NVIDIA A100 SXM4 40 GB vs AMD Radeon Pro WX Vega M GL
GPU · Specs Compare
Graphics Card
Release Date
May 2020
Apr 2018
Generation
Tesla Ampere
Vega
Type
AI GPU
Mobile
Bus Interface
PCIe 4.0 x16
IGP
Clock Speeds
Base Clock
1095 MHz
931 MHz
Boost Clock
1410 MHz
1011 MHz
Memory Clock
1215 MHz
700 MHz
Memory
Memory Size
40GB
4GB
Memory Type
HBM2e
HBM2
Memory Bus
5120bit
1024bit
Bandwidth
1555GB/s
179.2GB/s
Render Config
SM Count
108
—
Shading Units
6912
1280
TMUs
432
80
ROPs
160
32
Tensor Cores
432
—
L1 Cache
192 KB (per SM)
16 KB (per CU)
L2 Cache
40 MB
1024 KB
Compute Units
—
20
Theoretical Performance
Pixel Rate
225.6 GPixel/s
32.35 GPixel/s
Texture Rate
609.1 GTexel/s
80.88 GTexel/s
FP16 (half)
77.97 TFLOPS
2.588 TFLOPS
FP32 (float)
19.49 TFLOPS
2.588 TFLOPS
FP64 (double)
9.746 TFLOPS
161.8 GFLOPS
Graphics Processor
GPU Name
GA100
Polaris 22
Architecture
Ampere
GCN 4.0
Foundry
TSMC
GlobalFoundries
Process Size
7 nm
14 nm
Transistors
54.2 billion
5 billion
Die Size
826 mm²
208 mm²
GPU Variant
—
VegaM MGL XL
Board Design
TDP
400W
65W
Suggested PSU
800 W
—
Outputs
No outputs
No outputs
Power Connectors
None
—
Graphics Features
DirectX
N/A
12 (12_0)
OpenGL
N/A
4.6
OpenCL
3.0
2.1
Vulkan
N/A
1.2
CUDA
8.0
—
Shader Model
N/A
6.4
Benchmark
Blender
2,186
—
▲ Specs Compare
Data sourced from official specifications and benchmark tests.