Intel Data Center GPU Max Subsystem vs NVIDIA Rubin Ultra GPU
GPU · Specs Compare
Graphics Card
Release Date
Jan 2023
Jan 2027
Generation
Data Center GPU
Server Rubin
Type
Professional
Desktop
Bus Interface
PCIe 5.0 x16
PCIe 6.0 x16
Clock Speeds
Base Clock
900 MHz
700 MHz
Boost Clock
1600 MHz
2267 MHz
Memory Clock
1565 MHz
1900 MHz
Memory
Memory Size
128GB
1024GB
Memory Type
HBM2e
HBM4e
Memory Bus
8192bit
32768bit
Bandwidth
3205GB/s
31.1TB/s
Render Config
Shading Units
16384
57344
TMUs
1024
1792
ROPs
0
24
Tensor Cores
1024
1792
RT Cores
128
—
L1 Cache
64 KB (per EU)
256 KB (per SM)
L2 Cache
408 MB
128 MB
SM Count
—
448
Theoretical Performance
Pixel Rate
0 MPixel/s
54.41 GPixel/s
Texture Rate
1638 GTexel/s
4062.5 GTexel/s
FP16 (half)
52.43 TFLOPS
520.0 TFLOPS
FP32 (float)
52.43 TFLOPS
260.0 TFLOPS
FP64 (double)
52.43 TFLOPS
65.00 TFLOPS
Graphics Processor
GPU Name
Ponte Vecchio
GR110
Architecture
Generation 12.5
Rubin Ultra
Foundry
Intel
TSMC
Process Size
10 nm
3 nm
Transistors
100 billion
670 billion
Die Size
1280 mm²
2912 mm²
Board Design
TDP
2400W
3600W
Suggested PSU
2800 W
4000 W
Outputs
No outputs
No outputs
Power Connectors
1x 16-pin
—
Graphics Features
DirectX
12 (12_1)
N/A
OpenGL
4.6
N/A
OpenCL
3.0
3.0
Vulkan
N/A
N/A
Shader Model
6.6
N/A
CUDA
—
10.7
▲ Specs Compare
Data sourced from official specifications and benchmark tests.