Report an Error

AMD MI100

AMD's MI100 GPU uses the CDNA 1.0 architecture and is made using a 7 nm+ production process at TSMC. With a die size of 420 mm² and a transistor count of 19,000 million it is a very big chip. MI100 does not support DirectX. For GPU compute applications, OpenCL version 2.0 can be used. It features 7680 shading units, 480 texture mapping units and 64 ROPs.

Graphics Processor

GPU Name
MI100
Architecture
CDNA 1.0
Foundry
TSMC
Process Size
7 nm+
Transistors
19,000 million
Density
45.2M / mm²
Die Size
420 mm²
Released
Unknown

Graphics Features

DirectX
N/A
OpenGL
4.6
OpenCL
2.0
Vulkan
N/A
Shader Model
N/A
Compute
GFX9
ROCm
No

Render Config

Shading Units
7680
TMUs
480
ROPs
64
Compute Units
120
L1 Cache
16 KB per CU
L2 Cache
8192 KB
Max. TDP
300 W

All CDNA 1.0 GPUs

  • AMD MI100

AMD GPU Architecture History

Graphics cards using the AMD MI100 GPU

Name Chip Memory Shaders TMUs ROPs Base Clock Boost Clock Memory Clock
AMD Radeon Instinct MI100 MI100 XL 32 GB 7680 480 64 1000 MHz 1334 MHz 1200 MHz

MI100 GPU Notes

Architecture Codename: Arcturus
Graphics/Compute: GFX9 (gfx908)