capcom game titles over the past several years.

MT Framework is an exclusive seventh generation game engine built to be used with games developed for the PlayStation 3 and Xbox 360, and PC ports. MT stands for "Multi-Thread", "Meta Tools" and "Multi-Target". Originally meant to be an outside engine, but none matched their specific requirements in performance and flexibility. Games using the MT Framework are originally developed on the PC and then ported to the other two console platforms.

On the PC version a special bonus called Turbo Mode is featured, giving the game a slightly faster speed, and a new difficulty called Legendary Dark Knight Mode is implemented. The PC version also has both DirectX 9 and DirectX 10 mode for Microsoft Windows XP and Vista Operating Systems.

It's always nice to be able to compare the results we receive here at Benchmark Reviews with the results you test for on your own computer system. Usually this isn't possible, since settings and configurations make it nearly difficult to match one system to the next; plus you have to own the game or benchmark tool we used.

Devil May Cry 4 fixes this, and offers a free benchmark tool available for download. Because the DMC4 MT Framework game engine is rather low-demand for today's cutting edge multi-GPU video cards, Benchmark Reviews uses the DirectX 10 test set at 1920x1200 resolution to test with 8x AA (highest common AA setting available between GeForce and Radeon video cards) and 16x AF. The benchmark runs through four different test scenes, but scenes #2 and #4 usually offer the most graphical challenge.

Devil May Cry 4 doesn't stress the GPU to the extent that other game engines do. This isn't to say that the graphics don't look good, because they do, it's just that the MT Framework game engine is very well optimized. The Juniper GPU inside ATI's Radeon HD 5770 produced 72 FPS, which is extremely close to the 79 FPS rendered by our overclocked Palit GeForce GTX 260 Sonic video card. The ATI Radeon HD 4890 matched the overclocked ASUS GeForce GTX 285 TOP video card performance with 89 FPS. ATI's Cypress GPU found in the Radeon HD 5850 and 5870 certainly stood out from the crowd.

The Sapphire ATI Radeon HD 5850 produced 108 FPS, which equates to 21% better performance than the GeForce GTX 285. Sapphire's ATI Radeon HD 5870 rendered 126 FPS in the second benchmark scene, outperforming the NVIDIA GTX 285 counterpart by nearly 42%. Despite wearing a NVIDIA The Way It's Meant to be Played (TWIMTBP) logo, the Capcom MT Framework game engine appears to enjoy ATI's latest Cypress and Juniper GPUs.

The dual-GPU GeForce GTX matches performance to the single-GPU Radeon HD 5870, yet still falls way behind the twin Cypress-XT (Hemlock) GPUs inside the Radeon HD 5970.

Far Cry 2 Benchmark

Ubisoft has developed Far Cry 2 as a sequel to the original, but with a very different approach to game play and story line. Far Cry 2 features a vast world built on Ubisoft's new game engine called Dunia, meaning "world", "earth" or "living" in Farci. The setting in Far Cry 2 takes place on a fictional Central African landscape, set to a modern day timeline.

The Dunia engine was built specifically for Far Cry 2, by Ubisoft Montreal development team. It delivers realistic semi-destructible environments, special effects such as dynamic fire propagation and storms, real-time night-and-day sun light and moon light cycles, dynamic music system, and non-scripted enemy A.I actions.

The Dunia game engine takes advantage of multi-core processors as well as multiple processors and supports DirectX 9 as well as DirectX 10. Only 2 or 3 percent of the original CryEngine code is re-used, according to Michiel Verheijdt, Senior Product Manager for Ubisoft Netherlands. Additionally, the engine is less hardware-demanding than CryEngine 2, the engine used in Crysis.

However, it should be noted that Crysis delivers greater character and object texture detail, as well as more destructible elements within the environment. For example; trees breaking into many smaller pieces and buildings breaking down to their component panels. Far Cry 2 also supports the amBX technology from Philips. With the proper hardware, this adds effects like vibrations, ambient colored lights, and fans that generate wind effects.

