The RTX 4060 Ti’s specifications were released by kopite7kim a few days ago, stating that it uses the AD-106 GPU with 4352 CUDA cores and is equipped with 8GB of 18Gbps GDDR6 memory with an estimated memory bit width of 128bit. The L2 cache of the GPU was increased to 32MB and the power consumption of the card is 160W.

For comparison, Nvidia’s current RTX 3070 graphics card has 5888 GPU CUDA cores and 8GB of 256-bit GDDR6 memory, consuming 220W. The RTX 3060 Ti has 4864 CUDA cores and 8GB of 256-bit GDDR6 or GDDR6X memory, consuming 200W.

The RTX 4060 Ti has fewer CUDAs compared to the RTX 3060 Ti, and the memory bit width is cut in half. With the new architecture and new process, its raster performance should be able to reach RTX 3070 level. In DLSS 3 frame generation technology, RTX 4060 Ti performance is expected to be a bit higher.

NVIDIA just released the RTX 4070 Ti models at CES and is currently preparing RTX 4070 models, RTX 4060 series may have to wait until the release of the Taipei Computer Show in the middle of this year.

The post NVIDIA RTX 4060 Ti graphics card performance for RTX 3070 level appeared first on TechGoing.