NVIDIA has been experiencing great momentum ever since it launched its GTX 10series with Pascal architecture of graphics cards for PC gamers in 2016. Since then, NVIDIA’s revenue from its gaming segment has climbed 138%, and the stock price has skyrocketed 1,208%.
And a few months ago (Aug-2018), NVIDIA announced its new 2000-series of Graphics card with Turing architecture at Gamescom and the name of this series cards is Nvidia GeForce RTX.
These new graphics cards are anticipated to take graphics fidelity in video games to the new heights of realism, and could potentially place NVIDIA in a tough competitive position in the short term against its main rival AMD Vega and RX cards.
Now the exclusive question is:
Table of Contents
What makes the Nvidia RTX special?
Nvidia declared that the RTX series of cards offers “up to 6x the performance of previous-generation graphics cards“, which is quite a big claim seeing just how powerful the Pascal-powered GTX 1080 and its 1000 series was.
But the special thing about this is, that it adds innovative technologies that bring about lighting systems considered to still be years away from being practically possible and the name of this technology is RayTracing but all of it comes at a price.
The new-gen graphics cards are more expensive than their predecessors, even factoring in recent pricing dilemmas encountered by graphics card buyers all over the world. And here’s the breakdown of what these cards are and what they do for their price.
The New RTX Graphics Cards Models
That was slight of a surprise, as with past generations, Nvidia has staggered the showcase and announcement of these cards over an extended period of time.
Where in the GTX-series we saw the GTX 1080 release first, followed by the 1070 a month later, and the GTX 1080 Ti a year after that, the 2080 and 2080 Ti were initially slated to go on sale on (“or around”) Sep 20, while the 2070 may release in October 2018.
Nvidia’s new cards introduce support for both AI-powered anti-aliasing and Raytracing. Ray tracing is the important new capability with this generation. But before you even get a game installed on your PC, Nvidia’s DLSS ( Deep Learning Super-Sampling ) leverages the company’s supercomputer to scan games before they are released and work out the most effective way to render graphics.
These cards have a couple of various prices for each, with Huang (CEO of NVIDIA) claiming that third-party reference cards (that aren’t overclocked & given new custom cooling solutions) would be cheaper and somewhat lower-clocked than Founders Edition cards.
The Nvidia GPUs price at launch is $1,200 for the RTX 2080 Ti, $800 for the RTX 2080, and $600 for the RTX 2070. Third party cards will be noticeably cheaper at $1,000, $700, and $500 sequentially, but that is still much more expensive than previous generation cards have been.
Typically I would assume more mid-range cards at more fair prices to debut a couple of months after the flagship GPUs. A theoretical RTX 2050 and RTX 2060 could show up before the end of the year, but Nvidia has yet to make any official statement to such an effect.
What is RAY TRACING
What is ray tracing you may ask? That question just became far more important for PC gamers, as the new Nvidia GeForce RTX series is adding dedicated hardware to accelerate RayTracing.
The RTX 2080, 2080 Ti & RTX 2070 graphics cards are using the Nvidia’s new Turing architecture, which declares to be the most revolutionary jump in GPUs that we’ve seen in a long time—perhaps ever.
What is ray tracing?
Raytracing is a rendering method that can generate incredibly lifelike lighting effects. Basically, an algorithm can follow the path of light and then imitate the way that the light interacts with the virtual objects it ultimately hits in the computer-generated world.
We’ve seen in-game lighting effects enhance more and more realistic over the years, but the advantages of ray tracing are less about the light itself and more about how it mixes with the game world.
Ray tracing allows for dramatically more lifelike reflections and shadows, along with much-better scattering and translucence.
The algorithm takes into account where the light bumps and calculates the interaction much like the human eye would process reflections, real light, and shadows, for example. The way light hits objects in the world also influences which colors you see.
With sufficient computational power, it’s possible to create incredibly realistic CG images that are nearly equal from life. But that’s the problem: even a high end gaming PC only has so much GPU power to work with, let alone a modern game console.
Ray tracing is utilized widely when developing computer graphics imagery for movies and TV shows, but that’s because big studios can provide the power of an entire server farm (or cloud computing) to get the job done. And even then, it can be a lengthy, difficult process. Doing it on the fly has been far too demanding for existing gaming hardware.
Rather, video games use rasterization, which is a much faster way to render computer graphics. It converts the 3D graphics into 2D pixels to illustrate shadows, light, reflections etc on your screen, but rasterization then requires shaders to draw reasonably lifelike lighting effects.
The results don’t look as real or lifelike as they would with ray tracing. The gains of this technology apparently won’t seem individually mind-blowing, but the combined enhancements could really upgrade the realism of interactive game worlds.
Performance of RTX and Comparison with GTX
Nvidia’s next generation of graphics cards comes with huge promises. The RTX 2080 is assumed to be up to 73 percent faster than the GTX 1080 in some games, and it’s designed to finally fulfill the ultimate goal of 4K gaming at 60 fps.