There is a benchmark tool in the PC version of Far Cry 2, which offers an excellent array of settings for performance testing. Benchmark Reviews used the maximum settings allowed for DirectX 10 tests, with the resolution set to 1920x1200. Performance settings were all set to 'Very High', Render Quality was set to 'Ultra High' overall quality, 8x anti-aliasing was applied, and HDR and Bloom were enabled.

Although the Dunia engine in Far Cry 2 is slightly less demanding than CryEngine 2 engine in Crysis, the strain appears to be extremely close. In Crysis we didn't dare to test AA above 4x, whereas we used 8x AA and 'Ultra High' settings in Far Cry 2. The end effect was a separation between what is capable of maximum settings, and what is not. Using the short 'Ranch Small' time demo (which yields the lowest FPS of the three tests available), we noticed that there are very few products capable of producing playable frame rates with the settings all turned up.

Inspecting the performance at 1920x1200 resolution, it appears that every graphics card we tested can handle higher quality settings and post-processing effects in Far Cry 2. ATI's Radeon HD 5770 produces 29.0 FPS, which is extremely close to the 31.4 FPS delivered by the Radeon HD 4890. Palit's GeForce GTX 260 Sonic performed at 36.4 FPS, followed by the ASUS GeForce GTX 285 TOP with 43.0 FPS. The Radeon HD 5850 leads the overclocked GTX 285 by only about one frame, and produced 43.9 FPS on average. The Sapphire ATI Radeon HD 5870 DirectX 11 video card 21161-00-50R topped the single-GPU Far Cry 2 performance chart with 51.3 FPS at 1920x1200, with a 19% lead over NVIDIA's direct competition.

Moving on to dual-GPU comparisons, the NVIDIA GeForce GTX 295 displayed 56 FPS at 1920x1200 and edged out the HD5870's 51.3 FPS. This is fine if you're comparing those two cards (which sell at very different price points), but the ATI Radeon HD 5970 leaps over the GTX295 by almost 37% and produces a 76.6 frame rate under the same load. Perhaps that's The Way It's Meant To Be Played.

Resident Evil 5 Tests

Built upon an advanced version of Capcom's proprietary MT Framework game engine to deliver DirectX 10 graphic detail, Resident Evil 5 offers gamers non-stop action similar to Devil May Cry 4, Lost Planet, and Dead Rising. The MT Framework is an exclusive seventh generation game engine built to be used with games developed for the PlayStation 3 and Xbox 360, and PC ports. MT stands for "Multi-Thread", "Meta Tools" and "Multi-Target". Games using the MT Framework are originally developed on the PC and then ported to the other two console platforms.

On the PC version of Resident Evil 5, both DirectX 9 and DirectX 10 modes are available for Microsoft Windows XP and Vista Operating Systems. Microsoft Windows 7 will play Resident Evil with backwards compatible Direct3D APIs. Resident Evil 5 is branded with the NVIDIA The Way It's Meant to be Played (TWIMTBP) logo, and receives NVIDIA GeForce 3D Vision functionality enhancements.

NVIDIA and Capcom offer the Resident Evil 5 benchmark demo for free download from their website, and Benchmark Reviews encourages visitors to compare their own results to ours. Because the Capcom MT Framework game engine is very well optimized and produces high frame rates, Benchmark Reviews uses the DirectX 10 version of the test at 1920x1200 resolution. Super-High quality settings are configured, with 8x MSAA post processing effects for maximum demand on the GPU. Test scenes from Area #3 and Area #4 require the most graphics processing power, and the results are collected for the chart illustrated below.

Resident Evil 5 has really proved how well the proprietary Capcom MT Framework game engine can look with DirectX 10 effects. The Area 3 and 4 tests are the most graphically demanding from this free downloadable demo benchmark, but the results make it appear that the Area #3 test scene performs better with NVIDIA GeForce products compared to the Area #4 scene that favors ATI Radeon GPUs. Although this benchmark tool is distributed directly from NVIDIA and GeForce Forceware drivers likely have optimizations written for the Resident Evil 5 game, there doesn't appear to be any favoritism towards GeForce products over Radeon counterparts from within the game itself.

