Radeon Instinct MI100 and GeForce RTX 4070 Ti SUPER - video card comparison
Radeon Instinct MI100
GeForce RTX 4070 Ti SUPER
Architecture
CDNA 1.0 (2020)
Ada Lovelace
GPU code name
Arcturus
AD103-275
Market segment
Desktop
Desktop
Release date
November 2020
January 2024
Interface
PCIe 4.0 x16
PCIe 4.0 x16
Supplementary power connectors
2x 8-pin
1x 16-pin
Boost clock speed
1502 MHz
2610 MHz
Core clock speed
-
2340 MHz
Manufacturing process technology
7 nm
5 nm
Number of transistors
50,000 million
45,900 million
Pipelines / CUDA cores
7680
8448
Power consumption (TDP)
300 Watt
285 Watt
Texture fill rate
721.0
689.0
Memory type
HBM2
GDDR6X
Maximum RAM amount
32 GB
16 GB
Memory bus width
4096 Bit
256 Bit
Memory clock speed
2.4 GB/s
21000 MHz
Memory bandwidth
1,229 GB/s
716.8 GB/s
Shared memory
-
-
Display Connectors
No outputs
1x HDMI 2.1, 3x DisplayPort 1.4a
HDMI
-
+
DirectX
N/A
12 Ultimate (12_2)
CUDA
-
8.9
OpenCL
2.1
3.0
OpenGL
N/A
4.6
Shader Model
N/A
6.7
Vulkan
N/A
1.3