Nvidia Launches The $349 Geforce Rtx 2060
While Nvidia didn’t share configuration information about the test system, the stated results were generated on the Arras multiplayer map. The $399 RTX 2060 Super is more in line with the previous RTX 2070 spec-wise, instead of the plain 2060. It now has 8GB of RAM, 256 more CUDA cores and a healthy base clock speed bump up to 1,470MHz instead of 1,365MHz. NVIDIA says the 2060 Super is around 15% faster than the 2060, and it almost keeps up with the original 2070. That’s a compelling amount of performance for that price, especially if you don’t mind overclocking to eke out a bit more speed. You’ll find three DisplayPort connections, an HDMI port and USB-C “VirtualLink” on both cards.
But for right now, Nvidia’s software continues to be better, accelerating GPU performance not just in games, but in traditional work tasks. That was until I got to our Blender test, in which we time how long it takes a GPU to render a 3D object in Blender. Thanks to a recent software update that gtx 1080 8gb provided better support for programs like Blender, the 2060 not only beat the Vega 56, but even the 2080 Ti, which I tested only a couple of months ago. It has been a fairly crazy ride with NVIDIA the past few months, roughly 4-5 months ago the GeForce RTX 2080 and 2080 Ti became available.
It’s cheaper than the 1070 or 1070 Ti and on par with the similarly priced AMD Vega 56. Only unlike the 56, the 2060 will support ray tracing when games finally start sticking it in. The RX 590 is, admittedly, still a lot cheaper than the (with most cards going for the same price as Nvidia’s 6GB GTX 1060 at time of writing), and offers a smidge more performance than both its GTX rival and RX 580 predecessor.
Among these contenders, it’s important to note that besides being the cheapest, the 2060 also has the least memory. It ships with 6GB of GDDR6 memory, while the 2080 ships with 8GB and the 2080Ti ships with 11GB. The 1080, being the oldest, ships with 8GB of slower GDDR5X memory, while AMD opted for 8GB of HBM2 memory, which tends to be faster than both GDDR5X and GDDR6. But that’s just what the RTX 2060 can do in terms of raw performance. Add in all of other clever RTX gubbins, such as their really very nifty DLSS support , and some of those aforementioned toughies become really quite tamable. Take Final Fantasy XV, which stuttered along at 34-39fps on Average at 4K with all of Nvidia’s extra effects turned off, but whizzed up to a much smoother 43-45fps when I turned on DLSS in its anti-aliasing menu.
TFLOPS indicates how many trillion FP32 floating point operations the graphics card can perform per second. There it is, in all its glowing green glory.The gap was even bigger in Shadow of the Tomb Raider, where the RTX 2060 was more than 10 frames ahead of the GTX 1070Ti on Highest with its SMAATx2 anti-aliasing option enabled. Here, it delivered frame rates ranging between 58-75fps in the game’s Cozumel Caves section, compared to the GTX 1070Ti’s 47-63fps. Even the GTX 1080 only managed between 56-79fps when I put together my Shadow of the Tomb Raider graphics performance guide, which is pretty damn amazing considering how much it costs. You get two DisplayPort 1.4a ready outputs on the RTX 2006, as well as one HDMI 2.0, one USB-C, and one DVI-DL.So what are you actually getting with the RTX 2060? It’s numbers time – although some of these figures may vary if you decide to get one of the third party cards instead of the Nvidia Founders Edition I’ve got here.
For months, AMD has been hyping how much faster they are than the first RTX cards. The only solution for NVIDIA was a surprise batch of faster cards, which, in turn, led to AMD announcing a price drop at the last minute. The GeForce RTX™ 2060 is powered by the NVIDIA Turing™ architecture, bringing incredible performance and the power of real-time ray tracing and AI to the latest games and to every gamer. The GeForce packs a slightly cut-down version of the TU106 GPU found inside the RTX 2070. The card has only 20 percent fewer CUDA cores than its bigger brother, but crams in a whopping 33 percent more CUDA cores than its predecessor, the GTX 1060. And extensive tweaks found in Nvidia’s Turing GPU architecture make the RTX 2060’s cores much more capable than the GTX 1060’s, as we’ll see in the benchmarks section.
It replaces the original model, and you can think of it as a slightly slower version of the RTX 2080. The 2070 Super gets 256 more CUDA cores, 16 additional texture units, and faster clock speeds . All of that hardware makes it faster than the GTX 1080 Ti, NVIDIA’s last-gen cream of the crop. NVIDIA’s new RTX Super graphics cards are clearly a response to AMD’s new midrange GPUs, the Radeon RX 5700 and 5700 XT.