
Radeon Instinct MI100 and GeForce RTX 3080 - video card comparison
Radeon Instinct MI100
GeForce RTX 3080
Architecture
CDNA 1.0 (2020)
Ampere (2020−2022)
GPU code name
Arcturus
Ampere GA102
Market segment
Desktop
Desktop
Release date
November 2020
September 2020
Interface
PCIe 4.0 x16
PCIe 4.0 x16
Supplementary power connectors
2x 8-pin
1x 12-pin
Boost clock speed
1502 MHz
1710 MHz
Core clock speed
-
1450 MHz
Manufacturing process technology
7 nm
8 nm
Number of transistors
50,000 million
28,300 million
Pipelines / CUDA cores
7680
8704
Power consumption (TDP)
300 Watt
320 Watt
Texture fill rate
721.0
465.1
Memory type
HBM2
GDDR6X
Maximum RAM amount
32 GB
10 GB
Memory bus width
4096 Bit
320 Bit
Memory clock speed
2.4 GB/s
19000 MHz
Memory bandwidth
1,229 GB/s
760.3 GB/s
Shared memory
-
-
Display Connectors
No outputs
1x HDMI, 3x DisplayPort
HDMI
-
+
DirectX
N/A
12 Ultimate (12_2)
CUDA
-
8.5
OpenCL
2.1
2.0
OpenGL
N/A
4.6
Shader Model
N/A
6.5
Vulkan
N/A
1.2