Nvidia will not arrange regular evaluations for RTX 4060 Ti 16GB, and partners have no intention of providing samples
In May of this year, NVIDIA unveiled a new generation of mainstream GPUs based on the Ada Lovelace architecture, including the GeForce RTX 4060, RTX 4060 Ti 8GB, and RTX 4060 Ti 16GB. Their respective Manufacturer’s Suggested Retail Prices (MSRPs) are $299, $399, and $499. The GeForce RTX 4060 and RTX 4060 Ti 8GB have already hit the market, while the RTX 4060 Ti 16GB is rumored to be set for release on July 18th.
NVIDIA did not prepare a GeForce RTX 4060 Ti 16GB Founder Edition for this launch, instead fully entrusting the custom, non-public products to its partners. However, it was previously noted that few manufacturers chose to customize this model, and their marketing enthusiasm was rather low. In fact, official attention to this GPU has been minimal; VideoCardz reports that NVIDIA did not arrange the usual review lifting process for the new graphics card.
Many manufacturers are not planning to provide samples for review and are even uncertain when they will introduce corresponding graphics cards. It is anticipated that the model and supply volume of the GeForce RTX 4060 Ti 16GB will be limited, a tepid approach reminiscent of the situation when the previous generation RTX 3080 12GB was launched. Players interested in purchasing the RTX 4060 Ti 16GB may find a few product options available at launch.
The GeForce RTX 4060 Ti 16GB is equipped with an AD106-351 GPU and uses a PCIe 4.0 x8 interface. The full version of the chip has 36 SM groups, 4608 CUDA cores, 36 RT cores, and 144 Tensor cores. However, the RTX 4060 Ti only enables 34 of these SM groups, resulting in 4352 CUDA cores, 34 RT cores, and 136 Tensor cores. In addition, it includes an 8th-generation NVENC encoder and a 5th-generation NVDEC decoder, supporting AV1 hardware encoding and decoding.
The new product has a base clock of 2310 MHz, and a boost clock of 2535 MHz, and is paired with 16GB of GDDR6 memory. The memory has a 128-bit width, a memory speed of 18Gbps, a memory bandwidth of 288 GB/s, and a total card power consumption of 165W, an increase of 5W from the 8GB version. NVIDIA states that, with the aid of a 32MB L2 cache, the effective memory bandwidth is boosted to 554 GB/s.
