Logo   Banner   TopRight
TopUnder
Transparent
NVIDIA GeForce GTX 660 Round-Up: MSI, ZOTAC, GB
Transparent
Date: Sep 13, 2012
Section:Graphics/Sound
Author: Marco Chiappetta
Transparent
Introduction and Specifications

In the past few months, we have taken a look at no less than four new high-end GPUs from NVIDIA based on the company’s Kepler microarchitecture. NVIDIA kicked things off with the GeForce GTX 680 back in May and followed up a few weeks later with the monstrous, dual-GPU powered GeForce GTX 690, which remains the fastest graphics card in existence. A few days after that, the GeForce GTX 670 hit the scene and last month marked the arrival of the GeForce GTX 660 Ti. While all of these GeForce GTX 600 series cards differ in their delivered performance levels, they do have something major in common--they are all based on the same GK104 GPU.

Today, NVIDIA is announcing two more GeForce GTX 600 series graphics cards, but unlike their older brothers, the new GeForce GTX 660 and GeForce GTX 650 are based on two different GPUs, the GK106 and GK107, respectively. The GK106 and GK107 are both based on Kepler, but they are scaled down in comparison to the more powerful GK104. They offer the same feature set (save for Boost technology on the GK107), but have fewer cores, texture units, and ROPs, among other things, and simply perform at a lower performance level as a result. As such, they are also more affordable, as you might expect.

Below are the main features and specifications of the “non-Ti” GK106-based GeForce GTX 660, followed by some more detail on the GPU itself. Take a gander at the high-level stuff and then we’ll move on to the GeForce GTX 650 and all of the performance data you’re probably itching to see…


NVIDIA GeForce GTX 660 Reference Card.

NVIDIA GeForce GTX 660
Specifications & Features
Processing Units
Graphics Processing Clusters 3
SMXs 3
CUDA Cores 960
Texture Units 80
ROP Units 24
Clock Speeds
Base Clock 980 MHz
Boost Clock 1033 MHz
Memory Clock (Data Rate) 6008 MHz
L2 Cache Size 384KB
Memory
Total Video Memory 2048MB
Memory Interface 192-bit
Total Memory Bandwidth 144.2 GB/s
Texture Filtering Rate (Bilinear) 78.4 GigaTexels/sec
Physical & Thermal
Fabrication Process 28 nm
Transistor Count 2.54 Billion
Connectors 2 x Dual-Link DVI, 1 x mini-DP, 1 x HDMI
Form Factor Dual Slot
Power Connectors 1 x 6-pin
Recommended Power Supply 450 watts
Thermal Design Power (TDP) 140 watts
Thermal Threshold 98° C

Although the GK106 features all of the same technology as NVIDIA’s more powerful Kepler-based graphics processors, this new GPU is somewhat smaller and scaled-down versus its higher-end counterparts.


GK106: GeForce GTX 660 GPU Block Diagram

Posted above is a high-level block diagram of the GK106 GPU powering the GeForce GTX 660. The actual chips are manufactured using TSMC’s 28nm process node and are comprised of approximately 2.54B transistors. The GPU features three Graphics Processing Clusters (GPC) with three SMXs and a total of 960 CUDA cores. There are also 80 texture units and 24 ROPs within the GPU, along with 384K of L2 cache. Memory is linked to the GPU via a 192-bit memory interface. In comparison to the GeForce GTX 660 Ti, the new 660 has a similar memory interface and cache configuration, but the 660 has fewer CUDA cores, texture units, and ROPs, which results in lower compute performance and fillrate, but similar memory bandwidth.

That block diagram sure looks a little funky though, doesn't it? NVIDIA mentioned during our briefing that the 960 CUDA cores exposed on the GK106 are the chip's full complement, but if that's truly the case, one of the GPCs is half the size of the others. Obviously, it's possible NVIDIA designed the chip that way from the start. However, there could also be 192 additional CUDA cores laying dormant on the chip for one reason or another.

  
More Pics of the NVIDIA GeForce GTX 660 Reference Card.

NVIDIA’s reference specifications call for a base GPU clock of 980MHz on the GeForce GTX 660, with a boost clock of 1033MHz and memory clock of 1502MHz (6008Mhz effective). At those frequencies, GeForce GTX 660 card will offer up to 144.2GB/s of memory bandwidth and 78.4GTexes/s of textured fillrate. Many of NVIDIA’s partners, however, are ready with factory overclocked models (which we’ll show you on the next page), that will offer somewhat higher performance characteristics.

