Introduction

NVIDIA today introduces their GeForce RTX 4060 Ti graphics card and we're reviewing the company's in-house GeForce RTX 4060 Ti Founders Edition. The new RTX 4060 Ti is part of a series consisting of the RTX 4060 Ti we're reviewing today, and its 16 GB memory variant that comes out in July, alongside the RTX 4060 non-Ti. All three are designed to plow through the latest games at 1080p resolution, with fairly high settings. NVIDIA stops short of saying that you can max out your game with this card and its 8 GB memory, though we'll find out if you really need double the memory in July, with the 16 GB model.



The GeForce RTX 4060 Ti is based on the same Ada Lovelace graphics architecture as the rest of the RTX 40-series that NVIDIA has been rolling out in a systematic top-down manner, beginning with the enthusiast segment (RTX 4090 and RTX 4080), coming down to the performance segment (RTX 4070 Ti and RTX 4070), and now the mid-range, with the RTX 4060 Ti, RTX 4060 Ti 16 GB, and the RTX 4060.

The new GeForce Ada Lovelace graphics architecture debuts the third generation of RTX, NVIDIA's ground-breaking invention in the consumer graphics space, which combines certain real-time ray traced elements with conventional raster 3D graphics, to significantly improve visual realism. These include ray traced lighting, reflections, shadows, global-illumination, and motion-blur. Even this bit of ray tracing requires enormous compute power, and so NVIDIA has developed fixed-function hardware in the form of RT cores, which it has been generationally enhancing in performance and capabilities.

The GeForce RTX 4060 Ti debuts the new 5 nm AD106 silicon to the desktop space. This chip already powers the mobile RTX 4070 in its maxed out form. The desktop RTX 4060 Ti stops a touch short of doing so, and uses 34 out of 36 streaming multiprocessors (SM) physically present on the silicon, which work out to 4,352 CUDA cores, 34 RT cores, 136 Tensor cores, 48 ROPs, and 136 TMUs. The most fascinating aspect about the RTX 4060 Ti is its memory—you get the same 8 GB as the previous-generation RTX 3060 Ti, but across a generationally narrower memory bus width of 128-bit. We'll provide NVIDIA's explanation about why this shouldn't alarm you, in the Architecture page.

The NVIDIA GeForce RTX 4060 Ti Founders Edition retains the dual-axial flow-through design of Founders Edition graphics cards for the past couple of generations now. Fresh air is drawn in from one of the card's two fans, guided through an intricate set of heatsinks, and exhausted by the second fan. The Founders Edition card retains the modern 16-pin ATX 12VHPWR power connector, which should simplify power connections with modern PSUs. For the rest, the company includes a power adapter that converts two 8-pin PCIe connectors to one of these.

The defining feature of GeForce RTX 4060 Ti, which should sweeten the deal for you compared to buying a previous-generation graphics card, has to be DLSS 3 Frame Generation. Ada GPUs have the ability to leverage the new Optical Flow Accelerator component and its AI-acceleration capabilities, to generate entire alternate frames entirely using AI, and without involving the graphics rendering machinery. This should, in theory, double the frame-rates in games that take advantage of the feature.

NVIDIA is pricing the GeForce RTX 4060 Ti at an MSRP of USD $399, which is the same exact launch price as its predecessor, the RTX 3060 Ti. The company is encouraging its custom-design board partners to have products at the $399 MSRP, by giving those cards a earlier review NDA of today (May 23). Custom-design cards priced above this $399 price will see their reviews go live tomorrow (May 24), when both kinds of RTX 4060 Ti go on sale.

Short 5-Minute Summary of this Review

Our goal with the videos is to create short summaries, not go into all the details and test results, which can be found on the following pages of this review.