As is usual with graphics cards shortly before their announcement and immediate launch, benchmarks begin to emerge from what would be production samples or versions sent for review. Therefore, the first supposed benchmarks of NVIDIA GeForce RTX 4060 Ti graphics cards have already appeared.
These benchmarks have appeared in the Geekbench database. The NVIDIA GeForce RTX 4060 Ti benchmark leak shows 15% better performance than the 3060 Ti. The NVIDIA GeForce RTX 4060 Ti graphics card is the next solution for the consumer segment to be released later this month, wanting to be a direct replacement for NVIDIA's RTX 3060 Ti at a similar price for the 8GB variant.
NVIDIA GeForce RTX 4060 Ti graphics card benchmark
The NVIDIA GeForce RTX 4060 Ti has been tested on a Z790 platform with the Intel Core processor i9-13900K. The Geekbench 5 database confirms that this is a 34 SM piece with 4252 cores and a maximum frequency of 2540 MHz. The memory is also 8 GB at 18 Gbps. It has not been possible to verify if it is an overclocked variant as some custom models tend to be, or if it is a reference model.
The GPU scored 146.170 points in the Geekbench 5 CUDA test. The GeForce RTX 3060 Ti scored between 125 and 130.000 points on average at the time, making the 4060 Ti 15% faster. But at the same time, the RTX 3070 scores close to 150.000 points, which means that the 4060 Ti could end up being a bit slower or on par with the 3070 in terms of actual gaming performance.
In this comparison it is not possible to use AMD cards, because CUDA is optimized for NVIDIA cards. Only OpenCL and Vulkan metrics can be used to compare NVIDIA and AMD cards against each other.
So it seems that NVIDIA's GeForce RTX 4060 Ti would be a small improvement over the 3060 Ti in native raster performance. The improved ray tracing, DLSS 3 and added functions such as AV1 would be the main selling points of this model compared to the previous one.
Source: Wccftech
