GeForce RTX 3080 Ti and Radeon HD 8670 OEM - video card comparison
GeForce RTX 3080 Ti
Radeon HD 8670 OEM
Architecture
Ampere (2020−2022)
GCN 1.0 (2012−2020)
GPU code name
Ampere GA102
Oland
Market segment
Desktop
Desktop
Release date
September 2020
January 2013
Interface
PCIe 4.0 x16
PCIe 3.0 x8
Supplementary power connectors
1x 12-pin
None
Boost clock speed
1667 MHz
1050 MHz
Core clock speed
1370 MHz
1000 MHz
Floating-point performance
-
806.4 gflops
Manufacturing process technology
8 nm
28 nm
Number of transistors
28,300 million
950 million
Pipelines / CUDA cores
10240
384
Power consumption (TDP)
350 Watt
75 Watt
Texture fill rate
532.8
25.20
Memory type
GDDR6X
GDDR5
Maximum RAM amount
12 GB
2 GB
Memory bus width
384 Bit
128 Bit
Memory clock speed
19000 MHz
4600 MHz
Memory bandwidth
912.4 GB/s
73.6 GB/s
Shared memory
-
-
Display Connectors
1x HDMI, 3x DisplayPort
1x DVI, 1x HDMI, 1x VGA
HDMI
+
+
DirectX
12 Ultimate (12_2)
12 (11_1)
CUDA
8.6
-
OpenCL
3.0
1.2
OpenGL
4.6
4.6
Shader Model
6.6
5.1
Vulkan
1.2
1.2.131