Even so, the ATI Radeon HD 5770 rendered 36 FPS in test scene 3, while jumping to 47 FPS in test scene 4. This loosely indicates that lower-end graphics cards can still play Resident Evil 5 at 1920x1200, and produce good 30+ frame rates with maximum settings. For these results however, it seems that driver optimizations between manufacturers could account for the disparity among test scenes, although the Resident Evil 5 game itself 'normalizes' in the two other (less demanding) scenes.

Much of the test results in Resident Evil 5 were identical to performance standings in our other tests. The GTX260 and HD4890 produce the same frame rates, while the HD5850 and GTX285 push and pull between tests. The Radeon HD5870 keeps up with the GTX295, while the Radeon HD 5970 performs ahead of them all with a 38% lead. With a 123 FPS peak frame rate in scene 4, the HD5970 certainly delivers enough performance to guarantee smooth fast motion on even larger display resolutions.

STALKER Call of Pripyat Benchmark

The events of S.T.A.L.K.E.R.: Call of Pripyat unfold shortly after the end of S.T.A.L.K.E.R.: Shadow of Chernobyl. Having discovered about the open path to the Zone center, the government decides to hold a large-scale military "Fairway" operation aimed to take the CNPP under control. According to the operation's plan, the first military group is to conduct an air scouting of the territory to map out the detailed layouts of anomalous fields location. Thereafter, making use of the maps, the main military forces are to be dispatched. Despite thorough preparations, the operation fails. Most of the avant-garde helicopters crash. In order to collect information on reasons behind the operation failure, Ukraine's Security Service send their agent into the Zone center.

S.T.A.L.K.E.R.: CoP is developed on X-Ray game engine v.1.6, and implements several ambient occlusion (AO) techniques including one that AMD has developed. AMD's AO technique is optimized to run on efficiently on Direct3D11 hardware. It has been chosen by a number of games (e.g. BattleForge, HAWX, or the new Aliens vs Predator) for the distinct effect in it adds to the final rendered images. This AO technique is called HDAO which stands for ‘High Definition Ambient Occlusion' because it picks up occlusions from fine details in normal maps.

Put in simple terms, ambient light occlusion can be described as the parts of the scene where light finds it hard to reach. In the real world, light has to bounce off many surfaces in order to reach some places. Classically this problem is solved with a radiosity technique but this is usually too expensive for real-time applications. For this reason, various screen space techniques have been invented to emulate the effect of ambient occlusion.

S.T.A.L.K.E.R. Call of Pripyat is a video game based on the DirectX 11 architecture and designed to use high-definition SSAO. If Benchmark Reviews were to test this game with the developers recommended settings for desired gaming experience, only the ATI Radeon HD 5000-series video cards would be tested. Having the good fortune to experience this free benchmark demo run through all four tests (Day, Night, Rain, Sun Shafts) with the highest settings possible (HDAO mode with Ultra SSAO quality), the Radeon HD5970 produced 30+ FPS (in the Day test) while the GTX295 looked like a slide show with 6 FPS. Needless to say, this wouldn't be the most educational way of testing video cards for our readers. For this reason alone, we reduced quality to DirectX 10 levels, and ran tests with SSAO off and then enabled with Default settings on our collection of higher-end video cards.

Without SSAO and using DX10 lighting, the HD4890 actually comes amazingly close to the Radeon HD 5850 with 42 FPS in the 'Day' test run. The overclocked ASUS GeForce GTX 285 TOP rendered only 38 FPS on DX10 settings, which isn't very convincing argument for NVIDIA's current top-end product. The Radeon HD 5870 performed on-par with the GeForce GTX 295 using DX10 lighting and produced 51 FPS, but then the Radeon HD 5970 comes in and tacks on 42% gain for 75.1 FPS. That might not seem like good news for NVIDIA GeForce products, but it only gets worse.

