AMD Radeon HD 7950 Launch Review

Author: Chris Ledenican
Editor: Howard Ha
Publish Date: Monday, January 30th, 2012
Originally Published on Neoseeker (http://www.neoseeker.com)
Article Link: http://www.neoseeker.com/Articles/Hardware/Reviews/AMD_HD_7950/
Copyright Neo Era Media, Inc. - please do not redistribute or use for commercial purposes.

Just a month ago AMD released the fastest single GPU solution available, the Radeon HD 7970 utilizing AMD's Generation Core next (GCN) architecture which sported the first 28nm GPU on the market, and also the first to support DirectX 11.1. Today we are continuing our look at the Northern Islands graphics cards with the next iteration in the series, the HD 7950.

Like the HD 7970, the HD 7950 is based on AMD’s GCN architecture and is also built on a 28nm node. However, the HD 7950 comes with less streaming processors, slightly reduced clock speeds and requires less power. In total, the HD 7950 comes packed with 1792 stream processors, 32 raster units, 112 texture units and a GPU engine clock speed of 800MHz. Additionally, the HD 7950 has a 3GB memory buffer with a default clock speed of 1250MHz which runs on a 384-bit bus, giving the card a total memory bandwidth of 240GB/s.

While these specifications are slightly reduced in comparison to the HD 7970, they are still extremely robust. Just from the specifications alone, we can see the HD 7950 is potentially on track to becoming the second fastest single GPU graphics card on the market. Since the HD 7950 is the little brother of the HD 7970, the price tag is also reduced; this model is coming into the channel with a $449 USD price tag, making it $100 cheaper than the HD 7970.

Over the last week the Neoseeker lab has been inundated with reference and aftermarket HD 7950 graphics cards. Being the hardware fanatics we are, we decided to not only put up a reference review, but also a three part review featuring the reference model and AIB models. We'll even have CrossFire and TriFire reviews up very soon! Be sure to check them all out to get the clearest picture of the how AMD's latest 28nm part stacks up from all angles.

AMD's Graphics Core Next is a scalable architecture designed to be optimized for both graphics horsepower and compute power. The GPU is compiled of 32 Compute Units, dual Geometry engines, 8 Render Back-Ends, 768KBs of read/write L2 cache, and a 384-bit memory bus all running on a PCI Express 3.0 x16 bus interface and packing in a total 4.81 billion transistors. Like the HD 6900 series, the dual geometry engines each have their own Rasterizer and Render Back-End units, but share the 32 Compute Units. You can basically think of these Geometry engines as having dual cores inside the GPU.

As mentioned earlier, the Southern Islands GPU is designed to work as both a graphics and compute engine. This is due to the ACE engine which posits three devices in one GPU that are completely asynchronous from one other: the graphics pipeline, the direct compute pipeline and two parallel pipelines inside the GPU. Each of these run independently and asynchronously to the primary graphics pipeline. This allows the GPU to process an intensive compute simultaneously with the graphics pipeline as it is running an intensive 3D application, and maximizes the utilization of the available 3.71TFLOPS of compute power. Paired with this is the processing core and dual DMA engines which allow the HD 7970 to saturate the Gen 3.0 PCI-Express interface with a TeraFlop of bidirectional double precision.

On top of this, all of the GDDR5 memory is protected by single error correction and double error detection when used in a compute environment. All the internal SRAM has the same capabilities, and also have ECC data protection.

As many are wondering exactly how the efficiency of CGN compares to VLIW4, AMD has stated that we should see an more performance per millimeter with the GCN architecture than what was available in previous generations. According to Eric Demers from AMD, this should represent a peak physical improvement of up to 7 to 7.5 times over previous generation architecture like that used in the Radeon HD 6970. The Tahiti architecture also includes an improved Gen 9 tessellator unit that increases the tessellation throughput via more efficient vertex re-use, larger parameter caches and improved off-chip buffering. All of this gives the GCN architecture up to 4 times the throughput compared to the Radeon HD 6900 series.

The basic compute unit includes all the instruction, wavefront and scheduling; essentially this unit can be thought of as its own core. Each compute unit has four sub-units that run a way-front of 64-bit vector lanes over four cycles that are completely independent of each other. This is a huge departure from the VLIW engine, which had 5 or 4 math units that executed the individual instructions in a parallel process. Since the new GCN architecture runs fully scalar, it eliminates the compilation issues of the VLIW design.

