How Much Does it Cost?
The Nvidia RTX 5090 is launching this month, with the flagship RTX 5090 and the RTX 5080 available from January 30th, priced at $1,999 (£1,939) and $999 (£979) respectively. The RTX 5070 Ti and RTX 5070 will follow in February with prices set at $749 (£729) and $549 (£539). UK prices are estimated at this time, please wait for official figures.
Nvidia Founders Editions will be available directly from Nvidia and select retailers, while partner cards from manufacturers like ASUS and MSI will also hit the market, and we’ll have reviews of those for you as soon as possible. You can check for up to date stock and prices at Scan.co.uk here.
Overview
So let’s break this down a little because with a card like this, there are multiple angles to look at. The first is that this is the most powerful GPU in the world. There’s no doubt about that, but it comes with a caveat, though it’s a pretty obvious one. If you’re slapping down $2000 on a 5090, you’ve probably got a 4K high refresh-rate monitor and you also would expect the very best performance. Alternatively, and this is something we’ve not touched on, you’re a professional looking for Titan-like performance in GeForce clothing. Regardless, money is likely not an obstacle and instead, you want the best of the best. Using the 5090 at anything other than 4K, or perhaps 5K2K almost seems criminal. So who knows, maybe 8K gaming will finally take off at some point soon. Drilling down further, we can break it up into rasterisation, ray tracing and upscaling, and they all have their merits, but also their downsides.
Rasterisation
When it comes to rasterisation, we have strong performance, though as I said many times, the 1% lows do seem like they need some work, but through optimisations and driver updates, I feel that this can be fixed, and in quite a timely manner. It’s also not a huge leap as I’m sure a lot of consumers will expect for rasterisation, but I can also see why. From NVIDIA’s point of view, they are hedging their bets on upscaling through AI and leveraging that, which I’ll move on to shortly. There is still an uplift without upscaling, and maybe I’m wrong, but it feels like we’re almost at some kind of metaphorical limit for what we can expect from generation to generation. Perhaps that’s why other avenues are being explored to give better performance, better visuals and better latency, and all of those factors combined does equal a pretty damn impressive experience.
Ray Tracing
Ray tracing has also seen an uplift and is almost ushered into the same boat as rasterisation now. We’ve had raytracing for quite a few years now, and it’s made huge strides in terms of what’s possible, and how we get that next level of immersion. The 5090 can run games at respectable framerates without upscaling when raytracing is turned on, but game developers are going to want to keep pushing this to its limits by not just having certain elements of raytracing, but going all in with full ray tracing, including global illumination, reflections, shadows, and ambient occlusion—all calculated in real-time. This level of fidelity brings an unparalleled sense of realism to games, making environments feel more dynamic and lifelike and that is still considered next level.
The NVIDIA 5090, with its improved architecture and raw power, is finally starting to make this vision feasible without significant compromises. However, as developers push towards fully ray-traced worlds, the challenge will shift towards optimizing performance and ensuring these visual enhancements don’t come at the cost of gameplay fluidity or accessibility for players with mid-range hardware; something the 5090 doesn’t have to worry about.
For gamers, the combination of hardware like the 5090 and advances in software technologies like DLSS means raytracing is no longer just a luxury; it’s becoming the standard for next-gen gaming experiences, though I guess a big question comes down to how far developers can push raytracing before the hardware has to catch up once again.
Upscaling and Frame Gen
Which leads me onto upscaling, and more importantly, Multi-Frame Generation (MFG), a groundbreaking feature that we now have with the Blackwell architecture. MFG takes upscaling to an entirely new level by not only predicting intermediate frames for smoother motion but also leveraging AI to generate entire frames, reducing and enhancing the rendering workload on the GPU.
It works in conjunction with traditional upscaling methods like DLSS to deliver those stunning visuals while also maintaining high framerates, even in demanding scenarios. The big difference with MFG, means that the 5090 is capable of rendering at lower native resolutions while making the output look as if it were produced at 4K or even 8K, and this improves on what we’ve seen before by offering an incredible balance of performance and image quality.
This personally, is a game-changer for ray tracing-heavy games, as it allows developers to implement advanced features like full raytraced environments without worrying about crippling performance and based on seeing what it can do, the real marvel of MFG lies in its ability to handle unpredictable scenes with rapid changes. Unlike traditional upscaling, which sometimes struggles with motion artefacts, MFG adapts to each frame intelligently, therefore preserving detail and minimizing visual distortions. While it’s ideal for gaming, it also has a place in creative applications like 3D rendering and VR too.
Reflex
While MFG on its own should and does introduce drawbacks in terms of latency, Reflex comes to the rescue here and shows that working in tandem, we’re now able to reach higher levels of visual fidelity, with higher frame rates, and lower latency and this all gives us a glimpse into the future of real-time graphics.
Should Buy One?
Is the RTX 5090 for everyone? No, of course not. It’s positioned as a premium-tier product and is designed for enthusiasts and professionals who need the highest levels of performance and cutting-edge technology. However, with the release of the RTX 5080, 5070 Ti, and 5070 coming soon, NVIDIA is making sure that the Blackwell architecture is accessible across a range of price points and performance tiers.
Every product line benefits from a flagship model, and the RTX 5090 can be described as a showcase of NVIDIA’s technological capabilities. I’m looking forward to seeing what the rest of the stack has to offer, as well as seeing more and more games take advantage of what the 50 series as a whole has to offer.
The RTX 5090 is more than just a product; it’s a demonstration of what’s possible at the forefront of GPU technology. While it’s not practical for every user, there’s nothing that can compete with it, and with the improvements I spoke about, I think it’s only going to get better.
