GeForce RTX 3050 Ti e RTX 3050 it’s two o’clock new GPUs from Nvidia to expand the audience of gamers able to play the most modern titles, accelerating not only the calculations related to ray tracing in hardware but also by offering full support to technology DLSS. Coming up there is a host of models from all partners who will bring the calculation of RTX notebooks to over 140 designs, both for gamers and for content creators (Nvidia Studio).
The two new graphics chips (the GPU is an 8nm GA107) based on architecture Ampere they turn to notebook with a starting price of $ 799, capable of ensuring Full HD gaming performance with medium details and active DLSS above 60 fps, even in titles with ray tracing support (in the case of the RTX 3050 Ti, as per slide). The two new GPUs will take the place of the GTX 1650 Ti and 1650 solutions in the Nvidia solutions chessboard.
Click to enlarge
The GeForce RTX 3050 offer 2048 CUDA core, but also 64 Tensor core e 16 RT core. As for the frequencies, depending on the TGP (from 35 to 80W) the GPU reaches a Boost Clock between 1057 and 1740 MHz. 4 GB of GDDR6 memory on 128-bit bus.
The most powerful model, the RTX 3050 Ti, differs in having 2560 CUDA core, 80 Tensor core e 20 RT core, as well as for slightly different GPU operating frequencies between 1035 and 1695 MHz. The rest, as you can also see in the table, is mirrored to the 3050.
| GPU | RTX 3080 mobile | RTX 3070 mobile | RTX 3060 mobile | RTX 3050 Ti mobile | RTX 3050 mobile |
| CUDA core | 6144 | 5120 | 3840 | 2560 | 2048 |
| Tensor core | 192 | 160 | 120 | 80 | 64 |
| RT core | 48 | 40 | 30 | 20 | 16 |
| Boost Clock (MHz) | 1245 – 1710 | 1290 – 1620 | 1283 – 1703 | 1035 – 1695 | 1057 – 1740 |
| TGP (W) | 80 – 150+ | 80 – 125 | 60 – 115 | 35 – 80 | 35 – 80 |
| Memory | 16 GB GDDR6 8 GB GDDR6 |
8 GB GDDR6 | 6 GB GDDR6 | 4 GB GDDR6 | 4 GB GDDR6 |
| Bus | 256 bit | 256 bit | 192 bit | 128 bit | 128 bit |
In addition to games, the new GeForce will also find space in all those multipurpose notebooks with Studio certification, even solutions with 1440p displays. In fact, over 60 apps for creators are “accelerated” thanks to the use of the Tensor cores present in the GPU, in order to speed up some operations thanks to artificial intelligence compared to the graphics chip integrated in the processor. This means that in terms of performance, in a load with DaVinci Resolve where a Core i7-11850H with integrated GPU takes 2 hours, an RTX 3050 Ti shortens the processing time to just 17 minutes, while the RTX 3060 takes only 14.
And it is always thanks to this technology that Nvidia has expanded the capabilities of Broadcast, the app that allows you to transform your home into a recording studio, allowing you to change the background, have the camera follow your movements and more: the new features include Removal of room echo and video noise, the ability to filter out noises such as dog barking or cicadas, as well as multi-effect support.