Nvidia GeForce RTX 4080 12GB Model Tipped to Launch Again as the RTX 4070 Ti

Nvidia paused the launch of its 12GB RTX 4080 graphics card last month, after backlash over naming and pricing. However, it appears the recently launched GPU from the American technology firm could return under a different model number. According to details shared by a well-known leaker the "unlaunched" GPU will return as the Nvidia GeForce RTX 4070 Ti in January, as Nvidia attempts to fix the confusing naming around two RTX 4080 cards that had different specifications.

Well known tipster kopite7kimi (Twitter: @kopite7kimi) leaked details about the purported return of the 12GB Nvidia RTX 4070 GPU. "The original RTX 4080 12GB will become RTX 4070 Ti instead," they said, while replying to a question related to a January 2023 release date with a thumbs up.

The tipster previously predicted the specifications of the Nvidia RTX 3090 and RTX 3080 GPUs, and that the RTX 4090 GPU would require 450 watts of power, according to a report by The Verge.

It would be logical to see the 12GB RTX 4080 make a comeback under the RTX 4070 Ti name because the 16GB model was so different from the 12GB model. Nvidia received harsh criticism for labelling the 12GB model as an RTX 4080, which was supposed to start at $899 and include 7,680 CUDA Cores, a 2.31GHz base clock that boosts up to 2.61GHz, 639 Tensor-TFLOPs, 92 RT-TFLOPs, and 40 Shader-TFLOPs.

On the other hand, the 16GB Nvidia GeForce RTX 4080 is much more powerful, with 9,728 CUDA Cores, a base clock of 2.21GHz that boosts up to 2.51GHz, 780 Tensor-TFLOPs, 113 RT-TFLOPs, and 49 Shader-TFLOPs of power.

Back in September, Nvidia unveiled the first two GPUs in its long-awaited GeForce RTX 40 series: the GeForce RTX 4090 and GeForce RTX 4080. The new GPUs are based on the new 'Ada Lovelace' architecture. Nvidia has promised up to 4 times the performance of the previous 'Ampere' architecture, along with improved power efficiency.

The company had also announced its new DLSS3 image upscaling technology which is claimed to be able to generate entire frames independent of a PC's CPU. Nvidia says its third-generation ray tracing cores and fourth-gen Tensor cores drastically speed up ray tracing and floating-point acceleration which enable "the age of neural rendering".