In addition, the scalar runs in parallel to the unit and can issue instructions of its own, allowing complex operations to be moved to the scalar for improved efficiency. The unit also includes a 16KB L1 cache that has both read and write functions allowing the textures to run through the cache, as opposed to entering the cache to be processed and then passed on to the back-end before being exported back into the cache. This function can now be handled exclusively though the L1 cache via the read/write functions.

Moving over to this form of computing doesn't necessarily translate into improved graphics performance, but since the VLIW compiling was not necessary efficient, it will improve the total compute power, i.e. parallel processing of the GPU. The compute unit on the other hand has four independent wavefronts running in parallel, as well as a scalar programing model at the lane level to ensure that all instructions running through the GPU automatically work. This eliminates all port conflicts, simplifying the compiler and instructions, thus improving the compute performance dramatically.

When it comes to the memory, each L1 cache has 64 bytes of bandwidth per clock and the HD 7970 has a total of 32, giving it up to 2TB/s of bandwidth. As mentioned before, each L1 cache has both read and write functions, something new to the Southern Islands architecture. However, each L1 still has a L2 cache to fall back on, but the L2 also now includes both read and write functions. The HD 7970 comes packed with twelve L2 caches that feed back into the L1 cache which are also 64 bytes per clock. In total this gives the HD 7970 around 710GB/s of cache bandwidth, with the default GPU clock speed of 925MHz for both reads and writes.

The GPU also includes a 16KB instruction cache and 32KB scalar data cache that are shared per four compute units, which as mentioned earlier are also backed by the L2 cache. Additionally, each compute unit has has its own registers and local data share. There is also a global data share unit that works as a manage buffer on the chip to allow sharing between any wavefront on the chip.

The Southern Islands architecture also includes a texture mapping technique called Partially Resident Texture, or PRT. Essentially what this feature do is take advantage of all the memory hardware available, and turn the local frame buffer into a local texture cache. So, what does this mean? First, the local graphics memory can behave like a hardware-managed cache where texture data can be streamed in on demand. This prevents stuttering as the pages are brought in, and texture stream has the ability to handle the process more efficiently.

PRT also Improves the memory efficiency and image quality with very large, detailed textures. This allows for texture sizes up to 32 TB (16k x 16k x 8k x 128-bit), done by turning the textures into 64KB chunks that are dynamically selected and loaded into the memory as needed. So, essentially the textures that are not going to be displayed are not loaded. PRT also translates through the page table's every request. The data is rendered if it is available, and if not the application can manage the textures instead. This allows it to dynamically opt to use lower resolution bitmaps for the lower resolution which will make the textures slightly blurry, but there will be no lag time in these situations.

Like the HD 6900 series, the Southern Islands graphics cards come equipped with AMD PowerTune technology. PowerTune is basically a means to set a predefined TDP by adjusting the clock speeds in real time. The way in which PowerTune is utilized is very different than the on-board regulation chips used on NVIDIA’s GTX 500 series. NVIDIA’s power management system monitors the power coming from the rails, while AMD’s technology instead relies on performance counters that are embedded throughout the GPU. These performance counters have an internal algorithm that dynamically calculate how much power is being used, and adjust accordingly. This allows PowerTune to maintain the power draw at the predefined level, effectively eliminating huge surges in power from occurring. Since games operate at a lower peak power rate than benchmarking applications such as Kombuster, in-game performance will not be negativity affected.

AMD has also introduced a new feature called "Zero Core Power" which maximizes the idle power consumption of the board. When a discrete GPU is in a static screen state, it works to minimize idle power by enabling a host of active power saving functions including (but not limited to); clock gating, power gating, memory compression, and a host of other features. However, GPUs with AMD’s exclusive ZeroCore Power technology can take energy savings to entirely new heights by completely powering down the core GPU while the rest of the system remains active.

Along with the changes to the architecture, AMD is also introducing Eyefinity 2.0. For the most part, the changes are being made at the driver level, but there is one new feature being added to the Southern Islands graphics cards like the HD 7970.