The GeForce GTX 660 has a TDP of 140 watts and requires a single, supplemental 6-pin PCI express power feed. Cards are two-slot wide, but sport a short PCB, like the GeForce GTX 660 Ti and many GeForce GTX 670 cards. The output configuration is the same 2 x Dual-Link DVI, 1 x mini-DP, and 1 x HDMI setup as the GeForce GTX 660 Ti as well.
 

Transparent
GeForce GTX 660 Cards: GB, MSI, Zotac

For the purposes of this article, we got our hands on a trio of retail-ready GeForce GTX 660 cards from MSI, Gigabyte, and Zotac, in addition to the reference card we showed you on the previous page. First up, here’s a look at the MSI GeForce GTX 660 Twin Frozr III cooler



  
MSI GeForce GTX 660 Twin Frozr III

Although it is overclocked and sports a fully custom design, the MSI GeForce GTX 660 Twin Frozr III's cooler is the standout feature on this card. The dual-slot Twin Frozr III sports high-density heatsinks, with dual 80mm cooling fans, and thick copper heatpipes that run from the cooler’s base up through the heatsink fins. We should also point out that the cooler’s base is made of pure copper and the entire assembly is nickel-plated. The Twin Frozr III’s cooler has also been outfitted with custom “propeller blade” fans that reportedly push 20% more air than previous designs.

The customizations on the MSI GeForce GTX 660 TwinFrozr III don’t stop at the cooler. MSI has also overclocked the card, with 1033MHz (base) and 1098MHz (boost) clocks for the GPU. The memory runs at the same speed as reference models, however, at 1502MHz (6008MHz effective). The MSI GeForce GTX 660 also has a custom PWM that can handle more current than reference designs and it’s outfitted with all solid caps and supper ferrite chokes. Supplemental power is handled by a single 6-pin connector and the outputs on the card consist of a pair of DVI outputs and single DP and HDMI outputs.

Bundled with the MSI GeForce GTX 660 Twin Frozr III are a quick installation guide and user’s manual, a PCI Express 6-pin power adapter, a DVI to VGA adapter, and of course a driver / utility disc. Also available for the card is MSI’s excellent Afterburner performing tuning and monitoring tool, which is available for download right from MSI’s website and allows for additional overclocking and voltage adjustments.



  
Gigabyte GeForce GTX 660 OC Version with Windforce Cooler

Next up we have the Gigabyte GeForce GTX 660 OC Version with Windforce cooling. As is the case with MSI’s offering, although Gigabyte’s card is also overclocked, the real attraction here is the Windforce cooler. Underneath a pair of oversized fans sits an array of aluminum heatsink fins, linked to a copper base via multiple 6mm, copper heat-pipes. The cooler’s dual fans blows air directly onto the heatsinks, where some is diverted into the case and some exhausted outside through the vents in the case bracket. The base of the cooler is shaped in such a way that it reportedly helps minimize turbulence and better direct the airflow through the fins. As you’ll see a little later, the Windforce cooler also does an excellent job of keeping temperatures in check, and it’s nice and quiet too.

The Gigabyte GeForce GTX 660 OC with Windforce cooler ships with its 2GB of memory clocked at the same 6008MHz (effective data rate) of reference cards, but with a base GPU clock of 1033MHz and a boost clock of 1098MHz. Outputs on the card are the same as the reference version as well (dual DVI, HDMI, DP), and Gigabyte’s offering requires the same 6-pin supplemental power connectors of stock GTX 660 cards as well.

Included with the Gigabyte GeForce GTX 660 OC with Windforce cooler were a quick installation guide, driver / utility CD, and a power adapter.



  
ZOTAC GeForce GTX 660 AMP! Edition

Which brings up to the ZOTAC GeForce GTX 660 AMP! Edition. Like the Ti model we liked so much a few weeks back, the “non Ti” GTX 660 from ZOTAC sports a short PCB, which makes the card look somewhat petite.

Although the ZOTAC GeForce GTX 660 AMP! Edition actually utilizes a PCB similar in size to NVIDIA’s reference design, its cooler doesn’t protrude past the end of the card, which shortens the overall length. The ZOTAC GeForce GTX 660 AMP! Edition instead has a compact array of heatsink fins that sit just above the card’s components. The cooler is comprised of large aluminum heatsinks, linked together by an array of thick copper heat-pipes. Two large fans sit above the heatsinks in an angular shroud, blowing air down on the heatsinks and PCB. As you’ll see a little later, the cooler does its job fairly well and it’s relatively quiet too.

