NVIDIA CMP 40HX vs ATI Radeon HD 5870 Eyefinity 6
We compared two Desktop platform GPUs: 8GB VRAM CMP 40HX and 2GB VRAM Radeon HD 5870 Eyefinity 6 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA CMP 40HX 's Advantages
Released 10 years and 11 months late
Boost Clock1650MHz
More VRAM (8GB vs 2GB)
Larger VRAM bandwidth (448.0GB/s vs 153.6GB/s)
704 additional rendering cores
Lower TDP (185W vs 228W)
Score
Benchmark
FP32 (float)
CMP 40HX
+179%
7.603 TFLOPS
Radeon HD 5870 Eyefinity 6
2.72 TFLOPS
Graphics Card
Feb 2021
Release Date
Mar 2010
Mining GPUs
Generation
Evergreen
Desktop
Type
Desktop
PCIe 1.0 x4
Bus Interface
PCIe 2.0 x16
Clock Speeds
1470 MHz
Base Clock
-
1650 MHz
Boost Clock
-
1750 MHz
Memory Clock
1200 MHz
Memory
8GB
Memory Size
2GB
GDDR6
Memory Type
GDDR5
256bit
Memory Bus
256bit
448.0GB/s
Bandwidth
153.6GB/s
Render Config
-
Compute Units
20
36
SM Count
-
2304
Shading Units
1600
144
TMUs
80
64
ROPs
32
288
Tensor Cores
-
36
RT Cores
-
64 KB (per SM)
L1 Cache
8 KB (per CU)
4 MB
L2 Cache
512 KB
-
-
-
Theoretical Performance
105.6 GPixel/s
Pixel Rate
27.20 GPixel/s
237.6 GTexel/s
Texture Rate
68.00 GTexel/s
15.21 TFLOPS
FP16 (half)
-
7.603 TFLOPS
FP32 (float)
2.720 TFLOPS
237.6 GFLOPS
FP64 (double)
544.0 GFLOPS
Board Design
185W
TDP
228W
450 W
Suggested PSU
550 W
No outputs
Outputs
6x mini-DisplayPort 1.1
1x 8-pin
Power Connectors
1x 6-pin + 1x 8-pin
Graphics Processor
TU106
GPU Name
Cypress
TU106-100-A1
GPU Variant
Cypress XT
(215-0735033)
Turing
Architecture
TeraScale 2
TSMC
Foundry
TSMC
12 nm
Process Size
40 nm
10.8 billion
Transistors
2.154 billion
445 mm²
Die Size
334 mm²
Graphics Features
12 Ultimate (12_2)
DirectX
11.2 (11_0)
4.6
OpenGL
4.4
3.0
OpenCL
1.2
1.3
Vulkan
N/A
7.5
CUDA
-
6.8
Shader Model
5.0
