Radeon HD 3850 and GeForce 840A - video card comparison
Passmark - benchmark score Radeon HD 3850 vs GeForce 840A
Radeon HD 3850
404
GeForce 840A
1214
Radeon HD 3850
GeForce 840A
Architecture
TeraScale (2005−2013)
Maxwell (2014−2018)
GPU code name
RV670
GM108
Market segment
Desktop
Desktop
Release date
November 2007
March 2014
Interface
PCIe 2.0 x16
PCIe 3.0 x8
Supplementary power connectors
1x 6-pin
-
Boost clock speed
-
1124 MHz
Core clock speed
668 MHz
1029 MHz
Floating-point performance
427.5 gflops
863.2 gflops
Manufacturing process technology
55 nm
28 nm
Number of transistors
666 million
-
Pipelines / CUDA cores
320
384
Power consumption (TDP)
75 Watt
33 Watt
Texture fill rate
10.69
17.98
Memory type
GDDR3
DDR3
Maximum RAM amount
1 GB
2 GB
Memory bus width
256 Bit
64 Bit
Memory clock speed
1660 MHz
2002 MHz
Memory bandwidth
53.12 GB/s
16.02 GB/s
Display Connectors
2x DVI, 1x S-Video
No outputs
DirectX
10.1 (10_1)
12 (11_0)
CUDA
-
5.0
OpenCL
N/A
1.2
OpenGL
3.3
4.6
Shader Model
4.1
5.1
Vulkan
N/A
1.1.126