According to preliminary specs for Nvidia’s performance mainstream GeForce RTX 4060 Ti graphics card, the board doesn’t draw as much power as the company’s high-end offerings and features a short PCB that fits nicely in compact PCs. On the other hand, while this product may be among the best graphics cards, its performance and specs may be lower than some would like.
hardware leaker @Kopite7 you Late Tuesday, Nvidia’s GeForce RTX 4060 Ti was said to be based on an AD106 graphics processor with 4352 CUDA cores paired with 8 GB of 18 Gbps GDDR6 memory using a 128-bit interface. The power consumption of this product is around 220W, which is not much by today’s standards. The reference card uses a “very short” printed circuit board, so it can be easily installed in a compact PC.
The GeForce RTX 4060 Ti Founders Edition, on the other hand, continues to use the infamous 12VHPWR power connector. Why would you want a 16-pin connector that could potentially deliver 600W to a 220W part?
Nvidia GeForce RTX 40 Series Specifications
| row 0 – cell 0 | GPUs | FP32 CUDA core | memory configuration | TBP |
| GeForce RTX 4090 | AD102 | 16384 | 24GB 384bit 21GT/s GDDR6X | 450W |
| GeForce RTX 4080 | AD103 | 9728 | 16GB 256bit 22.4GT/s GDDR6X | 320W |
| GeForce RTX 4070 Ti | AD104 | 7680 | 12GB 192bit 21GT/s GDDR6X | 285W |
| GeForce RTX 4070 | AD104 | 5888 | 12GB 192bit 21GT/s GDDR6X | 250W |
| GeForce RTX 4060 Ti | AD106 | 4352 | 8GB 128bit 18GT/s GDDR6 | 220W |
Nvidia has yet to confirm specs for its upcoming products, so take them with a grain of salt. The gap between 4070 seems to be quite large. Assuming more or less similar clocks, we’re talking about at least 26% lower compute performance from the 4060 Ti. This is a good deal.
Considering the Ada Lovelace architecture improvements over the Ampere architecture, and the higher clocks made possible by TSMC’s 4N manufacturing technology, the AD106 GPU and GeForce RTX 4060 Ti board deliver significantly more compute power than the GeForce RTX 3060 Ti. You can expect it to be ready. However, the memory bandwidth of 288 GB/s is a whopping 41% lower compared to its predecessor. 32MB of L2 cache should help mitigate the difference, but the problem is that the new GeForce RTX 4060 Ti can rule out cases where it lags behind the GeForce RTX 3060 Ti with high resolution and/or anti-aliasing enabled. whether or not
To make matters worse, even the regular GeForce RTX 3060 has 4GB more memory than the upcoming GeForce RTX 4060 Ti, which is a bit strange. Given the 128-bit memory bus and relatively few of his CUDA cores, we speculate that Nvidia designed his AD106 primarily with laptops in mind. Anyway, I felt his 8 GB of VRAM on the 3060 Ti, 3070 and 3070 Ti was becoming the limiting factor. That was two years ago. A mainstream performance card that launches in 2023 and still uses 8GB of memory almost feels like an insult to gamers.
The use of AD106 in the GeForce RTX x060 Ti parts could of course increase the company’s profits, but from a gamer’s point of view, the performance and playability of upcoming games matter. The 4060 Ti is expected to easily surpass his existing 3060 Ti thanks to significantly higher clocks, but we’ll have to see how it holds up to other GPUs after launch.
Of course, this is all unofficial, preliminary information and may not be accurate or final. may become. But for now, the GeForce RTX 4060 Ti looks pretty pale and could only match his existing RTX 3070 Ti.
