Home GPU Comparison AMD Radeon Pro Vega 20 vs NVIDIA RTX 2000 Max-Q Ada Generation

AMD Radeon Pro Vega 20 vs NVIDIA RTX 2000 Max-Q Ada Generation

We compared two Mobile platform GPUs: 4GB VRAM Radeon Pro Vega 20 and 8GB VRAM RTX 2000 Max Q Ada Generation to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.

Main Differences

NVIDIA RTX 2000 Max-Q Ada Generation 's Advantages
Released 4 years and 4 months late
Boost Clock has increased by 13% (1455MHz vs 1283MHz)
More VRAM (8GB vs 4GB)
Larger VRAM bandwidth (256.0GB/s vs 189.4GB/s)
1792 additional rendering cores
Lower TDP (35W vs 100W)

Score

Benchmark

FP32 (float)
Radeon Pro Vega 20
3.284 TFLOPS
RTX 2000 Max Q Ada Generation +172%
8.94 TFLOPS
VS

Graphics Card

Nov 2018
Release Date
Mar 2023
Radeon Pro Mac
Generation
Quadro Ada-M
Mobile
Type
Mobile
PCIe 3.0 x16
Bus Interface
PCIe 4.0 x16

Clock Speeds

815 MHz
Base Clock
930 MHz
1283 MHz
Boost Clock
1455 MHz
740 MHz
Memory Clock
2000 MHz

Memory

4GB
Memory Size
8GB
HBM2
Memory Type
GDDR6
1024bit
Memory Bus
128bit
189.4GB/s
Bandwidth
256.0GB/s

Render Config

-
SM Count
24
20
Compute Units
-
1280
Shading Units
3072
80
TMUs
96
32
ROPs
48
-
Tensor Cores
96
-
RT Cores
24
16 KB (per CU)
L1 Cache
128 KB (per SM)
1024 KB
L2 Cache
12 MB

Theoretical Performance

41.06 GPixel/s
Pixel Rate
69.84 GPixel/s
102.6 GTexel/s
Texture Rate
139.7 GTexel/s
6.569 TFLOPS
FP16 (half)
8.940 TFLOPS
3.284 TFLOPS
FP32 (float)
8.940 TFLOPS
205.3 GFLOPS
FP64 (double)
139.7 GFLOPS

Graphics Processor

Vega 12
GPU Name
AD107
Vega 12 XTA
GPU Variant
-
GCN 5.0
Architecture
Ada Lovelace
GlobalFoundries
Foundry
TSMC
14 nm
Process Size
5 nm
Unknown
Transistors
18.9 billion
Unknown
Die Size
159 mm²

Board Design

100W
TDP
35W
-
Suggested PSU
-
No outputs
Outputs
Portable Device Dependent
-
Power Connectors
None

Graphics Features

12 (12_1)
DirectX
12 Ultimate (12_2)
4.6
OpenGL
4.6
2.1
OpenCL
3.0
1.2
Vulkan
1.3
-
CUDA
8.9
6.3
Shader Model
6.7

Related GPU Comparisons

Related News

© 2024 - TopCPU.net   Contact Us Privacy Policy