Notebook Check has shared the first benchmarks for NVIDIA’s next round of Ampere-based laptop GPUs, the GeForce RTX 3050 and GeForce RTX 3050 Ti. These GPUs feature 2,048 and 2,560 CUDA Cores, respectively, along with 4 GB of GDDR6 VRAM and TGPs ranging from 35 to 80 watts. Enthusiasts are once again advised to double check these GPUs’ specifications before buying, as their clock frequencies can vary wildly depending on subsystem power.
| Nvidia GeForce RTX 3050 laptop GPU | Nvidia GeForce RTX 3050 Ti laptop GPU | |
|---|---|---|
| CUDA arithmetic units | 2,048 | 2,560 |
| TGP | 35W 80W | 35W-80W |
| Clock frequencies (boost) | 35W: 1,057.5 MHz 50W: 1,455 MHz 60W: 1,500 MHz 80W: 1,740 MHz | 35W: 1,035 MHz 50W: 1,410 MHz 60W: 1,485 MHz 80W: 1,695 MHz |
| VRAM | 4 GB GDDR6 | 4 GB GDDR6 |
| interface | 128-bit | 128-bit |
| Bandwidth | 88 GB / S | 88 GB / S |
The memory configuration in particular suggests that both graphics chips are primarily intended for gaming in 1080p resolution. As our benchmark results show, even the faster GeForce RTX 3050 Ti does not quite achieve the performance of the GeForce RTX 2060 Max-Q , but a slightly better performance should be possible depending on the TGP.
Source: Notebook Check