Like the other cards featured here, ZOTAC has done some factory overclocking too. The ZOTAC GeForce GTX 660 AMP! Edition’s base and boost GPU frequencies are 993MHz and 1059MHz, respectively, but its memory is identical to NVIDIA’s reference specs and the other cards featured here. With those frequencies, the ZOTAC GeForce GTX 660 Ti AMP! Edition should trail the other cards slightly, but we’re talking about relatively miniscule differences in clock speeds here. We should also note that the ZOTAC card's got the same display output configuration of the others as well.

ZOTAC did go a step further than the other manufacturers featured here with their 660’s bundle, however. Included with the ZOTAC GeForce GTX 660 AMP! Edition, we found a user’s manual and quick installation guide, a coupon for the game TrackMania 2 Canyon, a driver utility disc, a ZOTAC case badge, a peripheral to 6-pin adapter and a DVI to VGA adapter.
 

Transparent
GeForce GTX 650 Details

We didn’t get our hands on a card for testing, but should mention that NVIDIA is also releasing the GeForce GTX 650 today. The GeForce GTX 650, however, is not based on the same GPU as the GeForce GTX 660. Although both are based on NVIDIA’s Kepler microarchitecture, the GTX 650 sports a smaller, more pared down GPU. Whereas the GeForce GTX 660 is built around the GK106 GPU, the GeForce GTX 650 is built around GK107.


NVIDIA's GeForce GTX 650 Reference Card

NVIDIA GeForce GTX 650
Specifications & Features
Processing Units
Graphics Processing Clusters 1
SMXs 2
CUDA Cores 384
Texture Units 32
ROP Units 16
Clock Speeds
Base Clock 1058 MHz
Boost Clock N/A
Memory Clock (Data Rate) 5000 MHz
L2 Cache Size 256KB
Memory
Total Video Memory 1024 or 2048MB
Memory Interface 128-bit
Total Memory Bandwidth 80 GB/s
Texture Filtering Rate (Bilinear) 33.9 GigaTexels/sec
Physical & Thermal
Fabrication Process 28 nm
Transistor Count 1.3 Billion
Connectors 2 x Dual-Link DVI, 1 x mini-HDMI
Form Factor Dual Slot
Power Connectors 1 x 6-pin
Recommended Power Supply 400 watts
Thermal Design Power (TDP) 64 watts
Thermal Threshold 98° C

Fundamentally the GK107 GPU is similar to the more powerful GK106 and GK104 used on more powerful cards like the GeForce GTX 670 / 680. The GK107 at the heart of the GeForce GTX 650, however, has been scaled back and features fewer CUDA cores, texture units, ROPs, L2, and memory partitions.


GK107: GeForce GTX 650 GPU Block Diagram

The GK107 GPU is comprised of roughly 1.3B transistors and is manufactured using TSMC’s 28nm process. The GPU features a single GPC (Graphics Processing Cluster) with 2 SMXs and a total of 384 CUDA cores. There are 32 texture units and 16 ROPs in the GPU, and it features 256K of L2 cache and a 128-bit memory interface.

The GK107 does not support NVIDIA’s GPU boost technology, so there is only a single GPU clock to report. Reference GTX 650’s will have a GPU frequency of 1058MHz with 1250MHz (5000MHz effective) memory. At those clocks, the GeForce GTX 650 offers up to 33.9GTexels/s of textured fillrate with 80GB/s of peak memory bandwidth. Needless to say, performance of this card won't be in the same league as the GTX 660 or more powerful cards, but the GTX 650 should make for a decent, entry-level DX11-class GPU.

GeForce GTX 650 cards require only a single 6-pin supplemental power connector and have a TDP of only 64W. Outputs on the card consist of 2 x Dual-Link DVI and 1 x mini-HDMI.
 

Transparent
Test System and Unigine Heaven v3

How We Configured Our Test Systems: We tested the graphics cards in this article on an Asus P9X79 Deluxe motherboard powered by a Core i7-3960X six-core processor and 16GB of G.SKILL DDR3-1866 RAM. The first thing we did when configuring the test system was enter the system UEFI and set all values to their "optimized" or "high performance" default settings and disabled any integrated peripherals that wouldn't be put to use. The memory's X.M.P. profile was enabled to ensure better-than-stock performance and the hard drive was then formatted and Windows 7 Ultimate x64 was installed. When the installation was complete we fully updated the OS and installed the latest DirectX redist, along with the drivers, games, and benchmark tools necessary to complete our tests.

