Home Intel Data Center GPU Max Subsystem

Intel Data Center GPU Max Subsystem

This is a GPU manufactured with Intel 10nm process, based on Intel Generation 12.5 architecture and released on Jan 2023. It features 100 billion transistors, 16384 Rendering cores and 128GB HBM2e memory, with 408MB L2 cache, theoretical performance of 52.43TFLOPS, with total power consumption of 2400W.

Graphics Card

[Report Issues]
Release Date
Jan 2023
Generation
Data Center GPU
Type
Professional
Bus Interface
PCIe 5.0 x16

Clock Speeds

[Report Issues]
Base Clock
900MHz
Boost Clock
1600MHz
Memory Clock
1565MHz

Memory

[Report Issues]
Memory Size
128GB
Memory Type
HBM2e
Memory Bus
8192bit
Bandwidth
3205 GB/s

Render Config

[Report Issues]
Shading Units
16384
TMUs
1024
ROPs
0
Tensor Cores
1024
RT Cores
128
L1 Cache
64 KB (per EU)
L2 Cache
408MB

Theoretical Performance

[Report Issues]
Pixel Rate
0 MPixel/s
Texture Rate
1638 GTexel/s
FP16 (half)
52.43 TFLOPS
FP32 (float)
52.43 TFLOPS
FP64 (double)
52.43 TFLOPS

Graphics Processor

[Report Issues]
GPU Name
Ponte Vecchio
Architecture
Generation 12.5
Foundry
Intel
Process Size
10 nm
Transistors
100,000 million
Die Size
1280mm²

Board Design

[Report Issues]
TDP
2400W
Suggested PSU
2800W
Outputs
No outputs
Power Connectors
1x 16-pin

Graphics Features

[Report Issues]
DirectX
12 (12_1)
OpenGL
4.6
OpenCL
3.0
Vulkan
N/A
Shader Model
6.6

Rankings

[Report Issues]
FP32 (float)
NVIDIA L20 48 GB GDDR6
59.35 TFLOPS
NVIDIA H800 SXM5 80 GB HBM3
59.3 TFLOPS
NVIDIA H100 CNX 80 GB HBM2e
53.84 TFLOPS
52.43 TFLOPS
52.43 TFLOPS
52.22 TFLOPS
51.48 TFLOPS

Related Comparisons

© 2024 - TopCPU.net   Contact Us Privacy Policy