Radeon HD 5970 and GeForce GTX 650 - video card comparison
Passmark - benchmark score Radeon HD 5970 vs GeForce GTX 650
Radeon HD 5970
2350
GeForce GTX 650
1801
Radeon HD 5970
GeForce GTX 650
Architecture
TeraScale 2 (2009−2015)
Kepler (2012−2018)
GPU code name
Hemlock
GK107
Market segment
Desktop
Desktop
Release date
November 2009
September 2012
Bus support
-
PCI Express 3.0
Interface
PCIe 2.0 x16
PCIe 3.0 x16
Supplementary power connectors
1x 6-pin + 1x 8-pin
One 6-pin
Core clock speed
725 MHz
1058 MHz
CUDA cores
-
384
Floating-point performance
2x 2,320.0 gflops
812.5 gflops
Manufacturing process technology
40 nm
28 nm
Number of transistors
2,154 million
1,270 million
Pipelines / CUDA cores
1600
384
Power consumption (TDP)
294 Watt
64 Watt
Texture fill rate
58.00
33.9 billion/sec
Memory type
GDDR5
GDDR5
Maximum RAM amount
1 GB
2 GB
Memory bus width
256 Bit
128-bit GDDR5
Memory clock speed
4000 MHz
5.0 GB/s
Memory bandwidth
128.0 GB/s
80.0 GB/s
Display Connectors
2x DVI, 1x mini-DisplayPort
One Dual Link DVI-I, One Dual Link DVI-D, One Mini HDMI
Multi monitor support
-
4 displays
HDMI
-
+
HDCP
-
+
Maximum VGA resolution
-
2048x1536
Audio input for HDMI
-
Internal
DirectX
11.2 (11_0)
12 (11_0)
CUDA
-
+
OpenCL
1.2
1.2
OpenGL
4.4
4.3
Shader Model
5.0
5.1
Vulkan
N/A
1.1.126
3D Blu-Ray
-
+
3D Gaming
-
+
3D Vision
-
+