Report an Error

Intel Data Center GPU Max Subsystem

Graphics Processor
Ponte Vecchio x4
Cores
16384 x4
TMUs
1024 x4
ROPs
N/A
Memory Size
128 GB x4
Memory Type
HBM2e
Bus Width
8192 bit x4
GPU Chip
GPU
The Data Center GPU Max Subsystem is a professional graphics card by Intel, launched on January 10th, 2023. Built on the 10 nm process, and based on the Ponte Vecchio graphics processor, the card supports DirectX 12. The Ponte Vecchio graphics processor is a large chip with a die area of 1280 mm² and 100,000 million transistors. Data Center GPU Max Subsystem combines four graphics processors to increase performance. It features 16384 shading units, 1024 texture mapping units, and 0 ROPs, per GPU. Also included are 1024 tensor cores which help improve the speed of machine learning applications. The card also has 128 raytracing acceleration cores. Intel has paired 512 GB HBM2e memory with the Data Center GPU Max Subsystem, which are connected using a 8192-bit memory interface per GPU (each GPU manages 131,072 MB). The GPU is operating at a frequency of 900 MHz, which can be boosted up to 1600 MHz, memory is running at 1565 MHz.
Being a dual-slot card, the Intel Data Center GPU Max Subsystem draws power from 1x 16-pin power connector, with power draw rated at 2400 W maximum. This device has no display connectivity, as it is not designed to have monitors connected to it. Data Center GPU Max Subsystem is connected to the rest of the system using a PCI-Express 5.0 x16 interface. The card measures 267 mm in length, and features a dual-slot cooling solution.

Graphics Processor

GPU Name
Ponte Vecchio
Architecture
Generation 12.5
Foundry
Intel
Process Size
10 nm
Transistors
100,000 million
Density
78.1M / mm²
Die Size
1280 mm²

Graphics Card

Release Date
Jan 10th, 2023
Generation
Data Center GPU
(Ponte Vecchio)
Successor
H3C Graphics
Production
Active
Bus Interface
PCIe 5.0 x16

Clock Speeds

Base Clock
900 MHz
Boost Clock
1600 MHz
Memory Clock
1565 MHz
3.1 Gbps effective

Memory

Memory Size
128 GB
Memory Type
HBM2e
Memory Bus
8192 bit
Bandwidth
3.21 TB/s

Render Config

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

Theoretical Performance

Pixel Rate
0 MPixel/s
Texture Rate
1,638 GTexel/s
FP16 (half)
52.43 TFLOPS (1:1)
FP32 (float)
52.43 TFLOPS
FP64 (double)
52.43 TFLOPS (1:1)

Board Design

Slot Width
Dual-slot
Length
267 mm
10.5 inches
TDP
2400 W
Suggested PSU
2800 W
Outputs
No outputs
Power Connectors
1x 16-pin

Graphics Features

DirectX
12 (12_1)
OpenGL
4.6
OpenCL
3.0
Vulkan
N/A
Shader Model
6.6
Aug 16th, 2024 06:46 EDT change timezone

New Forum Posts

Popular Reviews

Controversial News Posts