Last Updated on
The leaker @kopite7kimi posted a tweet showing what could be the specs of one of the most anticipated GPUs from Nvidia, the RTX 4060. According to the leaker, the RTX 4060 graphics card is expected to have a lower core count than its predecessor, the RTX 3060, with just 3,072 CUDA cores compared to the 3,584 on the previous model.
It will also feature fewer Tensor cores (96 compared to 112 on the 3060) and raytracing cores (24 versus 28). In fact, the core count is 17% lower than that of the RTX 3060. However, the upgrade from Ampere to Ada Lovelace architecture delivers some performance enhancements, such as access to DLSS 3.
|Ray tracing cores
In addition to a lower core count, the RTX 4060 is also expected to come with less VRAM than its Ampere counterpart, with 8GB of 18Gbps GDDR6 memory compared to the 12GB on the RTX 3060, although the memory is clocked faster. Nevertheless, the GPU will receive a significant boost in the form of a much-needed L2 cache upgrade.
This upgrade is massive, with the RTX 4060 reportedly featuring 24MB of L2 cache, a 700% increase from Ampere, while the RTX 3060 only has 3MB. However, it’s worth noting that this is still only half the L2 cache found on the RTX 4070 Ti. Despite these changes, the RTX 4060 is still expected to provide a solid gaming performance at an affordable price point.
Will RTX 4060 be better than 3070?
Despite rumors that the RTX 4060 will have lower specs than its predecessor, the RTX 3060, the new Ada architecture, and the increased L2 cache are expected to give the GPU a significant performance boost. Public benchmarks have shown that the RTX 4060 is 27% faster than the RTX 3060, making it a stronger option than the entry-level Ampere GPUs.
In fact, some benchmarks even suggest that the RTX 4060 could outperform the RTX 3070, a higher-end GPU in the previous generation. Therefore, despite the lower core count and VRAM, the RTX 4060 has the potential to provide a great gaming experience at a more affordable price point.