NVIDIA GeForce RTX 3070 vs NVIDIA CMP 50HX

What is the difference between NVIDIA GeForce RTX 3070 and NVIDIA CMP 50HX. Find out which graphics card has better performance.

GeForce RTX 3070

GeForce RTX 3070

CMP 50HX

Graphics Processor (GPU)

GA104	GPU Name	TU102
GA104-300-A1	Model number	TU102-100-A1
Ampere	Architecture	Turing
Samsung	Foundry	TSMC
8 nm	Process Size	12 nm
17,400 million	Transistors	18,600 million
392 mm²	Die Size	754 mm²

Graphics Card

Sep 1st, 2020	Release Date	Jun 24th, 2021
GeForce 30	Family	Mining GPUs
Active	Production	Active
PCIe 4.0 x16	Bus Interface	PCIe 3.0 x16

Memory

8 GB	Memory Size	10 GB
GDDR6	Memory Type	GDDR6
256 bit	Memory Bus	320 bit
448.0 GB/s	Bandwidth	560.0 GB/s

Performance

165.6 GPixel/s	Pixel fillrate	123.6 GPixel/s
317.4 GTexel/s	Texture fillrate	296.6 GTexel/s
20.31 TFLOPS (1:1)	FP16 (half) performance	22.15 TFLOPS (2:1)
20.31 TFLOPS	FP32 (float) performance	11.07 TFLOPS
317.4 GFLOPS (1:64)	FP64 (double) performance	346.1 GFLOPS (1:32)

Clock Speeds

1500 MHz	Base Clock	1350 MHz
1725 MHz	Boost Clock	1545 MHz
1750 MHz 14 Gbps effective	Memory Clock	1750 MHz 14 Gbps effective

Render Config

5888	Shading Units	3584
184	TMUs	192
96	ROPs	80
46	RT Cores	56
128 KB (per SM)	L1 Cache	64 KB (per SM)
4 MB	L2 Cache	5 MB
46	SM Count	56
184	Tensor Cores	448

Board Design

Dual-slot	Slot Width	Dual-slot
242 mm 9.5 inches	Length	267 mm 10.5 inches
112 mm 4.4 inches	Width	116 mm 4.6 inches
220 W	Thermal design power (TDP)	250 W
550 W	Suggested PSU	600 W
2x HDMI 2x DisplayPort	Display Connectors	No outputs
1x 12-pin	Power Connectors	2x 8-pin
PG142 SKU 10	Board Number	PG150 SKU 100

API support

12 Ultimate (12_2)	DirectX	12 Ultimate (12_2)
4.6	OpenGL	4.6
3.0	OpenCL	3.0
1.2	Vulkan	1.2
6.6	Shader Model	6.6
8.6	CUDA	7.5