Even though we've restrained the STALKER CoP tests to a meager DirectX 10 level, the benchmark allows us to add SSAO. There are three SSAO modes: Default, HBAO, and HDAO; this test uses Default. Each mode then has three SSAO levels of detail: Low, Medium, High, and specific to HDAO you can add Ultra. Our tests use the Default-High settings, which rank #3 out of 10 SSAO levels.

Again, the HD4890 (31.6 FPS) is somewhat close to the HD5850 (34.6 FPS), but the GeForce GTX 285 (21.6 FPS) is way below them both. The ATI Radeon HD 5870 flexes its lonely Cypress-XT GPU and squeezes out 41.9 FPS, all while looking down at the dual GT200-GPU NVIDIA GeForce GTX 295 video card. The GTX295 squeaks out 28.8 FPS using both GPUs, but then the ATI Radeon HD 5970 steps in and shows how it's to be done: 62.8 FPS (118% faster).

It seems evident that even DirectX 10 platforms favor the ATI Radeon 5000-series. It also seems that NVIDIA should have choose its words carefully when extolling DirectX 10 and condemning DirectX 11.

Unigine Heaven Benchmark

The Unigine "Heaven" DirectX 11 benchmark is a free publicly available tool that grants the power to unleash the graphics capabilities in Windows 7 or updated Vista Operating Systems. It reveals the enchanting magic of floating islands with a tiny village hidden in the cloudy skies. With the interactive mode, emerging experience of exploring the intricate world is within reach. Through its advanced renderer, Unigine is one of the first to set precedence in showcasing the art assets with tessellation, bringing compelling visual finesse, utilizing the technology to the full extend and exhibiting the possibilities of enriching 3D gaming.

The distinguishing feature in the Unigine Heaven benchmark is a hardware tessellation that is a scalable technology aimed for automatic subdivision of polygons into smaller and finer pieces, so that developers can gain a more detailed look of their games almost free of charge in terms of performance. Thanks to this procedure, the elaboration of the rendered image finally approaches the boundary of veridical visual perception: the virtual reality transcends conjured by your hand. The "Heaven" benchmark excels at providing the following key features:

• Native support of OpenGL, DirectX 9, DirectX 10 and DirectX 11
• Comprehensive use of tessellation technology
• Advanced SSAO (screen-space ambient occlusion)
• Volumetric cumulonimbus clouds generated by a physically accurate algorithm
• Dynamic simulation of changing environment with high physical fidelity
• Interactive experience with fly/walk-through modes
• ATI Eyefinity support

Just like we've already done in BattleForge and S.T.A.L.K.E.R. Call of Pripyat, the Unigine "Heaven" benchmark was reduced to DirectX 10 levels to make sure everyone had a fair chance. The Heaven benchmark is a free demo that makes use of the unigine game engine, and is designed to show off the most detailed cobblestone and smoke you've ever seen a graphics card generate... in DirectX 11.

Last years high-end ATI HD4890 does its best to compete with this year's lower mid-level Radeon graphics solution, yet somehow matches performance with the GeForce GTX 285 when 4x AA is applied (and bests it with 8x AA enabled). The HD5870 produced the same 41 FPS that the GeForce GTX 295 did, but when 4x AA is increased to 8x the results favor the HD5870 by more than 18%. Then comes the ATI Radeon HD 5970, which improves on the GTX295 by 54% with 4x AA and 81% when it's turned up to 8x AA.

While the Unigine Heaven benchmark is a sample for what a game developer could do with their engine, as of now it's merely a synthetic benchmark with the same punctuation as 3dMark Vantage. Nobody plays a benchmark as I frequently say, but it seems evident that everyone can expect to see great things come from a tool this detailed. For now though, those details only come by way of the ATI Radeon 5000-series.

Radeon HD 5970 Temperatures

Benchmark tests are always nice, so long as you care about comparing one product to another. But when you're an overclocker, or merely a hardware enthusiast who likes to tweak things on occasion, there's no substitute for good information. Benchmark Reviews has a very popular guide written on Overclocking Video Cards, which gives detailed instruction on how to tweak a graphics cards for better performance. Of course, not every video card has overclocking head room. Some products run so hot that they can't suffer any higher temperatures than they already do. This is why we measure the operating temperature of the video card products we test.

