GeForce RTX 3080 Ti and Radeon Instinct MI50 - video card comparison
GeForce RTX 3080 Ti
Radeon Instinct MI50
Architecture
Ampere (2020−2022)
GCN 5.1 (2018−2021)
GPU code name
Ampere GA102
Vega 20
Market segment
Desktop
Desktop
Release date
September 2020
November 2018
Interface
PCIe 4.0 x16
PCIe 4.0 x16
Supplementary power connectors
1x 12-pin
2x 8-pin
Boost clock speed
1667 MHz
1746 MHz
Core clock speed
1370 MHz
1200 MHz
Manufacturing process technology
8 nm
7 nm
Number of transistors
28,300 million
13,230 million
Pipelines / CUDA cores
10240
3840
Power consumption (TDP)
350 Watt
300 Watt
Texture fill rate
532.8
419.0
Memory type
GDDR6X
HBM2
Maximum RAM amount
12 GB
16 GB
Memory bus width
384 Bit
4096 Bit
Memory clock speed
19000 MHz
2 GB/s
Memory bandwidth
912.4 GB/s
1.02 TB/s
Shared memory
-
-
Display Connectors
1x HDMI, 3x DisplayPort
1x mini-DisplayPort 1.4a
HDMI
+
-
DirectX
12 Ultimate (12_2)
12 (12_1)
CUDA
8.6
-
OpenCL
3.0
2.1
OpenGL
4.6
4.6
Shader Model
6.6
6.7
Vulkan
1.2
1.3