This new feature gives the HD 7970 the ability to simultaneously output multiple, independent audio streams. Essentially this means each video source that has the ability to support audio will have its own dedicated audio signal. This allows a single HD 7970 to connect to multiple displays, each having its own audio signal. In total the graphics card can support up to five audio signals. You can be fragging people on one monitor while watching your favorite show on another, also connected to the Southern Islands graphics card. The technology also follows the video so if the source changes, the audio seamlessly switches to the other device as well. This is actually an interesting feature that really pushes the expansion options of the Radeon series forward.

The next feature is one that we have been hoping would come along for some time now, the merging of Eyefinity and HD3D. Unlike the previous feature, having a Southern Islands graphics card is not required to run Eyefinity in 3D. Instead this is a simple driver update that will enable the feature for all graphics cards that already support both technologies. Stereoscopic 3D technology from both AMD and NVIDIA is still niche at best, but we are glad AMD is moving forward and taking Eyefinty to its next logical step.

Another driver fix is the addition of flexible bezel compensation. The image below should give you a good idea of what this is all about. Essentially, it allows anyone to pair three non-identical monitors together and not have to worry about the images not lining up. Instead, the user can adjust the displays and still have the on screen image align perfectly across the displays.

AMD has also added a task bar positioning feature. Anyone that uses Eyefinity knows that previous drivers pushed the main desktop display to the leftmost screen. With this new positioning feature, the user can now pick which display the task bar is set to. Again, this is a huge improvement over the previous generation Eyefinity, and will make using this technology more convenient as the main desktop can now be manually configured to fit the individual needs of any Eyefinity user. 

Eyefinity 2.0 also includes a custom resolution feature allowing the display resolution to be manually set to best fit the users' needs. While most gamers will be happy simply setting the resolution to 5760x1080, there are a handful of people that prefer even greater customization. The new Eyefinity 2.0 adds support for 5x1 Landscape with 1920x1200 and 2560x1600 monitors. This means Eyefinity is no longer limited to monitors at or below 1080p, increasing the available display real estate even further.

The "Tahiti Pro" is the second GPU released in the Southern Islands series, and like the Tahiti XT it is designed for extreme performance. Within the core AMD has packed roughly 4.31 billion transistors into a 28nm node with a die size of 365mm² . The core features 1792 streaming processors, 32 ROPS and 112 texture units. Along with the 800MHz clock speed, this gives the card a total compute power of 2.87 TFLOPs. Additionally, the HD 7950 has the same 384-bit, 3GB GDDR5 fame buffer that the HD 7970 has, with the only exception being the clock speed on this model is set at 1250MHz (5Gbps QDR) instead of 1375MHz (5.5Gbps QDR).

AMD has carried over the same physical attributes and stylings used on the HD 7970, so visually there is very little that separates this model from the HD 7970. This means the HD 7950 posses the same red and black color scheme and uses a rear mounted blower style fan. Along with the aesthetics, the HD 7950 also includes the same rounded design of the back end of the shroud, designed to improve ventilation when the graphics card is being used in CrossFireX configuration. As far as the dimensions go, the reference HD 7950 is actually the same length as the HD 7970, making it 10.5" long.

At the rear are two CrossFireX connectors and a PCIe Generation 3.0 x16 lane.  By using the PCIe 3.0 interface, the board has double the maximum data rate over Gen 2.0, giving the card up to 32 GB/s of bi-directional bandwidth on an x16 connector. It is going to be hard for a single graphics card to saturate the PCIe Gen 3 interface with so much bandwidth, so the benefit will most likely only be noticeable with scaling multiple graphics cards together in CrossFireX configuration.

Like the 6900 series, the HD 7950 comes equipped with PowerTune technology. PowerTune is basically a power management system that maximizes the performance of the board via dynamic power adjustment. It does this by increasing the GPU clock speed in real time when the GPU detects power headroom, and throttling the clocks when a certain power limit is exceeded. This allows the board to adjust the clock speeds on a microsecond level. The maximum PowerTune rating for the HD 7950 is around 200W at load. Since this is less than the HD 7970, the HD 7950 swaps out the 8+6 pin power configuration for just two 6-pin connectors.