HotHardware's Test System
Intel Core i7 Powered

Hardware Used:
Intel Core i7-3960X
(3.3GHz, Six-Core)
Asus P9X79 Deluxe
(Intel X79 Express)

Radeon HD 7950 (w/ Boost)
Radeon HD 7870 GHz Edition
Radeon HD 7850
GeForce GTX 670
GeForce GTX 660 Ti
GeForce GTX 560 Ti 448 Core 
MSI GTX 660 Power Edition
Zotac GeForce GTX 660 AMP!
Gigabyte GTX 660 Windforce OC

16GB GSKILL DDR3-1866
Western Digital Raptor 150GB
Integrated Audio
Integrated Network

Relevant Software:
Windows 7 Ultimate x64
DirectX April 2011 Redist
ATI Catalyst v12.7B/12.8
NVIDIA GeForce Drivers v304.48/v306.23

Benchmarks Used:

Unigine Heaven v3
3DMark 11
Batman: Arkham City
Just Cause 2
Alien vs. Predator
Metro 2033
Lost Planet 2
Dirt: Showdown

Unigine Heaven v3.0 Benchmark
Pseudo-DirectX 11 Gaming


Unigine Heaven v3.0

Unigine's Heaven Benchmark v3.0 is built around the Unigine game engine. Unigine is a cross-platform, real-time 3D engine, with support for DirectX 9, DirectX 10, DirectX 11 and OpenGL. The Heaven benchmark--when run in DX11 mode--also makes comprehensive use of tessellation technology and advanced SSAO (screen-space ambient occlusion) It also features volumetric cumulonimbus clouds generated by a physically accurate algorithm and a dynamic sky with light scattering.

The new GeForce GTX 660 cards put up some good numbers in the Unigine Heaven benchmark. As expected, whether factory overclocked or not, the GeForce GTX 660 cards trailed their higher-end NVIDIA-based counterparts, but they competed very well with AMD's offerings. The GTX 660s just missed the mark set by the "stock" Radeon HD 7950 and sandwiched the Radeon HD 7870 depending on their clock speeds.
 

Transparent
3DMark 11 Performance

Futuremark 3DMark11
Synthetic DirectX Gaming


Futuremark 3DMark11

The latest version of Futuremark's synthetic 3D gaming benchmark, 3DMark11, is specifically bound to Windows Vista and WIndows 7-based systems due to its DirectX 11 requirement, which isn't available on previous versions of Windows. 3DMark11 isn't simply a port of 3DMark Vantage to DirectX 11, though. With this latest version of the benchmark, Futuremark has incorporated four new graphics tests, a physics tests, and a new combined test. We tested the graphics cards here with 3DMark11's Extreme preset option, which uses a resolution of 1920x1080 with 4x anti-aliasing and 16x anisotropic filtering.

Futuremark's 3DMark11 tells a very similar story to Unigine Heaven on the previous page. Here, the GeForce GTX 660 cards perform a few percentage points behind the GeForce GTX 660 Ti, but are nipping right at the stock Radeon HD 7950's heals. All of the GeForce GTX 660 cards came out in front of the Radeon HD 7870, however.
 

Transparent
Lost Planet 2, Just Cause 2 Performance

Lost Planet 2
DirectX 11 Gaming Performance


Lost Planet 2

A follow-up to Capcom’s Lost Planet : Extreme Condition, Lost Planet 2 is a third person shooter that takes place again on E.D.N. III ten years after the story line of the first title. We ran the game’s DX11 mode which makes heavy use of DX11 Tessellation and Displacement mapping and soft shadows. There are also areas of the game that make use of DX11 DirectCompute for things like wave simulation in areas with water. This is one game engine that looks significantly different in DX11 mode when you compare certain environmental elements and character rendering in its DX9 mode versus DX11. We used the Test B option built into the benchmark tool with all graphics options set to their High Quality values.

As is typically the case for NVIDIA-powered cards, the GeForce GTX 660s also performed well in Lost Planet 2. Here, the higher-clocked GeForce GTX 660 cards were able to overtake the stock Radeon HD 7950 and all of the cards were clearly faster than the Radeon HD 7870.

