NVIDIA has now introduced the new members of its family based on the ada lovelace architectureor what is the same, the series of NVIDIA GeForce RTX 40 GPUs. Specifically, we will talk about the new NVIDIA GeForce RTX 4060 Ti in its two variantss, the model with 8 GB of VRAM and the most capable model with 16 GB.
One of the enhancements to the Ada Lovelace architecture is the new memory subsystem, which significantly expands the capacity of the L2 cache to significantly improve VRAM access speeds over the previous generation.
This design reduces traffic across the memory bus and improves performance and power efficiency.
Both variants share specifications beyond the memory difference, and support all the technologies of the RTX 40 such as DLSS 3, (Deep Learning Super Sampling) that now reaches the medium, medium-high range, with its ability to generate intermediate frames through Artificial Intelligence acceleration together with DLSS’s own rescaling to improve performance.
Although this technology was already available in the RTX 4070, RTX 4080 and RTX 4090 and allows them to multiply FPS in games with high quality and effects such as Raytracing, it is possibly in this type of range that the RTX 4060 Ti and its younger sister represent. the RTX 4060, where you can get much more out of it to be able to play demanding titles without the need for such graphic demands. In fact, we have up to 1.7 times more performance when using DLSS 3, compared to conventional rendering at native resolution, while DLSS 2 stays at 1.15x more performance in the case of an RTX 3060 Ti.
With the release of the 16 GB memory version of the RTX 4060 Ti, it is striking that the RTX 4070 only offers 12 GB of memory. The reason is none other than, since the data bus of the 4070 is 192 bits, the next jump in memory capacity would be 24 GB, a very high figure for a graphics card of this range. In the case of the RTX 4060 Ti, with a 128-bit data bus, you can add 8 or 16 GB of memory, and that extra will allow better performance in games with large textures or high resolutions.
Technical characteristics of the NVIDIA GeForce RTX 4060 Ti
| RTX 4060Ti | RTX 4060 | RTX 3060Ti | RTX 3060 | |
|---|---|---|---|---|
| CUDA Core Shaders | 4,352.22 TFLOPS | 15 TFLOPs | 4,864.16 TFLOPS | 3,594.13 TFLOPS |
| MHz | 2,310 Base, 2,535 Boost | ? | 1,410 base, 1670 Boost | 1,320 base, 1,780 boost. |
| RT Cores | 34, 3rd Gen, 51 TFLOPS | 3rd Gen, 35 TFLOPS | 38, 2nd Gen, 32 TFLOPS | 28, 2nd Gen 25 TFLOPS |
| Tensor Cores | 136, 4th Gen, 353 TFLOPS | 4th Gen, 242 TFLOPS | 152, 3rd Gen, 130 TFLOPS | 112, 3rd Gen 102 TFLOPS |
| DLSS | DLSS 3 | DLSS 3 | DLSS 2 | DLSS 2 |
| NV Encoder | 8th Gen with AV1 | 8th Gen with AV1 | 7th Gen | 7th Gen |
| VRAM memory |
|
|||
| Cache and Bandwidth |
|
|
||
| TGP | 160W | 115W | 200W | 170W |
| Average Consumption at stake | 140W | 110W | 197W | 170W |
| Consumption at rest | 7W | 7W | 12W | 8W |
| Launch Price |
|
335 euros |
399 dollars |
329 dollars |
The NVIDIA GeForce RTX 4060 Ti will arrive in both custom designs by major builders, as well as a Founders Edition version of NVIDIA with its signature dual-fan design, one on each side of the card.
The price of the RTX 4060 Ti with 8 GB of memory will be $399 in the international market and will hit stores on May 24, while the RTX 4060 Ti model with 16 GB of memory will be $499 and will arrive to the market later, in the month of July.
End of Article. Tell us something in the Comments!
