Nvidia took the wraps off the GeForce GTX 1060, its new sweet-spot graphics card. Nvidia first card based on their Polaris architecture and the mainstream-focused Radeon RX 480. Powered by TSMC’s 16nm FinFET process and GlobalFoundries’ 14nm FinFET process respectively, AMD and NVIDIA have delivered better performance for lower prices, and with lower power consumption as well.
The GTX 1060 is NVIDIA’s cheapest card built on its 16 nanometer Pascal architecture, and given its performance claims, it seems to show the performance benefits of moving to that smaller manufacturing process. It packs in 6 gigabyte of GDDR5 RAM and a boost clock speed of 1.7GHz. NVIDIA is mum on the base clock at this point, but given what we’ve seen with the past cards, it is likely a few hundred megahertz slower. The company claims it will be overclockable to 2GHz without effort (something we also saw with the 1080 and 1070 Founders cards).
One thing that might not be immediately apparent is that GTX 1060’s PCB is much shorter than that of its senior siblings, and it is only the reference cooler that extends it to the familiar length of around 10 inches. That leaves open the potential for Nvidia’s hardware partners that come up with their own, more compact cooling solutions, especially with water-cooling sys and the like.
As it stands today, Nvidia’s GTX 1060 is hard to judge because, unusually for new graphics card launches, it has not been subject to a battery of embargoed pre-announcement reviews. That leaves us with little sense of how it’s raw numbers translate into gaming performance, and so we’ll just have to wait a little longer to establish if this card is indeed a worthy rival to AMD’s RX 480. Nvidia plans a global release for the $249 GTX 1060 on July 19th, which is when it’ll also be offering a limited Founders Edition for $299 as well.