Just Cause 2
DX10.1 Gaming Performance


Just Cause 2

Just Cause 2 was released in March '10, from developers Avalanche Studios and Eidos Interactive. The game makes use of the Avalanche Engine 2.0, an updated version of the similarly named original. It is set on the fictional island of Panau in southeast Asia, and you play the role of Rico Rodriquez. We benchmarked the graphics cards in this article using one of the built-in demo runs called Concrete Jungle. The test results shown here were run at various resolutions and settings. This game also supports a few CUDA-enabled features, but they were left disabled to keep the playing field level, since AMD's cards can't use them.

The trend we has seen in the first few benchmarks returned in Just Case 2. In this game, the new GeForce GTX 660 cards landed somewhere in between the stock Radeon HD 7950 and Radeon HD 7870, but only a few frames per second separated them all.
 

Transparent
Metro 2033 Performance

Metro 2033
DirecX11 Gaming Performance


Metro 2033

Metro 2033 is your basic post-apocalyptic first person shooter game with a few rather unconventional twists. Unlike most FPS titles, there is no health meter to measure your level of ailment, but rather you’re left to deal with life, or lack thereof, more akin to the real world with blood spatter on your visor and your heart rate and respiration level as indicators. The game is loosely based on a novel by Russian Author Dmitry Glukhovsky. Metro 2003 boasts some of the best 3D visuals on the PC platform and includes a DX11 rendering mode that makes use of advanced depth of field effects and character model tessellation for increased realism. This title also supports NVIDIA PhysX technology for impressive in-game physics effects. We tested the game at resolutions of 1920x1200 and 2560x1600 with adaptive anti-aliasing and in-game image quality options set to their High Quality mode, with DOF effects disabled.

By this point it should be clear which market segment NVIDIA was aiming for with the GeForce GTX 660. As was the case in most of our previous tests, the GeForce GTX 660 cards finished just behind the Radeon HD 7950, but ahead of the Radeon HD 7870 in the Metro 2033 benchmark.
 

Transparent
Batman: Arkham City Performance

Batman: Arkham City
DirectX Gaming Performance


Batman: Arkham City

Batman: Arkham City is a sequel to 2009’s Game of the Year winning Batman: Arkham Asylum. This recently released sequel, however, lives up to and even surpasses the original in many ways. The story takes place 18 months after the original game. Quincy Sharp, the onetime administrator of Arkham Asylum, has become mayor and convinced Gotham to create "Arkham City" by walling off the worst, most crime-ridden areas of the city and turning the area into a giant open-air prison. The game has DirectX 9 and 11 rendering paths, with support for tessellation, multi-view soft shadows, and ambient occlusion. We tested in DX11 mode with all in-game graphical options set to their maximum values, at various resolutions.

Although the actual framerate scores are difference, the performance trend in Batman: Arkham City mirrors many of our other tests. The new GeForce GTX 660 cards once again outpace the Radeon HD 7870, but hey can't quite catch the stock Radeon HD 7950, let alone the faster / sort-of-newer Radeon HD 7950 w/ Boost.
 

Transparent
Dirt: Showdown Performance

Dirt: Showdown
DirectX 11 Gaming Performance


Dirt: Showdown

Dirt Showdown is the latest in a string of great racing games from Codemasters. Like is predecessor, 2011's Dirt 3, this game sports impressive visuals with DX11 support. “Ultra” settings for shadow effects, global illumination, tessellation, and post processing elements, like depth of field, are available in the game, and in turn, crank up the workload on the graphics subsystem. The game engine also makes use of multi-core processors for higher performance on top-end systems. We tested the game configured with its Ultra graphics options with 4X anti-aliasing enabled at resolutions of 1920x1200 and 2560x1600.

We test Dirt: Showdown with all of its graphics-related options cranked up to their respective maximum values. One feature in particular--Global Illumination, which enhances lighting significantly--runs much faster on the AMD-powered cards. As such, the GeForce GTX 660 card, regardless of their clock speeds, trail every other card in this game, including the Radeon HD 7850.
 

Transparent
Alien vs. Predator Performance

Alien vs. Predator
DirectX 11 Gaming Performance


Alien vs. Predator

The Alien vs. Predator benchmark makes use of the advanced Tessellation, screen space ambient occlusion and high-quality shadow features, available with DirectX 11. In addition to enabling all of the aforementioned DirectX 11 related features offered by this benchmark, we also switched on 4X anti-aliasing along with 16X anisotropic filtering to more heavily tax the graphics cards being tested.

