Introduction

PNY's GeForce RTX 4060 XLR8 is a premium custom-design graphics card based on NVIDIA's latest mid-range GPU that's bound to sell in heaps. The XLR8 (pronounced: accelerate) tops the RTX 4060 with a large cooling solution designed to keep fan noise low when gaming; and while it sticks to NVIDIA-reference clock-speeds, you have headroom to overclock the card on your own. PNY is trying to undercut other premium custom-design RTX 4060 cards by offering their card at close-to-MSRP pricing, while giving you some of the features you expect from premium custom-designs.

The new GeForce RTX 4060 targets the vast majority of gamers that still play at 1080p resolution. It succeeds a long line of bestselling graphics cards by NVIDIA, including the RTX 3060, RTX 2060, and the GTX 1060, which have topped Steam Hardware Survey charts for popularity. Full HD remains a reasonable resolution for PC gamers to experience the eye-candy of today's AAA games, and so NVIDIA's job with the RTX 4060 is to provide them with a card that they can max out their games at 1080p, including with ray tracing, and take advantage of next-generation features.



The GeForce RTX 4060 is firmly a new-generation graphics card that's based on the Ada Lovelace graphics architecture. It has two big things going for it that make it preferable to previous-generation graphics cards. Firstly, it gets the full RTX 40-series feature-set, which means support for DLSS 3 Frame Generation, something that nearly doubles frame-rates on supported games by drawing entire frames using AI, and without involving the graphics rendering machinery; and secondly, since the RTX 4060 is manufactured on the 5 nm foundry process, it comes with the lowest typical gaming power figures for any graphics card from this generation, rated at just 115 W.

The GeForce RTX 4060 maxes out the 5 nm AD107 silicon that it is based on. In a surprising move, NVIDIA is using its smallest ASIC from this generation to make the RTX 4060, while its sibling, the RTX 4060 Ti, nearly maxes out the larger AD106 silicon. Armed with 24 streaming multiprocessors (SM), the RTX 4060 gets 3,072 CUDA cores, 96 Tensor cores, 24 RT cores, and 96 TMUs. The AD107 features 48 ROPs, and the same 128-bit GDDR6 memory interface as the AD106, which on the RTX 4060 is wired to 8 GB of video memory. Generationally, this is a 50% reduction in memory size and bus width, but NVIDIA assures us that the 8 times larger on-die cache makes up for it, by reducing video memory round-trips by 40% to 60% in NVIDIA's testing.

PNY is pricing the GeForce RTX 4060 XLR8 OC at a reasonable $320, or just $20 more than the NVIDIA MSRP. For the premium, you're offered a more capable cooling solution that's designed to run quieter than most MSRP cards, although the card does not come with any factory overclock. The GPU runs at the NVIDIA reference 2460 MHz, and the memory is untouched at 17 Gbps. It will be interesting to find out what the extra $20 is worth, having seen some of the MSRP cards that even offer overclocked speeds.

Short 10-Minute Video Comparing 10x RTX 4060

Our goal with the videos is to create short summaries, not go into all the details and test results, which can be found in our written reviews.