FurMark does do two things extremely well: drive the thermal output of any graphics processor higher than any other application of video game, and it does so with consistency every time. While I have proved that FurMark is not a true benchmark tool for comparing video cards, it would still work very well to compare one product against itself at different stages. FurMark would be very useful for comparing the same GPU against itself using different drivers or clock speeds, of testing the stability of a GPU as it raises the temperatures higher than any program. But in the end, it's a rather limited tool.

To begin my testing, I use GPU-Z to measure the temperature at idle as reported by the GPU. Next I use FurMark 1.7.0 to generate maximum thermal load and record GPU temperatures at high-power 3D mode. The ambient room temperature remained at a stable 20.0°C throughout testing, while the inner-case temperature hovered around 36°C. The ATI Radeon HD 5970 DirectX 11 video card recorded a chilly 36°C in idle 2D mode (matching the inner case temp), and increased to 88°C in sustained full 3D mode. Oddly enough, these temperatures weren't much different than the reference-design HD5870 at idle and load, proving that the longer (larger) thermal management solution has done a decent job of cooling the Hemlock GPUs.

The Radeon HD 5970 is made of two Cypress-XT GPUs, which each have a massive 334 mm² die size and offer a much larger footprint for cooling the 2.154 billion transistors when they're just sitting idle. So with this in mind, it's understandable to see an impressive low idle temperature. Loaded temperature is noteworthy, but overclockering the HD5970 will push this ever-closer to 100°C. Nevertheless, for a 40nm die the Cypress GPU and makes it clear that the ATI Radeon HD 5970 Hemlock GPU has thermal and power management under close control.

VGA Power Consumption

Life is not as affordable as it used to be, and items such as gasoline, natural gas, and electricity all top the list of resources which have exploded in price over the past few years. Add to this the limit of non-renewable resources compared to current demands, and you can see that the prices are only going to get worse. Planet Earth is needs our help, and needs it badly. With forests becoming barren of vegetation and snow capped poles quickly turning brown, the technology industry has a new attitude towards suddenly becoming "green". I'll spare you the powerful marketing hype that I get from various manufacturers every day, and get right to the point: your computer hasn't been doing much to help save energy... at least up until now.

To measure isolated video card power consumption, Benchmark Reviews uses the Kill-A-Watt EZ (model P4460) power meter made by P3 International. A baseline test is taken without a video card installed inside our computer system, which is allowed to boot into Windows and rest idle at the login screen before power consumption is recorded. Once the baseline reading has been taken, the graphics card is installed and the system is again booted into Windows and left idle at the login screen. Our final loaded power consumption reading is taken with the video card running a stress test using FurMark. Below is a chart with the isolated video card power consumption (not system total) displayed in Watts for each specified test product:

If a reference-design Radeon HD5850/HD5870 video cards each using only about 25W at idle, then the HD5970 should probably rest at about 50W idle, right? ATI claims that because of the revamped PLX ExpressLane CrossFire bridge that enables a new feature called 'CrossFire Ultra-low Power State', one GPU is shut down and placed into a suspended power state when the video card doesn't require the graphics processing power. AMD tested the idle power draw at around 42W, however our own tests indicate that the Radeon HD 5970 sips closer to 48W at idle. Although this isn't single-GPU efficiency, it's much better than the 74W consumed at idle by NVIDIA's GTX295.

Once 3D-applications begin to demand power from the GPU, electrical power consumption really begins to climb. Under full 3D load the ATI Radeon HD 5970 requires 299W, which is still better than the 302W consumed by the GeForce GTX 295 dual-GPU video card. Put this into a frames-per-watt formula using BattleForge, the ATI Radeon HD 5970 costs 5.35W per frame while the NVIDIA GeForce GTX is costing 14.43W per each frame rendered. In a long-lasting MMORPG gaming session with a title like BattleForge, you can see how the electrical overhead costs will eat away at your budget.

