ATI Radeon HD 5870 Eyefinity 6 vs NVIDIA RTX PRO 4000 Blackwell SFF
GPU · Specs Compare
Graphics Card
Release Date
Mar 2010
Aug 2025
Generation
Evergreen
Blackwell PRO W
Type
Desktop
Desktop
Bus Interface
PCIe 2.0 x16
PCIe 5.0 x8
Clock Speeds
Memory Clock
1200 MHz
1125 MHz
Base Clock
—
577 MHz
Boost Clock
—
1432 MHz
Memory
Memory Size
2GB
24GB
Memory Type
GDDR5
GDDR7
Memory Bus
256bit
192bit
Bandwidth
153.6GB/s
432.0GB/s
Render Config
Compute Units
20
—
Shading Units
1600
8960
TMUs
80
280
ROPs
32
96
L1 Cache
8 KB (per CU)
128 KB (per SM)
L2 Cache
512 KB
48 MB
SM Count
—
70
Tensor Cores
—
280
RT Cores
—
70
Theoretical Performance
Pixel Rate
27.20 GPixel/s
137.5 GPixel/s
Texture Rate
68.00 GTexel/s
401.0 GTexel/s
FP32 (float)
2.720 TFLOPS
25.66 TFLOPS
FP64 (double)
544.0 GFLOPS
401.0 GFLOPS
FP16 (half)
—
25.66 TFLOPS
Graphics Processor
GPU Name
Cypress
GB203
GPU Variant
Cypress XT(215-0735033)
—
Architecture
TeraScale 2
Blackwell 2.0
Foundry
TSMC
TSMC
Process Size
40 nm
5 nm
Transistors
2.154 billion
45.6 billion
Die Size
334 mm²
378 mm²
Board Design
TDP
228W
70W
Suggested PSU
550 W
250 W
Outputs
6x mini-DisplayPort 1.1
4x mini-DisplayPort 2.1b
Power Connectors
1x 6-pin + 1x 8-pin
None
Graphics Features
DirectX
11.2 (11_0)
12 Ultimate (12_2)
OpenGL
4.4
4.6
OpenCL
1.2
3.0
Vulkan
N/A
1.4
Shader Model
5.0
6.8
CUDA
—
12.0
▲ Specs Compare
Data sourced from official specifications and benchmark tests.