There’s good reason for the excitement, but I appreciate a lot of you are console gamers or filthy casuals not as invested in PC gaming culture as others, and may not understand what’s caused the excitement – hence, this piece going into a little more detail about some of the reported capabilities of the new cards.
The RTX in the series name stands for “Ray Tracing”, which Wikipedia describes as “a rendering technique for generating an image by tracing the path of light as pixels in an image plane and simulating the effects of its encounters with virtual objects”.
In practical terms, it is absolutely amazing for reflections (eg neon lights in puddles, muzzle flash on windows, sunlight reflecting off a lake, etc), and when paired with the Deep Learning Super Sampling (DLSS) abilities of the cards, resulted in a massive leap in visual appearance (games like Metro Exodus and Control look incredible with RTX on), as well as increasing performance in general (higher frame rates, smoother images, etc)
The technology was introduced two years ago in the Turing architecture – based GeForce RTX 2000 series. Being first generation technology, it wasn’t cheap, and being able to afford a top-of-the-line RTX 2080Ti card (approximate cost: AUD$2000) was something of a flexing point for particularly involved members of the PC gaming scene.
The next generation, based on Ampere architecture, is offering an exponential performance leap over the 2000 series cards, and for considerably less money – the RTX 3070 will be available in October for about AUD$809, and have similar (or even in some cases better) performance for less than half the price of the top-end 2000 series card.
The major benefit Nvidia are spruiking is the usual high-resolution, high-framerate capabilities (the top-end RTX 3090 will be able to run games at 8K resolution)
There are some other benefits to the new cards, and they’re going to be of particular interesting to competitive gamers too – notably, in the form of reducing system latency.
In a broad sense, system latency is the delay – and we’re talking milliseconds – from you clicking a mouse button or pushing a key and seeing its effect on screen. For the majority of us, it’s unnoticeable and irrelevant anyway (we’re talking maybe 20-50ms, after all) but in professional e-sports it can be the difference between hitting a target and being taken out yourself.
One of the innovations in the Ampere series chipsets is Nvidia Reflex –which, rather than being a lonely child waiting by the park, or in charge of finding treasure in the dark – is a latency reducing technology that aims to eliminate the render queue aspect of system latency.
Nvidia’s testing shows noticeable improvements in system latency – especially when using the RTX 3090 card – which will be a big deal to top-tier e-sports competitors.
There’s also a totally non-gaming application to all this, with the RTX cards allowing for better rendering of 3D images and video, faster encoding, and use of AI in the process; Nvidia claim the new cards offer around twice (or even more) the performance of the RTX 2080Ti in that space.
Of course, all this is based on data Nvidia have provided and while I have absolutely no reason to doubt them, the real proof will be when review units start making their way out into the wild and we get to see some ‘real-world’ tests coming online.
I’ve spoken with both Nvidia and Asus about obtaining one of the cards for testing and review, and will keep you posted on the situation.
In the meantime – between the new consoles and the RTX 3000 series cards, it’s looking to be a very good time to be a gamer indeed, no matter what your platform is.