NVIDIA Rubin Ultra GPU vs AMD Radeon Pro V620
GPU · Specs Compare
Graphics Card
Release Date
Jan 2027
Nov 2021
Generation
Server Rubin
Radeon Pro
Type
Desktop
Desktop
Bus Interface
PCIe 6.0 x16
PCIe 4.0 x16
Clock Speeds
Base Clock
700 MHz
1825 MHz
Boost Clock
2267 MHz
2200 MHz
Memory Clock
1900 MHz
2000 MHz
Memory
Memory Size
1024GB
32GB
Memory Type
HBM4e
GDDR6
Memory Bus
32768bit
256bit
Bandwidth
31.1TB/s
512.0GB/s
Render Config
SM Count
448
—
Shading Units
57344
4608
TMUs
1792
288
ROPs
24
128
Tensor Cores
1792
—
L1 Cache
256 KB (per SM)
128 KB per Array
L2 Cache
128 MB
4 MB
Compute Units
—
72
RT Cores
—
72
L3 Cache
—
128 MB
Theoretical Performance
Pixel Rate
54.41 GPixel/s
281.6 GPixel/s
Texture Rate
4062.5 GTexel/s
633.6 GTexel/s
FP16 (half)
520.0 TFLOPS
40.55 TFLOPS
FP32 (float)
260.0 TFLOPS
20.28 TFLOPS
FP64 (double)
65.00 TFLOPS
1267 GFLOPS
Graphics Processor
GPU Name
GR110
Navi 21
Architecture
Rubin Ultra
RDNA 2.0
Foundry
TSMC
TSMC
Process Size
3 nm
7 nm
Transistors
670 billion
26.8 billion
Die Size
2912 mm²
520 mm²
GPU Variant
—
Navi 21 XT(215-121000177)
Board Design
TDP
3600W
300W
Suggested PSU
4000 W
700 W
Outputs
No outputs
No outputs
Power Connectors
—
2x 8-pin
Graphics Features
DirectX
N/A
12 Ultimate (12_2)
OpenGL
N/A
4.6
OpenCL
3.0
2.1
Vulkan
N/A
1.3
CUDA
10.7
—
Shader Model
N/A
6.5
▲ Specs Compare
Data sourced from official specifications and benchmark tests.