Report an Error

NVIDIA GA107

NVIDIA's GA107 GPU uses the Ampere architecture and is made using a 8 nm production process at Samsung. GA107 supports DirectX 12 Ultimate (Feature Level 12_2). For GPU compute applications, OpenCL version 2.0 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 3072 shading units, 96 texture mapping units and 48 ROPs. Also included are 96 tensor cores which help improve the speed of machine learning applications. The GPU also contains 24 raytracing acceleration cores.
Further reading: Ampere Architecture Whitepaper

Graphics Processor

GPU Name
GA107
Codename
NV177
Architecture
Ampere
Foundry
Samsung
Process Size
8 nm
Transistors
unknown
Die Size
unknown
Released
Unknown

Graphics Features

DirectX
12 Ultimate (12_2)
OpenGL
4.6
OpenCL
2.0
Vulkan
1.2
CUDA
8.6
Shader Model
6.5
PureVideo HD
VP11
VDPAU
Feature Set k

Render Config

Shading Units
3072
TMUs
96
ROPs
48
SM Count
24
FP16 Units
3072
FP64 Units
64
INT32 Units
1536
Tensor Cores
96
RT Cores
24
SFUs
384
TPCs
12
GPCs
3
Tex L1 Cache
64 KB per SM
L1 Cache
128 KB per SM
L2 Cache
2048 KB
Max. TDP
90 W

All Ampere GPUs

NVIDIA GPU Architecture History

Graphics cards using the NVIDIA GA107 GPU

Name Chip Memory Shaders TMUs ROPs Base Clock Boost Clock Memory Clock
NVIDIA GeForce RTX 3050 GA107-300-A1 4 GB 2304 72 40 1545 MHz 1740 MHz 1750 MHz

GA107 GPU Notes

Speculation