New GeForce RTX 3050 Mobile Graphics Card with 1792 CUDA Cores and 4GB of VRAM

NVIDIA recently confirmed the launch of the GeForce RTX 3050 A, a mobile graphics card that promises impressive specs with 1,792 CUDA cores and 4GB of memory. This Ada-based model raises questions about its performance compared to its predecessors. Learn about the fascinating details of this new GPU and its potential impact on laptop performance.

NVIDIA RTX 3050 A, New Model with AD106 ADA GPU

The company confirmed with Notebookcheck that this GPU is indeed in development.

To follow up on this article about the supposed GeForce RTX 3000 model based on Ada Lovelace: It turns out that NVIDIA is planning to integrate the Ada GPU into its previous lineup. However, while the company has confirmed the initial specifications, it has not yet clarified that this is based on the Ada architecture.

To summarize: the RTX 3050 A (that's its name) is not a desktop GPU, but a mobile version. This model comes after the RTX 2050, which was based on the Ampere GPU, while the rest of the RTX 2000 series was based on Turing.

The new version, the RTX 3050 A (laptop GPU), features 1,792 CUDA cores, NVIDIA confirmed via NotebookCheck. This card should also have 4GB of memory on a 64-bit memory bus and a configurable TGP between 35 and 50 watts.

Update 2024/07/26 12:08 CET: NVIDIA has confirmed that the GeForce RTX 3050 (codenamed GeForce RTX 3050 A) is here and will have the following specs: 4GB of VRAM on a 64-bit bus, a TDP of 35-50W, and 1,792 CUDA cores.

– Check the notebook

Compared to other 3050 models, this new model has fewer cores and a shorter memory bus. However, in the absence of memory speed information, the bandwidth may not be noticeably lower if NVIDIA opts for faster memory.

Compared to the RTX 2050, the specs are similar in terms of memory. However, the “ancestor” offers more memory. Given the evolution of the architecture from Ampere to Ada, the new 3050A could have similar performance, depending on the workload. This will of course require independent testing, but the launch seems to be limited in scope. NVIDIA has not even made a public announcement about this. In fact, no laptop manufacturer has offered this GPU option yet, despite the driver being released more than 10 days ago.

Source: Checkbook