Nvidia recently launched its GeForce RTX 4060 and 4060 Ti desktop graphics cards, updating its most popular family of desktop GPUs with the Ada Lovelace architecture and Nvidia’s next-generation Deep Learning Super Sampling 3 (DLSS 3) technology. I have written previously on how Nvidia’s AI-driven DLSS 3 can take a demanding AAA game like Cyberpunk 2077, which is pretty much unplayable at high resolutions, and make it playable. There’s value in AI-driven gaming today. I believe that the future of gaming will be dictated by AI, there will be even more benefits to come.
In this article, I want to explain the value Nvidia has created through the innovative features in the GeForce RTX 4060 and 4060 Ti and how I see these graphics cards more as AI gaming cards. Let’s dive in.
AI is eating the world
Nvidia has given gamers many reasons to upgrade to the new GeForce RTX 40 series graphics cards, including the new Ada Lovelace architecture and DLSS 3.
Among other improvements, Nvidia has reduced the VRAM traffic per frame by adding more L2 cache compared to previous generations. This allows for more data to be stored in the faster L2 cache rather than needing to move across the memory bus to the VRAM. This leads to better performance, increased power efficiency and better ray-tracing results.
The biggest difference between DLSS 3 and its predecessor, DLSS 2, is that the new generation introduces a feature that uses AI to predictively generate new frames based on the geometry and effects of past frames. This frame-generation technology significantly increases the system’s performance by mitigating most of the workload handled via traditional renderings. Nvidia says that DLSS 3 reconstructs seven-eighths of the total displayed pixels. Once more, if we think about where this frame generation happens, the more information stored in the L2 cache from the game motion vectors, the optical flow field and the sequential game frames, the better results you will get thanks to DLSS 3.
GeForce RTX 4060 Ti: Ada Plus DLSS 3. Source: Nvidia
With these new graphics cards, Nvidia is showing that the future of gaming is less about improving raw hardware performance and more about making intelligent software improvements using AI. Moore’s law—which predicts that the number of transistors on a microchip will double with every new generation of silicon—is slowing down, so Nvidia, like every other chipmaker, must answer the question of how to keep increasing hardware performance to keep up with consumer demand for ever-faster software.
It reminds me a lot of the early days of ray tracing, when it was less about making graphics cards better in terms of raw performance and more about making them better in terms of value optimization. Software features like DLSS 3 can be optimized for lower-end GPUs as well as more premium chips, whereas most of the differentiated features between the higher-end and lower-end GPUs are at the hardware level.
The biggest difference between ray tracing and DLSS 3 is that Nvidia offered ray tracing at a premium, yet now Nvidia has slashed the retail price of the GeForce RTX 4060 by $30 compared to the RTX 3060. The RTX 4060 Ti will start at $399 for the 8GB VRAM model and $499 for the 12GB VRAM model, while the RTX 4060 will start at $299.
Nvidia is not stopping at DLSS 3 either. It has also previewed future AI capabilities that use neural rendering and neural textures. These AI technologies make the rendering crisper and better looking without sacrificing performance in frames-per-second.
Nvidia has added tremendous value across its entire GeForce RTX 40 series of desktop graphics by introducing Ada Lovelace and DLSS 3. Not only do the RTX 4060 and 4060 Ti have quadruple the TFLOPS compared to the RTX 2060 and twice the TFLOPS compared to the RTX 3060, but they also leverage AI to improve performance and enable systems to perform at higher resolutions.
Implementing frame generation with AI is an ingenious way for Nvidia to improve the performance of its graphics cards. I am excited to see the innovative ways that Nvidia will continue to use AI to improve gameplay in the future.