AMD Zhongshan Subor Z+ GPU vs NVIDIA RTX PRO 6000 Blackwell
GPU · Specs Compare
Graphics Card
Release Date
Aug 2018
Mar 2025
Generation
Console GPU
Blackwell PRO W
Type
Game console
Desktop
Bus Interface
—
PCIe 5.0 x16
Clock Speeds
Memory Clock
1200 MHz
1750 MHz
Base Clock
—
1590 MHz
Boost Clock
—
2617 MHz
Memory
Memory Size
8GB
96GB
Memory Type
GDDR5
GDDR7
Memory Bus
256bit
512bit
Bandwidth
153.6GB/s
1.79TB/s
Render Config
Compute Units
24
—
Shading Units
1536
24064
TMUs
96
752
ROPs
32
192
SM Count
—
188
Tensor Cores
—
752
RT Cores
—
188
L1 Cache
—
128 KB (per SM)
L2 Cache
—
128 MB
Theoretical Performance
Pixel Rate
41.60 GPixel/s
502.5 GPixel/s
Texture Rate
124.8 GTexel/s
1968.0 GTexel/s
FP16 (half)
7.987 TFLOPS
126.0 TFLOPS
FP32 (float)
3.994 TFLOPS
126.0 TFLOPS
FP64 (double)
249.6 GFLOPS
1.968 TFLOPS
Graphics Processor
GPU Name
Fenghuang
GB202
GPU Variant
Zhongshan Subor
—
Architecture
GCN 5.0
Blackwell 2.0
Foundry
TSMC
TSMC
Process Size
14 nm
5 nm
Transistors
Unknown
92.2 billion
Die Size
397 mm²
750 mm²
Board Design
TDP
100W
600W
Outputs
No outputs
4x DisplayPort 2.1b
Suggested PSU
—
1000 W
Power Connectors
—
1x 16-pin
Graphics Features
DirectX
12 (12_1)
12 Ultimate (12_2)
OpenGL
4.6
4.6
OpenCL
2.1
3.0
Vulkan
1.2
1.4
Shader Model
6.2
6.9
CUDA
—
12.0
Benchmark
3DMark Time Spy
—
42,269
3DMark Time Spy Extreme
—
27,891
3DMark Speed Way
—
15,642
▲ Specs Compare
Data sourced from official specifications and benchmark tests.