The Alien vs. Predator benchmark is heavily dependant on available memory bandwidth, and in this regard, the Radeons have an advantage. The new GeForce GTX 660 cards are able to outrun the Radeon HD 7850, but the more powerful 7870 and 7950 come out ahead.
 

Transparent
Power Consumption, Noise, Temps

Before bringing this article to a close, we'd like to cover a few final data points--namely power consumption, temperatures and noise. Throughout all of our benchmarking and testing, we monitored acoustics and tracked how much power our test system was consuming using a power meter. Our goal was to give you an idea as to how much power each configuration used while idling and also while under a heavy workload. Please keep in mind that we were testing total system power consumption at the outlet here, not just the power being drawn by the graphics cards alone.

Total System Power Consumption
Tested at the Outlet

Our power consumption numbers mirror many of the benchmark results from the previous pages. As you would expect, the higher clocked GeForce GTX 660 cards consume somewhat more more than the reference card under load conditions. While idle though, the differences are negligeble. Versus competing cards, like we've seen throughout, the GeForce GTX 660 sandwiches the Radeon HD 7870.


With relatively low power consumption characteristics, it should come as no surprise that all of the GeForce GTX 660 cards we tested were quiet and ran relatively cool. All of the cards proved to be quiet, as we never saw their fan speeds spin up past 50%. During moderate loads, these cards do produce a bit of noise, but they'll typically be drowned out by other components. As for their temperatures, the Gigabyte and MSI coolers clearly did the best job keeping temps low, but even the reference cooler is adequate.
 
Transparent
Our Summary and Conclusion

Performance Summary: We don’t have any performance numbers to discuss in regard to the GeForce GTX 650 unfortunately, but looking back through our numbers, summarizing the more powerful GeForce GTX 660’s performance is fairly straightforward. With a few exceptions, the GeForce GTX 660 performed somewhere in between the Radeon HD 7870 GHz Edition and stock, original Radeon HD 7950. Of course, as expected, the GeForce GTX 660 also finished a few percentage points behind the more powerful GeForce GTX 660 Ti across the board.

There is one other aspect of performance we want to quickly discuss as well. During many product briefings, NIVIDA likes to set the stage for a new GPU launch by discussing the expected performance improvements it will offer over previous generations of graphics cards. The GeForce GTX 660, for example, is targeting roughly the same market segment as the GeForce 9800 GT that launched a few years ago and the GeForce GTX 460, which landed in 2010. Current data suggests that many gamers are still using DX10-class products, which aren’t nearly as powerful as the current-generation of cards.

If you’ve got a GeForce 9800 GT or GTX 460-class graphics card and have been holding off on an upgrade, our testing suggests that you can expect a 2X – 4X performance improvement by moving to a card like the new GeForce GTX 660. In addition to increased performance, however, newer cards also offer other benefits, like lower-power, lower-noise (usually), and additional features like full DX11 support. In any case, if you were wondering how much faster a GTX 660 is than an older card like a 9800 GT or GTX 460, now you know.

 

The GeForce GTX 660 and GTX 650 should be available immediately from your favorite retailers at prices hovering around $229 and $109, respectively. We can't say how the GTX 650 fits in the current landscape without testing one, but at its expected price point, the GeForce GTX 660 is looking very strong. As our testing has shown, the GeForce GTX 660 is faster than the Radeon HD 7870 GHz Edition overall and it's also priced a few dollars lower than AMD's offerings, which is a clear win for NVIDIA. AMD and its partners are offering a free copy of Sleeping Dogs with select Radeon purchases right now, but ultimately the GeForce GTX 660 means another round of price cuts for the Radeon HD 7800 series are due.

Above is a chart comparing the specific GeForce GTX 660 cards from MSI, Gigabyte, and ZOTAC we tested. With only a few MHz separating the highest and lowes clocked cards, the end result are a group of cards that ultimately offer very similar performance. The coolers on the MSI and Gigabyte cards, however, offered lower temperature than ZOTAC's. But the ZOTAC card offers a somewhat smaller form factor and also includes a game. Considering they'll all be priced between $229 and $239, it's really a toss up. They're all solid offerings.

In the end, the new GeForce GTX 660 is another strong product in its category, from NVIDIA. The GPU offers great performance in its segment, it's relatively quiet, power consumption is in-line and the GeForce GTX 660 is priced competitively.  You can't ask for much more than that.


NVIDIA GeForce GTX 660
  • Good Performer
  • Competitively Priced
  • Relatively Power Friendly
  • None



Content Property of HotHardware.com