GPU Wiki Architecture

Ampere

Architecture notes pending source review.

Architecture Overview

Architecture notes pending source review.

Architecture name: Ampere
Launch era / years active: Pending source review.
Predecessor: Pending source review.
Successor: Pending source review.
Process nodes: Pending source review.
Important chips: Pending source review.
Memory technologies: Pending source review.
CUDA / RT / Tensor generation: Pending source review.
Consumer series: DRIVE, GeForce MX Series, GeForce RTX 20 Series, GeForce RTX 30 Series, GeForce RTX 40 Series, GRID, Mining GPUs, Nintendo Switch GPU, NVIDIA A-Series, NVIDIA RTX A Series, Tegra
Workstation / professional series: Pending source review.
Data center series: Pending source review.
Source review status: Pending source review.

Sort by

Cards Table

Reset filters

GPU	Architecture	Launch Date	CUDA Cores	Memory	Bus Width	TDP/TBP	Segment
GeForce RTX 3080	Ampere	2020-09-01	8,704	10 GB GDDR6X	320-bit	320 W	Desktop
GeForce RTX 3090	Ampere	2020-09-01	10,496	24 GB GDDR6X	384-bit	350 W	Desktop
GeForce RTX 4010	Ampere	2024-04-16	768	4 GB GDDR6	64-bit	50 W	Desktop
GeForce MX570	Ampere	2021-12-17	2,048	2 GB or 4 GB depending on OEM implementation GDDR6	64-bit	15 to 45 W, OEM dependent	Desktop
GeForce RTX 3050 6 GB	Ampere	2024-02-02	2,304	6 GB GDDR6	96-bit	70 W	Desktop
GeForce RTX 3090 Ti	Ampere	2022-01-27	10,752	24 GB GDDR6X	384-bit	450 W	Desktop
GeForce RTX 3080 12 GB	Ampere	2022-01-11	8,960	12 GB GDDR6X	384-bit	350 W	Desktop
GeForce RTX 3050 8 GB	Ampere	2022-01-04	2,560	8 GB GDDR6	128-bit	130 W	Desktop
GeForce RTX 3050 OEM	Ampere	2022-01-04	2,304	8 GB GDDR6	128-bit	130 W	Desktop
GeForce RTX 3080 Ti 20 GB	Ampere	2022-01-01	10,240	20 GB GDDR6X	320-bit	350 W	Desktop