Radeon 5000-Series Final Thoughts

As of 23 September 2009. AMD can claim that the ATI Cypress GPU in their Radeon HD 5870 video card can achieve 2.72 TeraFLOPS; making it more powerful than any other known microprocessor. Armed with 1600 shader cores, the 40nm Cypress-XT GPU claims to push video game frame rates well-beyond what NVIDIA offers from their GeForce GTX 285 counterpart. Putting two Cyrpress-XT GPUs together inside the Radeon HD 5970 has helped to further punctuate this lead over the competing GeForce graphics line.

ATI's next-generation Radeon HD 5800 graphics cards are also the world's first and only GPU to fully support Microsoft DirectX 11, the new gaming and compute standard that ships with the Microsoft Windows 7 operating system and is now support on Windows Vista. With the ATI Radeon HD 5800 series of graphics cards, PC users can expand their computing experience with ATI Eyefinity multi-display technology, accelerate their computing experience with ATI Stream technology, and dominate the competition with superior gaming performance and full support of Microsoft DirectX 11, making it a "must-have" consumer purchase just in time for launch of Microsoft's Windows 7 Operating System.

Reading the editorial articles surrounding the launch of ATI's Radeon 5800-series video cards has become entertainment in and of itself. Websites loyal to NVIDIA assert that the Cypress GPU is nothing more than an overpowered product trying to push DirectX-11 onto unwilling consumers, or that NVIDIA is doing more for gamers than ATI because they offer The Way It's Meant to be Played, GeForce 3D Vision, PhysX, and CUDA. Some websites have even taken the time to research the amount of progress AMD has had with their Stream technology, and then complain that ATI isn't doing enough to compete with NVIDIA in regard to GPGPU. All of this rhetoric amounts to a desperate attempt at hiding some very frightening facts.

NVIDIA was extremely vocal when Windows Vista launched with DirectX-10, and they couldn't over-emphasize how important Vista/DirectX-10 was going to be to gamers and that enthusiast should upgrade to their recently announced DirectX 10-compliant GeForce video card series. Oddly enough gamers didn't take to Windows Vista like NVIDIA had hoped, and as Windows 7 launched there was still a 52.6% market share using Windows XP compared to 36.4% using Vista (with 10% of the market already using a beta version of Windows 7). The DirectX-11 Direct3D API is native to the Windows 7 Operating System, a product Microsoft is releasing, not AMD. ATI has simply prepared for the launch of Windows 7 better than NVIDIA, and now the green machine claims nobody will buy a video card for DirectX-11.

From these developments ATI has distanced themselves ahead of NVIDIA by placing gamers first in their consideration, and have positioned the ATI 5000-series to introduce enthusiasts to a new world of DirectX 11 video games on the Microsoft Windows 7 Operating System. While most hardware enthusiasts are familiar with the back-and-forth competition between these two leading GPU chip makers, it might come as a surprise that NVIDIA actually states that DirectX-11 video games won't fuel video card sales, and have instead decided to revolutionize the military with CUDA technology. This kind of sentiment doesn't make sense, especially when you consider the work they've put into the NVIDIA GF100 GPU Fermi Graphics Architecture.

AMD has launched the Radeon 5870 as their first assault on their multi-monitor ATI Eyefinity Technology feature, using native HDMI 1.3 output paired with DisplayPort connectivity. The new Cypress GPU features the latest ATI Stream Technology, which is designed to utilize DirectCompute 5.0 and OpenCL code. These new features improve all graphical aspects of the end-user experience, such as faster multimedia transcode times and better GPGPU compute performance. AMD has already introduced a DirectCompute partnership with CyberLink, and the recent Open Physics Initiative with Pixelux promises to offer physics middleware built around OpenCL and Bullet Physics. This looks like ATI's recipe for success, since NVIDIA does won't have their GF100 GPU inside GeForce products until March 2010 to compete against the Radeon 5800 series or support DirectX-11. It doesn't help matters any that existing NVIDIA GPUs do not support OpenCL and DirectCompute 11 environments, leaving them out in the cold through these winter months.

