NVIDIA RTX PRO 4000 Blackwell Embedded vs AMD Radeon PRO W9070
GPU · Specs Compare
Graphics Card
Release Date
Unknown
Unknown
Generation
Blackwell-MW
Mining GPUs
Type
Mobile
Desktop
Bus Interface
PCIe 5.0 x16
PCIe 3.0 x16
Clock Speeds
Base Clock
975 MHz
1120 MHz
Boost Clock
1702 MHz
1206 MHz
Memory Clock
1750 MHz
2100 MHz
Memory
Memory Size
16GB
4GB
Memory Type
GDDR7
GDDR5
Memory Bus
256bit
256bit
Bandwidth
896.0GB/s
268.8GB/s
Render Config
SM Count
60
—
Shading Units
7680
2048
TMUs
240
128
ROPs
96
32
Tensor Cores
240
128
RT Cores
60
64
L1 Cache
128 KB (per SM)
16 KB (per CU)
L2 Cache
64 MB
2 MB
Compute Units
—
32
L3 Cache
—
64 MB
Theoretical Performance
Pixel Rate
163.4 GPixel/s
38.59 GPixel/s
Texture Rate
408.5 GTexel/s
154.4 GTexel/s
FP16 (half)
26.14 TFLOPS
4.940 TFLOPS
FP32 (float)
26.14 TFLOPS
4.940 TFLOPS
FP64 (double)
408.5 GFLOPS
308.7 GFLOPS
Graphics Processor
GPU Name
GB203
Polaris 20
Architecture
Blackwell 2.0
GCN 4.0
Foundry
TSMC
GlobalFoundries
Process Size
5 nm
14 nm
Transistors
45.6 billion
5.7 billion
Die Size
378 mm²
232 mm²
GPU Variant
—
Navi 48 XTW
Board Design
TDP
80W
180W
Outputs
Portable Device Dependent
1x HDMI 2.0b
Power Connectors
None
2x 8-pin
Suggested PSU
—
450 W
Graphics Features
DirectX
12 Ultimate (12_2)
12 (12_0)
OpenGL
4.6
4.6
OpenCL
3.0
2.1
Vulkan
1.4
1.3
CUDA
12.0
—
Shader Model
6.8
6.7
▲ Specs Compare
Data sourced from official specifications and benchmark tests.