AMD Radeon Instinct MI50 vs NVIDIA RTX A2000 Max-Q

GPU · Specs Compare

Products
GPU AMD Radeon Instinct MI50 AMD · 2018-11
GPU NVIDIA RTX A2000 Max-Q NVIDIA · 2021-04
Graphics Card
Release Date
Nov 2018
Apr 2021
Generation
Radeon Instinct
Ampere-MW
Type
Professional
Mobile
Bus Interface
PCIe 4.0 x16
PCIe 4.0 x16
Clock Speeds
Base Clock
1200 MHz
682 MHz
Boost Clock
1746 MHz
1207 MHz
Memory Clock
1000 MHz
1375 MHz
Memory
Memory Size
16GB
4GB
Memory Type
HBM2
GDDR6
Memory Bus
4096bit
128bit
Bandwidth
1.02TB/s
176.0GB/s
Render Config
Compute Units
60
—
Shading Units
3840
2560
TMUs
240
80
ROPs
64
48
RT Cores
32
20
L1 Cache
16 KB (per CU)
128 KB (per SM)
L2 Cache
4 MB
2 MB
L3 Cache
32 MB
—
SM Count
—
20
Tensor Cores
—
80
Theoretical Performance
Pixel Rate
111.7 GPixel/s
57.94 GPixel/s
Texture Rate
419.0 GTexel/s
96.56 GTexel/s
FP16 (half)
26.82 TFLOPS
6.180 TFLOPS
FP32 (float)
13.41 TFLOPS
6.180 TFLOPS
FP64 (double)
6.705 TFLOPS
96.56 GFLOPS
Graphics Processor
GPU Name
Vega 20
GA107
GPU Variant
Vega 20 GLXT
—
Architecture
GCN 5.1
Ampere
Foundry
TSMC
Samsung
Process Size
7 nm
8 nm
Transistors
13.23 billion
8.7 billion
Die Size
331 mm²
200 mm²
Board Design
TDP
300W
95W
Suggested PSU
700 W
—
Outputs
1x mini-DisplayPort 1.4a
Portable Device Dependent
Power Connectors
2x 8-pin
None
Graphics Features
DirectX
12 (12_1)
12 Ultimate (12_2)
OpenGL
4.6
4.6
OpenCL
2.1
3.0
Vulkan
1.3
1.3
Shader Model
6.7
6.8
CUDA
—
8.6
▲ Specs Compare Data sourced from official specifications and benchmark tests.