The ray-tracing technology introduced with the RTX 20 Series was a game-changer, allowing for lifelike lighting, improved detail and a huge boost in performance. But, like most new tech, it’s not until the second or third iteration that you really see the potential. That’s the case with the GeForce RTX 30 Series, with the headline RTX 3080 offering 2x the performance compared to the RTX 2080, and Nvidia claims it’s even more powerful than the high-end RTX 2080Ti too.
But despite what Nvidia calls the greatest generational leap ever, the new range of GPUs don’t cost the world – unless you’re talking about the RTX 3090, packing a whopping 26GB of graphics memory of course. Here’s all you need to know about the Nvidia GeForce RTX Series including pricing and the headline features of the next-gen GPUs. For more, take a look at which Nvidia GPU should I buy?.
RTX 30 Series release date
The headline GeForce RTX 3080 was released 17 September. The expensive beast of a GPU that is the RTX 3090 followed on 24 September, and they were joined by the cheaper RTX 3070 on 29 October, following a two-week delay due to stock issues.
Fast forward to 2 December and Nvidia released the RTX 3060 Ti, but not the standard 3060. Nvidia decided to wait until CES 2021 to reveal the new entry-level GPU, and teased a release towards the end of February.
There's no solid date in place just yet, but we'll update this section when we find out more.
How much does the RTX 30 Series cost?
The big question is around the price of the RTX 30 Series range, and the good news is that despite the incredible leap in terms of power, the prices are tempting. Compared to the £1200/$1200 RTX 2080 Ti, which is outperformed by the entire 3000 Series range, the new cards offer great value for money.
- Nvidia GeForce RTX 3060 - £299/$329
- Nvidia GeForce RTX 3060 Ti - £369, $399
- Nvidia GeForce RTX 3070 - £469, $499, €499
- Nvidia GeForce RTX 3080 - £649, $699, €699
- Nvidia GeForce RTX 3090 - £1399, $1499, €1499
What’s new with the RTX 30 Series?
Nvidia is calling the RTX 3080 the biggest generational leap ever when it comes to its graphics card range, and it’s not hard to see why.
The Ampere architecture in the Nvidia GeForce RTX 3080 offers 30 Shader TFLOPs, 38 RT-TFLOPs and 238 Tensor TFLOPs, offering a 2.7x, 1.7x and whopping 2.7x improvement compared to the Turing architecture in the RTX 2080. That’s a lot of numbers, and it’s quite daunting if you don’t know the ins and outs of GPU architecture, but what it essentially means is that it’s powerful. Really powerful.
In fact, Nvidia claims that the RTX 3080 offers 2x the performance and energy efficiency compared to the RTX 2080, with the new card able to render [email protected] with raytracing enabled without the need for DLSS 2.0. It’s even more powerful than the ultra-high-end RTX 2080Ti despite costing a little more than half the price.
The company showed off just how powerful it is by once again showcasing the Marbles at Night video during the announcement; while it ran at [email protected] on the RTX 2080Ti when it was originally shown at GTC 2020, the RTX 3080 can output the fully pay-traced video at [email protected] It’s a stunning clip, and everything seen is fully dynamic – the lighting, physics and more.
That’s possible thanks to a combination of a second-generation AI core, RTX I/O to reduce the bottleneck of decompressing data and the introduction of the new GDDR6X graphics memory. The latter offers a 2x improvement compared to the standard GDDR6 graphics memory, with Nvidia claiming the tech will revolutionise load times and the complexity of next-gen games that take advantage of it. There’s 10GB of GDDR6X memory available as standard with the RTX 3080.
There’s also an innovative new cooling system that works with the fans of the PC case to help keep the graphics card cool. It’ll draw cool air in from the bottom of the case and expel hot air near the top, where most outlet fans are found. There are two sets of independently controlled fans, but despite the increased airflow, the RTX 3080 is 3x quieter than the RTX 2080 Ti while managing to keep the GPU 20 degrees cooler too.
If you don’t need that level of power, the RTX 3070 may be of more interest. It offers the same design as the 3080, and sports the same GDDR6X graphics memory too, although it’s capped at 8GB. That being said, the RTX 3070 still performs better than the RTX 2080 Ti, offering a significant upgrade for most gamers at a cheaper cost.
There's also the RTX 3060 Ti, announced later in December, offering improved performance compared to the last-gen RTX 2080 Super and all the other benefits of the 30 Series with a great £369/$400 price tag. It offers 8GB of VRAM, but unlike the rest of the collection, it's GDDR6 and not the upgraded GDDR6X memory. Nvidia says it's perfect for 1080p and 1440p gamers, giving access to RTX performance on a budget.
Then, at CES 2021, Nvidia revealed the entry-level RTX 3060. The £299/$329 graphics card actually offers more VRAM than the 3060 Ti, at 12GB compared to 8GB, but with fewer CUDA cores at 3584 compared to 4864, suggesting the Ti model will have better RTX performance. No word yet on the exact release date, but Nvidia suggests it'll be released at the end of February 2021.
We cover the latest Nvidia RTX 3060 release date news separately for those who want to stay in the loop.
At the other end of the spectrum, for those that need the most graphical power possible, there's the RTX 3090, replacing the limited-edition Titan in previous-generation graphics card lines. The £1,399/$1,499 BFGPU isn’t for everyone, but with an eye-watering 26GB of GDDR6X memory, it's more than capable of running [email protected] – if you’ve got a TV that supports [email protected] anyway – and powering high-end multi-screen flight simulators with ease.
For more, take a look at the best gaming PCs and our pick of the best gaming monitors to get the most out of the new GPU range. If you're new to the world of GPUs, our What is a GPU? explainer may help.