In terms of raw horsepower:
- the RTX 2070 is roughly comparable to the GTX 1080;
- the RTX 2080 goes toe to toe with the GTX 1080 Ti;
- the RTX 2080 Ti is in a class of its own.
- The RTX 2070 and 2080 have 8GB of GDDR6 RAM;
- the 2080 Ti has 11GB RAM.
- And all three are based on the company’s new Turing architecture.
The RTX 2080 Ti Founders Edition can handle 14.2 TFLOPS ( trillion floating-point operations per second ), while the Turing TU102-Chip these new cards are based on pushes that count up to 16.3 TFLOPS. That’s accomplished through a mix of higher clock speeds and increased CUDA cores (the 2080 Ti has 4,352, the fully configured TU102 has 4,608.)
Nvidia RTX 2080 Ti vs GTX 1080 Ti – Performance
While we’ve not been able to check out the RTX 2080 Ti ourselves and say definitively whether or not you should throw down a grand or more for an RTX 2080 Ti, we can share some benchmarking results, presented by Nvidia.
Nvidia’s test rig for both graphics cards was the same – here’s a quick breakdown of what they used, plus frames per second (FPS) achieved on a number of DirectX 11 and DirectX 12 titles:
Processor: Intel Core i9-7900X 3.3GHz CPU
Motherboard: Asus X299 Rampage VI Apex
OS: Windows 10 (v1803) 64-bit
RAM: 16GB Corsair DDR4
Nvidia drivers: 411.38
Last update on 2018-12-15 at 22:28 / Affiliate links / Images from Amazon Product Advertising API
All these games were tested on high-end Monitor at 4K resolution Ultra setting. But if you play these games on 1440p & 1080p resolution you will see a huge increase in FPS depending on if you play games on a $200 gaming monitor or a high-end HDR monitor. But one thing I can say for sure is that in 1080p or 1440p gaming you will easily get 100+ FPS in most games with the RTX graphics card.
Ultimately, the RTX is around 20-30 percent faster than their last-gen cards, but in many experiments that gap can be as little as single-digit percentage scores, so performance enhancements will depend on the software being run.
Note that we’re expected to see some new motherboards hitting the market in the upcoming months which should be able to push frame rates even further. Intel’s rumored to be unveiling its 9th gen processors next month, so your brand-new 9th Gen CPU will be (you’d hope) significantly greater than a 7th Gen Core i9.
We don’t know if Nvidia overclocked its testing rig i9-7900X processor to achieve any of these scores, or if these were done with the 3.3GHz base clock.
This would have been impressive to see, as Nvidia says that compared to the 7-phase dual-FET power supply used on the GTX 1080 Ti Founders Edition, the 2080 Ti equivalent uses a new 13-phase iMON DrMOS power supply.
Nvidia says that this can dynamically adapt power flow so the power supply can run at peak efficiency during low load circumstances, standard gaming, and highest load when the GPU is overclocked. It’s also allegedly better at providing power to the GPU, as less energy is wasted as heat.
AI and Denoising
Here’s where Nvidia’s Turing architecture truly gets clever. As if the RT cores and enhanced CUDA cores aren’t enough, Turing has Tensor cores that can dramatically quicken machine learning computations.
In FP16 workloads, the RTX 2080 Ti FE’s Tensor cores work at 114 TFLOPS, compared to just 14 TFLOPS of FP32 on the CUDA cores. That’s fundamentally like ten GTX 1080 Ti cards waiting to crunch numbers.
These RT cores will work beside Turing’s Tensor cores, which usage AI “trained by supercomputers” to fill in the voids using a technique known as denoising — dramatically a new form of advanced anti-aliasing.
Huang also addressed the possibility of foveated rendering, which could help make virtual reality titles less hardware-intensive by directing the processing power where the gamer is looking and rendering everything in their peripherals at a lower quality level.
These latest technologies at the heart of the RTX-series architecture indicate that certain games will be capable to leverage real-time reflections & advanced anti-aliasing techniques like never before.
Noise, Cooling and Ports
While the hardware inside the RTX-series cards has improved, the outside has received a major overhaul too. After decades of a single fan, reference graphics cards, Nvidia has added dual fan setup to its design. It’s created a much more insular blower set up, no doubt to ensure that temperatures remain steady with the higher power demands of the new RTX series graphics cards.
Huang said that this is also much quieter graphics cards too. Even when entirely overclocked, we’re told that temperatures stay consistent and noise at a satisfactory level.
At the back end where you’ll attach up your display cable, the RTX-series offers a few port options. Alongside more standard DisplayPort 1.4 and HDMI 2.0b connectors, there’s also a virtual link; a USB-C-shaped port that is created to provide power and video to virtual reality headsets.
Ray tracing is unlikely to be utilized heavily by developers until there is a high enough number of RTX-capable cards out there and we still don’t actually know how capable other GPUs like the 10-series or AMD RX and Vega cards will be at managing it themselves. It may be that those cards still fight with the 2000-series in games without ray tracing, which will be the majority for years to come.
Image Sources: Nvidia Geforce