NVIDIA GeForce GTX 660 vs NVIDIA GeForce GTX 670

Compare NVIDIA GeForce GTX 660 3 GB vs NVIDIA GeForce GTX 670 4 GB, specs and GPU benchmark score. Which is the better graphics card for the money?

Benchmark Score
Other Benchmarks
Specifications
Comments (0)
Popular Comparisons

GPU Comparison

Quickly search and compare graphics cards

Benchmark Score

Overall Score

General gaming and workstation score

816 points

31% slightly better overall score

621 points

Flux Core frame rate

Volumetric ray casting test, a computationally expensive method of rendering high-quality scenes

18 FPS

30 FPS

12 FPS significantly higher Flux Core frame rate

Electron frame rate

Randomly generated noise sphere test

11 FPS

18 FPS

7 FPS significantly higher Electron frame rate

City frame rate

Procedurally generated city scene with voxel rendering

13 FPS

18 FPS

6 FPS significantly higher City frame rate

Clouds frame rate

Real-time noise calculation and ray marching test

9 FPS

14 FPS

4 FPS significantly higher Clouds frame rate

Free GPU Benchmark

Want to compare your graphics card against the GeForce GTX 660 and the GeForce GTX 670? Download our free and quick PC Performance Test.

Download GPU Benchmark

Other Benchmarks

3DMark Graphics score

High-end graphics benchmark

1317 points

1841 points

39% slightly higher 3DMark Graphics score

Blender score

Cycles Render (Samples per minute)

122.56 points

184.30 points

50% significantly higher Blender score

Geekbench 5 (CUDA) score

CUDA compute benchmark

8583 points

11969 points

39% slightly higher Geekbench 5 (CUDA) score

Geekbench 5 (Metal) score

Metal compute benchmark

5449 points

7421 points

36% slightly higher Geekbench 5 (Metal) score

Geekbench 5 (OpenCL) score

OpenCL compute benchmark

11381 points

14780 points

29% slightly higher Geekbench 5 (OpenCL) score

Geekbench 5 (Vulkan) score

Vulkan compute benchmark

10693 points

14358 points

34% slightly higher Geekbench 5 (Vulkan) score

Geekbench 6 (Metal) score

Metal compute benchmark

4311 points

6454 points

49% significantly higher Geekbench 6 (Metal) score

Geekbench 6 (OpenCL) score

OpenCL compute benchmark

11353 points

15314 points

34% slightly higher Geekbench 6 (OpenCL) score

Geekbench 6 (Vulkan) score

Vulkan compute benchmark

11858 points

16269 points

37% slightly higher Geekbench 6 (Vulkan) score

Specifications

Core Clock Speed

Nominal chip frequency

980 MHz

7% slightly higher core clock speed

915 MHz

Boost Clock Speed

Higher turbo frequency when power conditions are met

1032 MHz

5% slightly higher boost clock speed

980 MHz

Memory

Onboard memory size for textures and vertices

3 GiB

4 GiB

33% slightly more memory

Memory Bus Width

Number of parallel lines to the memory chips

192 Bit

256 Bit

33% slightly larger memory bus width

Memory Bandwidth

Data transfer speed between GPU core and memory

144.2 GB/s

192.3 GB/s

33% slightly higher memory bandwidth

TDP

Thermal Design Power: Measure of heat generated by the GPU

140 W

21% slightly lower TDP

170 W

Pixel Rate

Number of pixels that can be rendered per second

20.64 Gigapixels/s

27.44 Gigapixels/s

32% slightly higher pixel rate

Texture Rate

Number of textured pixels that can be rendered per second

82.56 Gigatexels/s

109.8 Gigatexels/s

32% slightly higher texture rate

Floating Point Performance

Raw number of floating point operations per second

1.98 TFLOPS

2.63 TFLOPS

32% slightly better floating point performance

Shading Units

Number of processors dedicated to shader processing

960

1344

40% slightly more shading units

Texture Mapping Units

Number of processors dedicated to applying textures

112

40% slightly more texture mapping units

Render Output Processors

Number of processors dedicated to final pixel rendering

33% slightly more render output processors

Other details

Rank

Ranking in the hardwareDB database

277th of 586

306th of 586

Memory Type

The type of memory used by this chip

GDDR5

DirectX Support

Maximum version of DirectX supported

11.0

OpenGL Support

Maximum version of OpenGL supported

4.5

GeForce GTX 660 vs GeForce GTX 670 comparison

In our benchmarks, the GeForce GTX 660 beats the GeForce GTX 670 in gaming performance.

Our database shows that the GeForce GTX 660 has a slightly higher core clock speed. The core clock speed (or base speed) is the frequency at which the GPU core runs. This metric makes sense when comparing GPUs of a similar architecture or generation. In addition, the GeForce GTX 660 also has a slightly higher boost clock speed. The boost clock speed is the frequency that the GPU core can reach if the temperature is low enough. The allows for higher performance in certain scenarios.

As stated by the manufacturer, the GeForce GTX 670 has slightly more memory with 4 GiB of memory compared to 3 GiB. If you're planning at playing games at high resolutions with high-quality textures, you will need a lot of memory.

In addition, the GeForce GTX 660 has a slightly lower TDP at 140 W when compared to the GeForce GTX 670 at 170 W. Heat output doesn't match power consumption directly but, it's a good estimate.

According to the results of the hardwareDB benchmark utility, the GeForce GTX 660 is faster than the GeForce GTX 670.