Intel Data Center GPU Max Subsystem vs NVIDIA A10 PCIe
We compared two Professional market GPUs: 128GB VRAM Data Center GPU Max Subsystem and 24GB VRAM A10 PCIe to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
Intel Data Center GPU Max Subsystem 's Advantages
Released 1 years and 9 months late
More VRAM (128GB vs 24GB)
Larger VRAM bandwidth (3205GB/s vs 600.2GB/s)
7168 additional rendering cores
NVIDIA A10 PCIe 's Advantages
Boost Clock has increased by 6% (1695MHz vs 1600MHz)
Lower TDP (150W vs 2400W)
Score
Benchmark
FP32 (float)
Data Center GPU Max Subsystem
+67%
52.43 TFLOPS
A10 PCIe
31.24 TFLOPS
Graphics Card
Jan 2023
Release Date
Apr 2021
Data Center GPU
Generation
Tesla
Professional
Type
Professional
PCIe 5.0 x16
Bus Interface
PCIe 4.0 x16
Clock Speeds
900 MHz
Base Clock
885 MHz
1600 MHz
Boost Clock
1695 MHz
1565 MHz
Memory Clock
1563 MHz
Memory
128GB
Memory Size
24GB
HBM2e
Memory Type
GDDR6
8192bit
Memory Bus
384bit
3205GB/s
Bandwidth
600.2GB/s
Render Config
-
-
-
-
SM Count
72
16384
Shading Units
9216
1024
TMUs
288
0
ROPs
96
1024
Tensor Cores
288
128
RT Cores
72
64 KB (per EU)
L1 Cache
128 KB (per SM)
408 MB
L2 Cache
6 MB
-
-
-
Theoretical Performance
0 MPixel/s
Pixel Rate
162.7 GPixel/s
1638 GTexel/s
Texture Rate
488.2 GTexel/s
52.43 TFLOPS
FP16 (half)
31.24 TFLOPS
52.43 TFLOPS
FP32 (float)
31.24 TFLOPS
52.43 TFLOPS
FP64 (double)
976.3 GFLOPS
Board Design
2400W
TDP
150W
2800 W
Suggested PSU
450 W
No outputs
Outputs
No outputs
1x 16-pin
Power Connectors
1x 8-pin
Graphics Processor
Ponte Vecchio
GPU Name
GA102
-
GPU Variant
GA102-890-A1
Generation 12.5
Architecture
Ampere
Intel
Foundry
Samsung
10 nm
Process Size
8 nm
100 billion
Transistors
28.3 billion
1280 mm²
Die Size
628 mm²
Graphics Features
12 (12_1)
DirectX
12 Ultimate (12_2)
4.6
OpenGL
4.6
3.0
OpenCL
3.0
N/A
Vulkan
1.3
-
CUDA
8.6
6.6
Shader Model
6.6