The HD 7950 also supports "Zero Core Power". Traditionally, anyone using multiple GPUs in a single system had to deal with a high power idle state, simply because each card was still actively drawing system power; each graphics card could produce 30+ watts of power even when the system wasn't under load. With "Zero Core Power", the extra graphics cards in a CrossFireX system are disabled, shutting down the fans and capping any voltage from going to the core. Since PowerTune works on a microsecond level, "Zero Core Power" will not interfere with gaming as all the GPUs can become active again in just microseconds.

The HD 7950 uses the same video output configuration was saw on the HD 7970. In total there are two Mini-DP connectors, a single HDMI 1.4a connector and a Dual-Link DVI connector. The HD 7950 also uses the same non-stacked DVI design as the HD 7970, which improves both the acoustics and exhaust rate giving the air a clearer path to travel out of the heatsink. Also, again just like the HD 7970, the HD 7950 will come bundled with a HDMI to DVI dongle, and mini-DP to DVI dongle that allow the card to support up to three DVI connections out of the box.

With the AMD HD 7950 AMD is continuing their robust support for multiple displays. In all they have included a HDMI port that uses the latest 1.2 standard which allows them to support up to three monitors per port (via MST Hub) as well as AMD's HD3D technology. The middle HDMI 1.4a connector also supports 3GHz speeds with frame packing. Essentially this allows the connection to run the frames faster, which is going to make viewing images and playing games smoother across the board. The HDMI and DP ports can also be paired together to support HD3D Surround.

The HD 7950 uses AMD's sixth generation vapor chamber design, which includes a multi-step vapor chamber technology. Essentaillyl what AMD has done is give the heatsink  two steps with three discrete levels. One of the levels comes down and sits directly on top the GPU, while another sits over the board and the third is located toward the back-end of the PCB. Additionally, AMD uses a second generation design for their phase changing thermal paste to improve the thermal performance, and according to AMD the TIM alone can affect temperatures by as much as 2°C to 3°C.

The HD 7950 also includes the same blower style fan seen on the HD 7970, but like its older sibling the fan utilizes optimized fan blade technology. Essentially what AMD has done is increase the size and width of the blades, allowing the fan to push higher CFM with a lower decibel rating. While the design is better than previous generations it can still produce a fair amount of noise at high RPM. However, more often than not the RPM level is low, so the fans acoustics are low as well.

The fan runs at 1.7A @ 12V DC and has a dual ball bearing design, but unfortunately we still have not been sent the exact CFM or dBA rating for the fan. The fan uses a 4-pin PWM design, allowing it to automatically adjust to the needs of the GPU, and like the other card in the series the stacked DVI connector has been removed for better ventilation and lower turbulence.

The HD 7950 is built on the same PCB as the HD 7970, but it boasts a few tweaks to compensate for the lower power consumption. The PCB has a 5-phase power design, all solid Japanese capacitors, and a CHL8228G voltage controller from the CHiL Semiconductor Corporation. The CHL8228G is a dual-loop digital multi-phase controller that can drive up to 8 phase units, and features Input Voltage Management to allow up to 3 input voltages to be monitored. This will ensure the card is adequately powered, and improves the overall power efficiency. Additionally, the HD 7950 has dual 6-pin power connectors, which along with the PCIe power slot provides the board with up to 225W of power.

Tahiti Pro owners will also get the dual BIOS toggle switch that was first introduced with the Cayman architecture. Essentially this switch allows the user to toggle back and forth between a protected and unprotected BIOS. By default, the switch is at the protected setting which runs the card at the default settings of the manufacturer. However, when the switch is set in the "unprotected" position, the BIOS can be flashed and the settings will be internally stored. This means the graphics card can boot with either the default or custom BIOS depending on the position of the switch. It is best to think of this feature as insurance for the card. If you flash the BIOS and the card runs into an issue, the BIOS can be reverted to the default settings simply by toggling the switch.

Overclocking:

Since the HD 7970 scaled above 1100MHz, we were expecting its little brother to put up a similar showing, albeit slightly lower due to the 225W current limitation. To overclock the GPU and memory, we used the latest version of MSI Afterburner which supports voltage control for 7000 series graphics cards with a CHL8228 voltage regulator. Since the HD 7950 does indeed have a CHL8228 regulator, we were able to increase the voltage and push the core to the BIOS cap of 1100MHz.

