NVIDIA GeForce RTX 4060 Ti and RTX 4060 announced - Pricing, Specifications, and Performance

GeForce RTX 4060 Ti is expected to launch later this month at a price of USD $399, excluding VAT. Its 16 GB variant is slated for a July release and will cost $499, excluding VAT. The price for the RTX 4060 is still under consideration.

The GeForce RTX 4060 Ti will be available with either 8GB or 16GB of memory. Despite the memory differences, the two versions share identical specifications. Both cards incorporate an AD106 GPU featuring 4,352 CUDA cores and a 32MB L2 cache. NVIDIA suggests that these video cards could provide up to 22 Tflops of shading processing power and have a memory bandwidth of 288GB/s. The total board power of the 4060 Ti stands at 160W, but NVIDIA estimates the average power consumption during gaming to be about 140W.

Details about the CUDA core count are still under wraps, although it's rumored to be 4,352. NVIDIA has disclosed that the RTX 4060 Ti will have a shader compute throughput of 22 TFLOP/s, a substantial 37.5% enhancement compared to the RTX 3060 Ti. The specifications reveal a narrower 128-bit memory bus on the RTX 4060 Ti, a departure from the 256-bit memory bus on the RTX 3060 Ti and the 192-bit on the RTX 4060. The GPU's memory sub-system has seen a major overhaul with "Ada," with a greater emphasis on on-die cache, enabling the use of slimmer memory interfaces and resulting in cost savings.

The RTX 4060 Ti, in both its memory versions, has a Total Graphics Power (TGP) of 160 W. This suggests that even the faster custom-design boards will likely only need a single 8-pin PCIe power connector. The GeForce RTX 4060, with a 15% higher shader throughput than the RTX 3060 and 6% lower than the RTX 3060 Ti, has a remarkable TGP of just 115 W, making it technically feasible for cards to have a single 6-pin PCIe power connector. DLSS 3 is a key feature of the RTX 4060 series, potentially setting it apart from heavily discounted previous-generation cards. The RTX 4060 Ti shows considerable generational performance upgrades over the RTX 3060 Ti, and games utilizing DLSS 3 Frame Generation are predicted to provide nearly twice the performance. Both the RTX 4060 Ti and RTX 4060 are geared towards optimal 1080p gaming, with capacity for handling 1440p gaming at high settings.

NVIDIA has formally revealed its mid-range RTX 40 lineup, which includes three graphics cards. The RTX 4060 Ti SKU will debut in two versions, boasting either 8 GB or 16 GB VRAM. Interestingly, the mid-range RTX 4060 will have more memory than the RTX 4070 series (12 GB). As per official statistics, the RTX 4060 Ti series will offer up to 22 TFLOPS of FP32 single-precision compute performance, while the non-Ti RTX 4060 is projected to provide about 15 TFLOPS of shader performance. This data is being compared with the RTX 3060 Ti and RTX 3060 series, which deliver 16 TFLOPS and 13 TFLOPS, respectively.

The TDP for the NVIDIA GeForce RTX 4060 Ti is 160 W (contrary to previous reports of 165 W), with an estimated average power consumption of around 140 W during gaming. The RTX 4060 will launch with a TDP of 115 W, with the typical power consumption for gaming projected to be around 110 W.

According to NVIDIA, the RTX 4060 Ti demonstrates an average performance increase of 15% over the RTX 3060 Ti when employing DLSS upscaling without Frame Generation. The company states that if the AI feature for frame interpolation is utilized, the card could be 70% faster on average than the 3060 Ti. In comparison to the RTX 2060 Super released in 2019, the RTX 4060 Ti could potentially offer an average speed increase of 70% with DLSS upscaling but without Frame Generation. With Frame Generation, this performance difference could escalate to 160%.