NVIDIA’s GeForce RTX 4060 Ti and RTX 4060 should be on the market as early as around May – Economical and efficient

Our colleagues from WCCFTECH had already reported that the GeForce RTX 4060 Ti AD106-350 and RTX 4060 AD107-400 GPUs with a power consumption of up to 115W are to be released in May. It is expected that these cards will hit the market after the release of the RTX 4070, which is scheduled for release on April 13 this year.  It is reported that the PCBs (Printed Circuit Boards) of the two cards will have some additional details. The NVIDIA GeForce RTX 4060 Ti is powered by an improved version of the Ada GPU architecture, which promises higher performance and efficiency. It is also rumored that the card will come with 8GB of GDDR6 memory, making it a potentially powerful choice for gamers and content creators in the lower mid-range.

The Ti variant is also expected to come with an improved cooling system to ensure that the card stays cool even during intensive use. This is an important factor to ensure that the card can also deliver high performance permanently and under load (Boost Steps). Another feature expected on the NVIDIA GeForce RTX 4060 Ti is support for ray tracing and AI-based technologies like DLSS (Deep Learning Super Sampling). These technologies significantly improve the graphics quality of games and other applications, making them more realistic overall. 

Possible NVIDIA GeForce RTX 4060 Ti specifications (rumor)

The CUDA cores are one of the most important components of the graphics card and enable the GPU to process data and execute tasks. With 4352 CUDA cores, the NVIDIA GeForce RTX 4060 Ti will be able to perform powerful calculations and deliver high-end gaming and VR experiences. The graphics card will also have an 8 GB GDDR6 memory, which will be driven via a 128-bit bus interface at 18 Gbit/s. This means that the card will offer an impressive 288 GB/s bandwidth, which is essential for fast and efficient loading of textures and graphics in games.

The GDDR6 memory is also faster and more energy-efficient than its GDDR5 predecessor. In addition, the NVIDIA GeForce RTX 4060 Ti comes with 32 MB of L2 cache on board the GPU, which is an 8x increase over the GeForce RTX 3060 Ti. The L2 cache is a fast memory area that will help the graphics card store and access frequently used data quickly to further improve performance.

The graphics card is expected to use the PG190 SKU 361 PCB. Leaked PCB photos revealed that the Founders Edition uses the standard 16-pin 12VHPWR connector, while custom models are once again free to use both standard 8-pin connectors and the newer 16-pin connectors. It is rumored that in the end it is still much cheaper to use only 8-pin connectors. As for TBP, the card will probably be between 150-160W, which is about 25% less than its predecessor, the RTX 3060 Ti.

Maybe, RTX 4060 Ti. pic.twitter.com/uA6w5u44CS

— Elysian Realm (@KittyYYuko) March 16, 2023

For the GeForce RTX 4060, NVIDIA is expected to use the AD107-400-A1 GPU, which is the fully functional AD107 Ada Die. Interestingly, the same configuration is also used by the GeForce RTX 4060 laptop GPU. The chip is expected to have 3072 CUDA cores and 8 GB of GDDR6 memory running at 18 Gbps over a 128-bit bus interface. This gives the card a bandwidth of 288 GB/s and all this with a reference TDP of 110 – 115 W. What also speaks for the notebook genes is the electrical connection of the PCIe port, which only uses 8 lanes instead of the 16.

Furthermore, the GPU is supposed to have 24 MB of cache (L2). The price of the NVIDIA GeForce RTX 4060 should be 35% lower than that of its predecessor, the RTX 3060 12 GB. While previous rumors suggested that the card would use a PG190 PCB, our information and that of WCCFTECH suggests that it will use the PG173 SKU 371 PCB design instead. Probably the PG190 variant is used by the FE and the PG173 by the reference AIC boards.

Like the RTX 4060 Ti, the non-Ti variant will also make use of the standard 8-pin connectors for the most part, and we can certainly expect to see some nice mini-ITX variants of the card as well, which will make for some really efficient micro-sized gaming PCs. Using the rumored “reference” boost frequency of up to 2535 MHz, the NVIDIA GeForce RTX 4060 Ti should offer a computational performance of around 22 TFLOPs, with the AIB models reaching a performance of almost 24 TFLOPs.

The 3070 Ti was priced at $599, so the 4060 Ti would have to be $399 to be a worthy successor to the 3060 Ti, which had the same MSRP. The biggest problem is that while the card performs well at 1080p, it becomes a liability at higher resolutions due to the weak 128-bit bus interface and the use of the slower G6 memory instead of the G6X memory. The card would represent an extremely efficient design in the 150W TDP range, but the rest remains to be seen until the the the card is launched and the reviews appear.

NVIDIA GeForce RTX 40 graphics card lineup (via WCCFTECH)

Source: WccfTech Twitter