To remain stable at this frequency, we increased the voltage to just over 1200mV, but even while 1100MHz did require some more juice, we were able to easily hit 1025MHz at the default voltage setting. The memory also scaled to the maximum level, capping out at 1575 (6300MHz effective). The final clock speeds we reached gave us a 18.2% increase in the GPU clock speed, while we able to pump up the memory by a beefy 20.5%, which bumped up the memory bandwidth to 304.2GB/s.

Hardware Configuration:

Drivers:

Benchmarks DX11:

Test Settings:

Usage:

Specifications:

(Note: All models might not be included in this review. The table below is to be used for comparison purposes)
AMD Specifications
Model
AMD Radeon HD 7970 AMD Radeon HD 7950
AMD Radeon HD 5870 AMD Radeon HD 6950 AMD Radeon HD 6970
Processing Cores
2048 1792 1600 1408 1536
Core Clock
925MHz 850MHz 850MHz 800MHz 880MHz
Memory Clock
1375MHz 1250MHz 1200MHz 1250MHz 1375MHz
Memory Interface
384-bit 384-bit 256-bit 256-bit 256-bit
Memory Type
3GB GDDR5 3GB GDDR5 1GB GDDR5 2GB GDDR5 2GB GDDR5
Fabrication Process
28nm 28nm 40nm 40nm 40nm
NVIDIA Specifications
Model
Nvidia GTX 460 Nvidia GTX 470 Nvidia GTX 480 Nvidia GTX 570 Nvidia GTX 580
Processing Cores
336 448 480 480 512
Core Clock
675MHz 607MHz 700MHz 742MHz 782MHz
Memory Clock
1100MHz 837MHz 924MHz 1250MHz 1002MHz
Memory Interface
256-bit 128-bit 320-bit 320-bit 384-bit
Memory Type
1GB GDDR5 1.25GB GDDR5 1.5GB GDDR5 1.25GB GDDR5 1.5GB GDDR5
Fabrication Process
40nm 40nm 40nm 40nm 40nm

Futuremark's latest 3DMark 2011 is designed for testing DirectX 11 hardware running on Windows 7 and Windows Vista. The benchmark includes six all new benchmark tests that make extensive use of all the new DirectX 11 features including tessellation, compute shaders and multi-threading.

In our first synthetic benchmark, the HD 7950 is significantly faster than the GTX 570 in the two highest presets and faster than the GTX580 when we benchmarked at the Extreme setting. From the results, we can see the HD 7950 should run on par with the GTX 580 in lower resolutions, and outpace it when the resolution is increased. In comparison to the GTX 570 though, the HD 7950 is going to beat it outright 100% of the time.

The results between the stock and overclocked benchmarks were quite striking. First off the HD 7950 jumped from a score of 9683 at 800Mhz to 12000, but more impressive is just how close the score was to that of the GTX 590 at 1100MHz. However, it is hard to gauge just where the overclocked HD 7950 will end up from a synthetic test, so for now lets not put too much stock in the overall increase.

Unigine Heaven became very popular very fast, because it was one of the first major DirectX 11 benchmarks. It makes great use of tessellation to create a visually stunning heaven.

The HD 7950 performed better than the stock GTX 580 in Ungine 2.5 at both the 1680x1050 and 1920x1080 resolutions, and tied with the GTX 580 at 2560x1600. So, just like the last benchmark, Ungine demonstrates how the HD 7950's closest competitor is not the GTX 570, but instead the GTX 590.

Just like the last synthetic benchmark, Unigine 2.5 scaled extremely well when the HD 7950 was overclocked to 1100MHz.

Aliens vs Predator is a DX11 Benchmark that runs though a scene straight out of the classic 80’s movie, Aliens. Since it uses DX11, it can often be more than a graphics card can handle.

The HD 7950 really had a good showing in our first real-world benchmark. The HD 7950 performed around 11% faster than the GTX 580 and a whopping 29% faster than the GTX 570 at the stock speeds. Once we overclocked the GPU, the HD 7950 managed to achieve a higher frame rate than the GTX 590 across the board.

