Radeon HD 4870 and GeForce 920A - video card comparison
Passmark - benchmark score Radeon HD 4870 vs GeForce 920A
Radeon HD 4870
1411
GeForce 920A
866
Radeon HD 4870
GeForce 920A
Architecture
TeraScale (2005−2013)
Kepler 2.0 (2013−2015)
GPU code name
RV770
GK208B
Market segment
Desktop
Desktop
Release date
June 2008
March 2015
Interface
PCIe 2.0 x16
PCIe 3.0 x8
Supplementary power connectors
2x 6-pin
-
Core clock speed
750 MHz
954 MHz
Floating-point performance
1,200.0 gflops
441.6 gflops
Manufacturing process technology
55 nm
28 nm
Number of transistors
956 million
915 million
Pipelines / CUDA cores
800
384
Power consumption (TDP)
150 Watt
33 Watt
Texture fill rate
30.00
30.53
Memory type
GDDR5
DDR3
Maximum RAM amount
1 GB
2 GB
Memory bus width
256 Bit
64 Bit
Memory clock speed
3600 MHz
2002 MHz
Memory bandwidth
115.2 GB/s
16.02 GB/s
Display Connectors
2x DVI, 1x S-Video
No outputs
DirectX
10.1 (10_1)
12 (11_0)
CUDA
-
3.5
OpenCL
1.1
1.2
OpenGL
3.3
4.6
Shader Model
4.1
5.1
Vulkan
N/A
1.1.126