NVIDIA GeForce GTX 660 vs NVIDIA GeForce GTX 650
We compared two Desktop platform GPUs: 2GB VRAM GeForce GTX 660 and 1024MB VRAM GeForce GTX 650 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA GeForce GTX 660 's Advantages
Boost Clock1032MHz
More VRAM (2GB vs 1024GB)
Larger VRAM bandwidth (144.2GB/s vs 80.00GB/s)
576 additional rendering cores
NVIDIA GeForce GTX 650 's Advantages
Lower TDP (65W vs 140W)
Score
Benchmark
FP32 (float)
GeForce GTX 660
+143%
1.981 TFLOPS
GeForce GTX 650
0.812 TFLOPS
Blender
GeForce GTX 660
+136%
111
GeForce GTX 650
47
OctaneBench
GeForce GTX 660
+107%
27
GeForce GTX 650
13
Graphics Card
Sep 2012
Release Date
Sep 2012
GeForce 600
Generation
GeForce 600
Desktop
Type
Desktop
PCIe 3.0 x16
Bus Interface
PCIe 3.0 x16
Clock Speeds
980 MHz
Base Clock
-
1032 MHz
Boost Clock
-
1502 MHz
Memory Clock
1250 MHz
Memory
2GB
Memory Size
1024MB
GDDR5
Memory Type
GDDR5
192bit
Memory Bus
128bit
144.2GB/s
Bandwidth
80.00GB/s
Render Config
-
-
-
-
-
-
960
Shading Units
384
80
TMUs
32
24
ROPs
16
-
-
-
-
-
-
16 KB (per SMX)
L1 Cache
16 KB (per SMX)
384 KB
L2 Cache
256 KB
-
-
-
Theoretical Performance
20.64 GPixel/s
Pixel Rate
8.464 GPixel/s
82.56 GTexel/s
Texture Rate
33.86 GTexel/s
-
-
-
1.981 TFLOPS
FP32 (float)
812.5 GFLOPS
82.56 GFLOPS
FP64 (double)
33.86 GFLOPS
Board Design
140W
TDP
65W
300 W
Suggested PSU
250 W
2x DVI
1x HDMI 1.4a
1x DisplayPort 1.2
Outputs
2x DVI
1x mini-HDMI 1.4a
1x 6-pin
Power Connectors
1x 6-pin
Graphics Processor
GK106
GPU Name
GK107
GK106-400-A1
GPU Variant
GK107-450-A2
Kepler
Architecture
Kepler
TSMC
Foundry
TSMC
28 nm
Process Size
28 nm
2.54 billion
Transistors
1.27 billion
221 mm²
Die Size
118 mm²
Graphics Features
12 (11_0)
DirectX
12 (11_0)
4.6
OpenGL
4.6
3.0
OpenCL
3.0
1.1
Vulkan
1.1
3.0
CUDA
3.0
5.1
Shader Model
5.1