Batman: Arkham City is the sequel to the smash hit, Batman: Arkham Asylum. The game was created with the Unreal 3 Engine, and includes areas with extreme tessellation, high res textures and dynamic lighting. Batman, also includes native support for PhysX and is also optimized for Nvidia 3DVision technology.

The reference HD 7950 was again faster than the GTX 580, but this time by just a few frames-per-second. Averaging it out, the HD 7950 was around 4% faster than the GTX 580, and 16% faster than the GTX 570. Once again, overclocking significantly increased the performance of the HD 7950 to the point where it rivaled the GTX 590.

Battlefield 3 is designed to deliver unmatched visual quality by including large scale environments, massive destruction, dynamic shadows. Additionally, BF 3 also includes character animation via ANT technology, which is also being utilized in the EA Sports franchise. All of this is definitely going to push any system its threshold, and is the reason so many gamers around the world are currently asking if their current system is up to the task.

Unlike the other games we benchmark, the performance of Battlefield 3 is tested during online game play. We ensure our results are accurate by running through each resolution four times before averaging the results.

Up to this point the HD 7950 was performing somewhere between 11% and 4% faster than the GTX 580, but at 800MHz it proved to be slightly slower than the GTX 580 in Battlefield 3. Compared to the GTX 580, the HD 7950 was between 3.5% and 10% slower, so in this benchmark the results were in favor of the GTX 580. However, the HD 7950 still easily outperformed the GTX 570 and HD 6970.

Crysis 2 is a first-person shooter developed by Crytek and is built on the CryEngine 3 engine. While the game was lacking in graphical fidelity upon its release, Crytek has since added feature such as D11 and high quality textures. This improved the in-game visuals substantially, which in turn pushes even high-end hardware to the max.

So far the results for the HD 7950 are almost a mirror-image of what we saw when we were testing the HD 7970. Some games show incredible scaling, which allowed both Tahiti based graphics cards to leave the Fermi architecture in the dust, while others simply underperformed. Sadly this is the case in Crysis 2, where the HD 7950 is only slightly faster than the HD 6950, and considerably slower than the GTX 580.

We believe this is a driver issue, as every new architecture has some teething issues when they are released. We fully expect these results to improve as AMD releases better optimized drivers for the Southern Islands series of graphics cards.

DiRT 3 is the third installment in the DiRT series and like it's predecessor incorporates DX11 features such as tessellation, accelerated high definition ambient occlusion and Full Floating point high dynamic range lighting. This makes it a perfect game to test the latest DX11 hardware.

The HD 7950 again fell slightly behind the GTX 580, but continues to display performance beyond the capabilities of the GTX 570 and HD 6970. In this benchmark the 7950 was between 3% and 5% faster than the GTX 570, and 20% faster than the HD 6950.

Metro 2033 puts you right in the middle of post apocalyptic Moscow, battling Mutants, rivals and ratio-active fallout. The game is very graphics intensive and utilizes DX11 technology, making it a good measure of how the latest generation of graphics cards perform under the latest standard.

Both the Southern Islands graphics cards we have tested thus far have shown excellent performance in Metro 2033. In this benchmark the HD 7950 managed to perform up to 21% better than the GTX 580 at the highest resolution, and 15% better at the two lower resolution settings. Metro 2033 is a benchmark that runs best when the graphics card has substantial memory bandwidth, so while the HD 7950ès results are impressive they were also expected.

Total War: Shogun 2 is a game that creates a unique game-play experience by combining both real-time and turn-based strategy. The game is set in 16th-century feudal Japan and gives the player control of a warlord battling various rival factions. Total War: Shogun 2 is the first in the series to feature DX11 technologies to enhance the look of the game, but with massive on-screen battles it can stress even the highest-end graphics cards.

The HD 7950 scores another slim win over the GTX 580, and we can more or less say the two cards performed evenly in this benchmark. Breaking it down, the HD 7950 was once again able to perform better than the HD 6950 by an average of 18%, while it performed around 20% better than the GTX 570.

The overclocked results of the HD 7950 continue to kick ass. This is yet another benchmark where the overclocked HD 7950 was able to perform at either the same range as, slightly better than, the GTX 590. That makes five out of seven real-world benchmarks were the single GPU HD 7950 was able to keep up with the dual GPU GTX 590.

