NVIDIA RTX PRO 6000 Blackwell Server vs AMD Radeon Pro Duo Polaris
We compared two Desktop platform GPUs: 96GB VRAM RTX PRO 6000 Blackwell Server and 16GB VRAM Radeon Pro Duo Polaris to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA RTX PRO 6000 Blackwell Server 's Advantages
Released 7 years and 11 months late
Boost Clock2617MHz
More VRAM (96GB vs 16GB)
Larger VRAM bandwidth (1.79TB/s vs 224.0GB/s)
21760 additional rendering cores
AMD Radeon Pro Duo Polaris 's Advantages
Lower TDP (250W vs 600W)
Score
Benchmark
FP32 (float)
RTX PRO 6000 Blackwell Server
+2099%
126 TFLOPS
Radeon Pro Duo Polaris
5.728 TFLOPS
Graphics Card
Mar 2025
Release Date
Apr 2017
Server Blackwell
Generation
Radeon Pro GCN
Desktop
Type
Desktop
PCIe 5.0 x16
Bus Interface
PCIe 3.0 x16
Clock Speeds
1590 MHz
Base Clock
-
2617 MHz
Boost Clock
-
1750 MHz
Memory Clock
1750 MHz
Memory
96GB
Memory Size
16GB
GDDR7
Memory Type
GDDR5
512bit
Memory Bus
256bit
1.79TB/s
Bandwidth
224.0GB/s
Render Config
-
Compute Units
36
188
SM Count
-
24064
Shading Units
2304
752
TMUs
144
192
ROPs
32
752
Tensor Cores
-
188
RT Cores
-
128 KB (per SM)
L1 Cache
16 KB (per CU)
128 MB
L2 Cache
2 MB
-
-
-
Theoretical Performance
502.5 GPixel/s
Pixel Rate
39.78 GPixel/s
1968 GTexel/s
Texture Rate
179.0 GTexel/s
126.0 TFLOPS
FP16 (half)
5.728 TFLOPS
126.0 TFLOPS
FP32 (float)
5.728 TFLOPS
1.968 TFLOPS
FP64 (double)
358.0 GFLOPS
Board Design
600W
TDP
250W
1000 W
Suggested PSU
600 W
4x DisplayPort 2.1b
Outputs
1x HDMI 2.0b
3x DisplayPort 1.4a
1x 16-pin
Power Connectors
1x 6-pin + 1x 8-pin
Graphics Processor
GB202
GPU Name
Ellesmere
-
GPU Variant
Ellesmere Gemini GL
Blackwell 2.0
Architecture
GCN 4.0
TSMC
Foundry
GlobalFoundries
5 nm
Process Size
14 nm
92.2 billion
Transistors
5.7 billion
750 mm²
Die Size
232 mm²
Graphics Features
12 Ultimate (12_2)
DirectX
12 (12_0)
4.6
OpenGL
4.6
3.0
OpenCL
2.1
1.4
Vulkan
1.3
12.0
CUDA
-
6.8
Shader Model
6.7