Any GeForce DirectX-11 graphics solution is still at least two months away for NVIDIA and not expected until March 2010 at the soonest, which leaves very few options in the fiercely competitive discrete graphics market until then. So far NVIDIA's only counter-attack on ATI's new 5000-series product line has been the ultra low-end GeForce 200 (no letter designation) and GeForce GT 220 series meant to one-up integrated graphics. Integrated graphics? You read that correctly, NVIDIA launched a product so feeble that it competes with older Integrated graphics. Outstanding. Maybe you can enable triple-SLI and get GeForce GTS 250 performance out of them for a good game of Solitaire? This ulitimately increases the demand for Fermi to succeed in the desktop sector.

So what can NVIDIA do to compete with ATI until Fermi makes and appearance? Since DX11 is dominated by ATI until GT100 launches, it would seem that GeForce price reductions should be in order. Just not yet, apparently. It's not clear what NVIDIA is waiting for, but the price of their current GeForce family hasn't changed much since the ATI Radeon 5870/5850 launch. ATI's Eyefinity Technology is another tough nut to crack for NVIDIA, since their DirectX-10 products only offer dual-DVI output. As an alternative, NVIDIA GeForce owners can use the $300 Matrox TripleHead2Go add-on peripheral to spread the picture across up to three screens. Just make sure you're using a GeForce GTX-275 of higher, since the added resolution is more than the video card was designed to accommodate.

ATI Radeon HD 5970 Conclusion

Although our rating and final score are made to be as objective as possible, please be advised that every author perceives these factors differently at different points in time. While we do our best to ensure that all aspects of the product are considered, there are often times unforeseen market conditions and manufacturer changes which occur after publication that would render our rating obsolete. Please do not base your purchases solely on our conclusion, as it represents our product rating at the time of publication. Benchmark Reviews begins our conclusion with a short summary for each of the areas that we rate.

The first section rated in our conclusion is performance, which considers how effective the Sapphire ATI Radeon HD 5970 DirectX 11 video card performs with its "Hemlock" GPU in designated test operations against direct competitor products. Considering that the GeForce GTX 295 sells at the $500 price segment and is NVIDIA's most powerful desktop graphics product, the ATI Radeon HD 5970 has set the bar so high that it could be several months before any other product comes close. In each of our benchmarks we've ensured that these two products were tested head-to-head, and in terms of DirectX 10 performance the Radeon HD 5970 offered approximately 40% better performance in NVIDIA TWIMTBP video games (Crysis Warhead, Devil May Cry 4, Far Cry 2, and Resident Evil 5). Looking at games with SSAO available to them, the Radeon HD5970 leaped ahead of the GTX295 more than 60% most times, and sometimes as much as 200% depending on the game and resolution. These are all the DirectX 10 tests, because when DirectX 11 is utilized only the ATI Radeon 5000-series is able to participate. In terms of power efficiency, the ATI Radeon HD 5970 is far more conservative and offered 5.35W per each frame rendered compared to the GTX295 with 14.43W per frame. For the time being and at least through to the end of the year, ATI's Radeon HD 5970 is the most powerful desktop graphics product available, and the only brand of graphics card capable of playing DirectX 11 games.

Product appearance is relative to personal tastes, but the entire Radeon 5000-series looks very similar to one another with the only difference being overall length and some minor changes to the vent grill. Lately it seems that almost everything has been encased in a plastic housings with a label fixed to the top, so I'm rather used to the basic style lines. In the end though, ATI strikes a comfortable blend of elegance and flair that seems appealing to the senses.

Construction is solid, and unlike ATI had done to the 5800-series cards the HD5970 receives a full-length grill for venting out all heated air from the case. Gamers and overclockers alike should appreciate it when the video card exhaust outside of the computer case, because it affects everything else inside of it. Every inch of the Hemlock graphics card is covered, save for some small knockouts on the back for the two GPUs. Considering that serious overclocking to the Cypress-XT GPUs could result in a fast-approaching 100°C loaded temperature, ATI was within right to extend the housing to lengthy 12.1" and open the vent.