Temperature:

To measure core GPU temperatures, we run three in-game benchmarks and record the idle and load temperature according to the min and max temperature readings recorded by MSI Afterburner. The games we test are Crysis 2, Lost Planet 2 and Metro 2033. We run these benchmarks for 15 minutes each. This way we can give the included thermal solution and GPU time to reach equilibrium.

From a thermal standpoint we can see that AMD really took the Tahiti Pro GPU cooling seriously. When running the HD 7950 at the default settings, the temperature only reached 71°C. This put the higher end HD 7950 below even mid-range graphics cards such as the HD 6850. Since the GPU was able to remain nice and cool during our testing, the fan RPM level stayed consistently low even though we were benchmarking a card using the reference thermal solution.

Power Consumption:

To measure power usage, a Kill A Watt P4400 power meter was used. Note that the numbers represent the power drain for the entire benchmarking system, not just the video cards themselves. For the 'idle' readings we measured the power drain from the desktop, with no applications running; for the 'load' situation, we took the sustained peak power drain readings after running the system through the same in-game benchmarks we used for the temperature testing. This way we are recording real-world power usage, as opposed to pushing a product to it's thermal threshold.

When it came to performance-per-watt, the HD 7950 really delivered. At the default settings the idle and load temperatures were a sparse 104W and 276W, respectively. This made it more efficient than any of the mid-range or high-end Fermi graphics cards, and even put it near the level of the HD 6870 in terms of overall power consumption. However, after we increased the voltage to maintain stability at 1100MHz, the power rating increased to 345 watts. Even while there was a noticeable bump, the overclocked HD 7950 was still more efficient than the stock GTX 580.

Throughout the review we compared the HD 7950 to the HD 7970 on multiple occasions. The reason behind this is because both models are based on the same core architecture, but the HD 7950 has slightly lower specifications so it can fit into a more affordable pricing segment. To be sure, at $449 USD the HD 7950 is still one of the more expensive graphics cards on the market, but rightly so. In our labs, the AMD HD 7950 was able to outperform the similarly priced GTX 580 in just over half of our real-world benchmarks when paired with a 2600K and running at the base clock speed of 800MHz. This isn’t a remarkable feat in and of itself, but when you take into account the improved power and thermal efficiency of the HD 7950, it has a leg up over the GTX 580 and helps make it all that more attractive once you get into the $449 USD price range, for the time being anyway.

In addition to the impressive stock performance, the HD 7950 was designed for overclocking. The card AMD sent us easily scaled to 1100MHz after some adjustments were made to the stock voltage. With the GPU at this frequency, the HD 7950 was able to push more pixels than the stock HD 7970 and at times even competed aggressively with the dual core HD 6990 and GTX 590. Additionally, even with the core overclocked the total power consumption remained lower than that of the stock GTX 580. So again the HD 7950 displayed excellent performance relative to the amount of power consumed.

When testing the HD 7970, our biggest complaint was with the drivers; the performance fluctuated up or down depending on the benchmark, making it hard to gauge the exact performance difference between the HD 7970 and the GTX 580. The drivers appear to have been ironed out a bit this time around with the HD 7950, but once again the performance seesawed back and forth across the benchmarks. Breaking the performance differences down into percentages, we can see the HD 7950 is around 15 to 20% faster than the GTX 570, and just about even with the GTX 580. On the other side of the coin, the HD 7950 is around 15% faster than the HD 6970, 25 to 30% faster than the HD 6950 and roughly 20% slower than the HD 7970. Overall, the performance increase over the last generation is good, but the 500 series Fermi architecture remains highly competitive, even after being on the market for over a year.

Overall the HD 7950 is an excellent graphics card, and we expect the performance to increase as the drivers mature. When looking at the whole package, the HD 7950 offers superb gaming power, overclocking headroom and performance-per-watt, making it the best graphics card currently available with an MSRP of $449 USD. Still, the HD 7950 is expensive, but just like the HD 7970 the price is going to stick until NVIDIA can counter with their upcoming Kepler architecture. So, for the time being the ball is definitely in AMD's court.

»Neoseeker.com

Copyright Neo Era Media, Inc., 1999-2014.
All Rights Reserved.

Please do not redistribute or use this article in whole, or in part, for commercial purposes.