Home NVIDIA RTX A2000 Embedded

NVIDIA RTX A2000 Embedded

This is a GPU manufactured with Samsung 8nm process, based on Nvidia Ampere architecture. It features 2560 CUDA cores and 8GB GDDR6 memory, with 2MB L2 cache, theoretical performance of 8.253TFLOPS, with total power consumption of 60W.

Graphics Card

[Report Issues]
Release Date
Unknown
Generation
Quadro Mobile
Type
Professional
Bus Interface
PCIe 4.0 x16

Clock Speeds

[Report Issues]
Base Clock
1117MHz
Boost Clock
1612MHz
Memory Clock
1500MHz

Memory

[Report Issues]
Memory Size
8GB
Memory Type
GDDR6
Memory Bus
128bit
Bandwidth
192.0 GB/s

Render Config

[Report Issues]
SM Count
20
Shading Units
2560
TMUs
80
ROPs
48
Tensor Cores
80
RT Cores
20
L1 Cache
128 KB (per SM)
L2 Cache
2MB

Theoretical Performance

[Report Issues]
Pixel Rate
77.38 GPixel/s
Texture Rate
129.0 GTexel/s
FP16 (half)
8.253 TFLOPS
FP32 (float)
8.253 TFLOPS
FP64 (double)
129.0 GFLOPS

Graphics Processor

[Report Issues]
GPU Name
GA107S
Architecture
Ampere
Foundry
Samsung
Process Size
8 nm
Transistors
Unknown
Die Size
Unknown

Board Design

[Report Issues]
TDP
60W
Outputs
No outputs
Power Connectors
None

Graphics Features

[Report Issues]
DirectX
12 Ultimate (12_2)
OpenGL
4.6
OpenCL
3.0
Vulkan
1.3
CUDA
8.6
Shader Model
6.6

Rankings

[Report Issues]
FP32 (float)
8.253 TFLOPS
8.218 TFLOPS
8.192 TFLOPS
8.192 TFLOPS
8.192 TFLOPS
8.186 TFLOPS
NVIDIA Tesla T4 16 GB GDDR6
8.141 TFLOPS
Blender
1980
Intel Arc A580 8 GB GDDR6
1929
1919
1910

Related Comparisons

© 2024 - TopCPU.net   Contact Us Privacy Policy