Home NVIDIA CMP 40HX

NVIDIA CMP 40HX

This is a GPU manufactured with TSMC 12nm process, based on Nvidia Turing architecture and released on Feb 2021. It features 10.8 billion transistors, 2304 CUDA cores and 8GB GDDR6 memory, with 4MB L2 cache, theoretical performance of 7.603TFLOPS, with total power consumption of 185W.

Graphics Card

[Report Issues]
Release Date
Feb 2021
Generation
Mining GPUs
Type
Desktop
Bus Interface
PCIe 3.0 x16

Clock Speeds

[Report Issues]
Base Clock
1470MHz
Boost Clock
1650MHz
Memory Clock
1750MHz

Memory

[Report Issues]
Memory Size
8GB
Memory Type
GDDR6
Memory Bus
256bit
Bandwidth
448.0 GB/s

Render Config

[Report Issues]
SM Count
36
Shading Units
2304
TMUs
144
ROPs
64
Tensor Cores
288
RT Cores
36
L1 Cache
64 KB (per SM)
L2 Cache
4MB

Theoretical Performance

[Report Issues]
Pixel Rate
105.6 GPixel/s
Texture Rate
237.6 GTexel/s
FP16 (half)
15.21 TFLOPS
FP32 (float)
7.603 TFLOPS
FP64 (double)
237.6 GFLOPS

Graphics Processor

[Report Issues]
GPU Name
TU106
GPU Variant
TU106-100-A1
Architecture
Turing
Foundry
TSMC
Process Size
12 nm
Transistors
10,800 million
Die Size
445mm²

Board Design

[Report Issues]
TDP
185W
Suggested PSU
450W
Outputs
No outputs
Power Connectors
1x 8-pin

Graphics Features

[Report Issues]
DirectX
12 Ultimate (12_2)
OpenGL
4.6
OpenCL
3.0
Vulkan
1.3
CUDA
7.5
Shader Model
6.6

Rankings

[Report Issues]
FP32 (float)
7.926 TFLOPS
7.675 TFLOPS
NVIDIA CMP 40HX 8 GB GDDR6
7.603 TFLOPS
7.589 TFLOPS
7.516 TFLOPS
7.465 TFLOPS
Blender
NVIDIA CMP 50HX 10 GB GDDR6
1384
1316
1295
NVIDIA CMP 40HX 8 GB GDDR6
1294
1231
1229
1202

Related Comparisons

© 2024 - TopCPU.net   Contact Us Privacy Policy