NVIDIA GeForce RTX 2070 vs NVIDIA CMP 40HX

What is the difference between NVIDIA GeForce RTX 2070 and NVIDIA CMP 40HX. Find out which graphics card has better performance.

Graphics Processor (GPU)

TU106GPU NameTU106
TU106-400A-A1Model numberTU106-100-A1
TuringArchitectureTuring
TSMCFoundryTSMC
12 nmProcess Size12 nm
10,800 millionTransistors10,800 million
445 mm²Die Size445 mm²

Graphics Card

Oct 17th, 2018Release DateFeb 25th, 2021
GeForce 20FamilyMining GPUs
ActiveProductionActive
499 USDLaunch Price699 USD
PCIe 3.0 x16Bus InterfacePCIe 3.0 x16

Memory

8 GBMemory Size8 GB
GDDR6Memory TypeGDDR6
256 bitMemory Bus256 bit
448.0 GB/sBandwidth448.0 GB/s

Performance

103.7 GPixel/sPixel fillrate105.6 GPixel/s
233.3 GTexel/sTexture fillrate237.6 GTexel/s
14.93 TFLOPS (2:1)FP16 (half) performance15.21 TFLOPS (2:1)
7.465 TFLOPSFP32 (float) performance7.603 TFLOPS
233.3 GFLOPS (1:32)FP64 (double) performance237.6 GFLOPS (1:32)

Clock Speeds

1410 MHzBase Clock1470 MHz
1620 MHzBoost Clock1650 MHz
1750 MHz
14 Gbps effective
Memory Clock1750 MHz
14 Gbps effective

Render Config

2304Shading Units2304
144TMUs144
64ROPs64
36RT Cores36
64 KB (per SM)L1 Cache64 KB (per SM)
4 MBL2 Cache4 MB
36SM Count36
288Tensor Cores288

Board Design

Dual-slotSlot WidthDual-slot
229 mm
9 inches
Length229 mm
9 inches
113 mm
4.4 inches
Width111 mm
4.4 inches
35 mm
1.4 inches
Height35 mm
1.4 inches
175 WThermal design power (TDP)185 W
450 WSuggested PSU450 W
1x DVI
1x HDMI
2x DisplayPort
1x USB Type-C
Display ConnectorsNo outputs
1x 8-pinPower Connectors1x 8-pin
PG160 SKU 52Board NumberPG161 SKU 100

API support

12 Ultimate (12_2)DirectX12 Ultimate (12_2)
4.6OpenGL4.6
3.0OpenCL3.0
1.2Vulkan1.2
6.6Shader Model6.6
7.5CUDA7.5