While most consumer buy a discrete graphics card for the sole purpose of PC video games, there's a very small niche who expect extra features beyond basic frame rates. AMD isn't the market leader in GPGPU functionality, but their ATI Stream Technology is the only one designed to utilize DirectCompute 5.0 and OpenCL code. ATI Eyefinity technology is impressive, as it demonstrates yet another dimension of visual experience that the competition cannot offer. But for most consumers, it's the added connectivity that really counts: native DisplayPort and HDMI interfaces. As a whole, the ATI Radeon HD 5890 video card offers more raw functionality than we can find from competing GeForce products.

As of February 2010, supplies of the Radeon HD 5970 'Hemlock' video card seem very scarce. NewEgg offers listings of the HD5970 from Sapphire, Diamond, PowerColor, and ASUS; and they're all priced around $679. The single-GPU Radeon HD 5870 still sells at NewEgg for $409.99 (and similar elsewhere) which makes it possible for hardware-savvy gamers who might consider buying two individual cards and putting them in a CrossFire set for similar performance and price.

In summary, there's a long future ahead for the entire Radeon HD 5000 series; and overclocking enthusiasts will be just as pleased with the Radeon HD 5970 as gamers will be, DirectX 11 is the future of gaming, whether the competition likes it or not, and ATI has a huge head-start on absorbing an early market share. Eyefinity is a nice touch and it certainly adds to the gaming experience, but there's such an incredibly small portion of potential users for the technology that shear graphics power is a better sales pitch. The ATI Radeon HD 5970 is perfect for high-performance gamers interested in unparalleled frame rates, and multi-monitor power users wanting triple-panel diversity. Overclockers can push this 'unlocked' Hemlock video card well beyond it's specifications, mainly because that's what it was designed for. The HD5970 is the 'Black Edition' of ATI's Radeon series, and the closest competing product is more than 40% behind.

AMD Warranty Statement: AMD's product warranty does not cover damages caused by overclocking, even when overclocking in enabled via AMD software. Unlike previous generations of ATI Radeon graphics cards, the overclocking potential of the ATI Radeon HD 5970 graphics card has been left uncapped. You can control the amount of overclocked to which you subject your graphics card. Please use discretion and caution when overclocking this video card.

Pros:

+ Fastest DirectX 11 compatible graphics accelerator available
+ Consumes 48W at idle and 299W under full load
+ Eyefinity Technology through DisplayPort and DVI (x2)
+ Outstanding performance for ultra high-end games
+ Fan exhausts all heated air outside of case
+ 3200 GPU cores (total) at 725 MHz
+ 2 GB of 1000 MHz 256-bit GDDR5 vRAM
+ Highly overclockable Hemlock GPUs
+ Very-quiet cooling fan under loaded operation
+ Supports CrossFireX functionality

Cons:

- 12.1" Long graphics card may not fit some cases
- Maximum post-processing Anti Aliasing is limited to 8x
- Very expensive premium-level product

Ratings:

• Performance: 10.0
• Appearance: 9.00
• Construction: 9.75
• Functionality: 9.50
• Value: 6.75

Final Score: 9.0 out of 10.

Excellence Achievement: Benchmark Reviews Golden Tachometer Award.

Questions? Comments? Benchmark Reviews really wants your feedback. We invite you to leave your remarks in our Discussion Forum.

Related Articles:

• G.Skill ECO 4GB DDR3 Memory Kit F310666CL7D
• HIS Radeon HD6950 IceQ-X Turbo-X Video Card
• Enermax Aurora and Caesar Scissor-switch Keyboards
• ASUS RT-N66U Dual Band Wireless-N900 Gigabit Router
• ASUS ENGTX560 Ti DCII TOP Video Card
• Seagate Cheetah 15K.7 SAS Hard Drive ST3600057SS
• D-Link DIR-655 Xtreme N Gigabit Router
• ASUS GeForce GTX 660Ti DirectCU-II TOP
• ASUS P6T Deluxe OC Palm Edition X58 Motherboard
• 120GB OCZ Vertex 3.20 Solid State Drive