A few days ago, the RTX 4060 Ti 16GB version was officially put on sale, with an overseas recommended price of US $499 and a domestic price of 3899 yuan. some AIC manufacturers launched the same products as the recommended price, but of course some expensive ones, such as the RTX 4060 Ti 16GB Raptor reached 4799 yuan, which was the same as the recommended price of RTX 4070. As for its performance, I can only say that it is a bit difficult to comment.

On the recent MSI Insider program, MSI officially conducted live tests on GeForce RTX 4060 Ti 16GB, including "Cyberpunk 2077", "F1 23", "Rainbow 6: siege", "Fortnite", "CS:GO" and "Legacy of Hogwarts", as well as popular online games and large-scale 3A games.

As a result, the big memory does not bring any advantages, in addition to "CS:GO" has a more obvious lead, in several other games, although the gap is very small, but its performance is not even as good as the 8GB version.I have also seen some other measured results on the Internet, except for games such as "biochemical Evil 4: remake" and "Part1, the Last Survivor", which basically can not be separated from the 8GB version.

In fact, this result is also expected. After all, compared with the 8GB version, the other core parameters of the 16GB version remain unchanged except for the doubling of the apparent memory capacity, and the performance is still subject to parameters such as core size and explicit memory bit width.

As for the possibility of insufficient memory in high resolution, I think the probability is very small, because under the ultra-high resolution + high picture quality setting, the bottleneck of performance will not be on the video memory.

The card is clearly aimed at users in a number of professional areas, and Nvidia has previously said that in addition to games, GeForce RTX graphics cards are also used worldwide in 3D animation, video editing, motion graphics, photography, graphic design, architectural visualization, STEM, live streaming and AI.

Some of the applications used in these industries may benefit from additional large graphics memory. For example, when editing a 4K or 8K timeline in Premiere, or making a large building scene in the D5 renderer. For gamers, there is absolutely no need to spend more money on these obvious memories.

Let's briefly talk about the RTX 4060 Ti graphics card itself. Compared with RTX 3060 Ti, its theoretical performance is improved by about 10%, which is close to the level of RTX 3070.Let's briefly talk about the RTX 4060 Ti graphics card itself. Compared with RTX 3060 Ti, its theoretical performance is improved by about 10%, which is close to the level of RTX 3070.

On the one hand, the core specification is cut down, RTX 3060 Ti uses the GA 104core, and to the RTX 4060 Ti uses the AD 106core; on the other hand, it is obvious that the RTX 3060 Ti has the 256bit bit width, while the RTX 4060 Ti only has the 128bit.

On the one hand, the core specification is cut down, RTX 3060 Ti uses the GA 104core, and to the RTX 4060 Ti uses the AD 106core; on the other hand, it is obvious that the RTX 3060 Ti has the 256bit bit width, while the RTX 4060 Ti only has the 128bit.

In short, Nvidia does not want RTX 40 series to have a substantial upgrade compared with RTX 30 series products of the same level, there is that kind of leapfrog upgrade, RTX 30 series and 40 series products will be in a state of complementary plug-in, forming a richer product line for consumers to choose.

Finally, this big memory version is not for the vast number of gamers, but will be useful for professional creative users. And my ideal large memory version, should be 12GB memory, 192bit memory bit wide, more friendly to gamers, but also more expensive money.