Report an Error

NVIDIA GA10B

Default
GA10B
NVIDIA's GA10B GPU uses the Ampere architecture and is made using a 8 nm production process at Samsung. GA10B supports DirectX 12 Ultimate (Feature Level 12_2). For GPU compute applications, OpenCL version 3.0 and CUDA 8.6 can be used. Additionally, the DirectX 12 Ultimate capability guarantees support for hardware-raytracing, variable-rate shading and more, in upcoming video games. It features 2048 shading units, 64 texture mapping units and 32 ROPs. Also included are 64 tensor cores which help improve the speed of machine learning applications.
Further reading: Ampere Architecture Whitepaper

Graphics Processor

Released
Feb 2023
GPU Name
GA10B
Architecture
Ampere
Foundry
Samsung
Process Size
8 nm
Transistors
unknown
Die Size
unknown

Graphics Features

DirectX
12 Ultimate (12_2)
OpenGL
4.6
OpenCL
3.0
Vulkan
1.3
CUDA
8.6
Shader Model
6.7
WDDM
3.1
Tensor Cores
3rd Gen
PureVideo HD
VP11
VDPAU
Feature Set K

Render Config

Shading Units
2048
TMUs
64
ROPs
32
SM Count
16
FP16 Units
2048
FP64 Units
64
INT32 Units
1024
Tensor Cores
64
Tex L1 Cache
64 KB per SM
L1 Cache
128 KB per SM
L2 Cache
256 KB
Max. TDP
60 W
SFUs
256

All Ampere GPUs

NVIDIA GPU Architecture History

Graphics cards using the NVIDIA GA10B GPU

Name Chip Memory Shaders TMUs ROPs GPU Clock Memory Clock
NVIDIA Jetson Orin NX 8 GB
TE980M-A1 8 GB 1024 32 16 765 MHz 1600 MHz
NVIDIA Jetson Orin Nano 8 GB
8 GB 1024 32 16 625 MHz 1067 MHz
NVIDIA Jetson Orin Nano 4 GB
4 GB 512 16 8 625 MHz 1067 MHz
NVIDIA Jetson AGX Orin 32 GB
32 GB 1792 56 24 930 MHz 1600 MHz
NVIDIA Jetson AGX Orin 64 GB
64 GB 2048 64 32 1300 MHz 1600 MHz
NVIDIA Jetson Orin NX 16 GB
TE980M-A1 16 GB 1024 32 16 918 MHz 1600 MHz

GA10B GPU Notes

Tensor Cores: 3rd Gen
PureVideo HD: VP11
VDPAU: Feature Set K
Jun 2nd, 2024 02:57 EDT change timezone

Latest GPU Drivers

New Forum Posts

Popular Reviews

Controversial News Posts