NVIDIA L40 CNX vs AMD Zhongshan Subor Z+ GPU
GPU · Specs Compare
Graphics Card
Release Date
Oct 2022
Aug 2018
Generation
Tesla Ada
Console GPU
Type
Professional
Game console
Bus Interface
PCIe 4.0 x16
—
Clock Speeds
Base Clock
1005 MHz
—
Boost Clock
2475 MHz
—
Memory Clock
2250 MHz
1200 MHz
Memory
Memory Size
24GB
8GB
Memory Type
GDDR6
GDDR5
Memory Bus
384bit
256bit
Bandwidth
864.0GB/s
153.6GB/s
Render Config
SM Count
142
—
Shading Units
18176
1536
TMUs
568
96
ROPs
192
32
Tensor Cores
568
—
RT Cores
142
—
L1 Cache
128 KB (per SM)
—
L2 Cache
48 MB
—
Compute Units
—
24
Theoretical Performance
Pixel Rate
475.2 GPixel/s
41.60 GPixel/s
Texture Rate
1406 GTexel/s
124.8 GTexel/s
FP16 (half)
89.97 TFLOPS
7.987 TFLOPS
FP32 (float)
89.97 TFLOPS
3.994 TFLOPS
FP64 (double)
1406 GFLOPS
249.6 GFLOPS
Graphics Processor
GPU Name
AD102
Fenghuang
GPU Variant
AD102-???-A1
Zhongshan Subor
Architecture
Ada Lovelace
GCN 5.0
Foundry
TSMC
TSMC
Process Size
4 nm
14 nm
Transistors
76.3 billion
Unknown
Die Size
608 mm²
397 mm²
Board Design
TDP
300W
100W
Suggested PSU
700 W
—
Outputs
1x HDMI 2.13x DisplayPort 1.4a
No outputs
Power Connectors
1x 16-pin
—
Graphics Features
DirectX
12 Ultimate (12_2)
12 (12_1)
OpenGL
4.6
4.6
OpenCL
3.0
2.1
Vulkan
1.3
1.2
CUDA
8.9
—
Shader Model
6.6
6.2
▲ Specs Compare
Data sourced from official specifications and benchmark tests.