Every year, Nvidia hosts a three-day event about all things graphics processor-y, called the GPU Technology Conference or GTC, for short. Originally aimed at rendering and GPGPU, the presentations, talks, and demonstrations are now all firmly in the AI camp. However, for 2026, PC gamers might have something to look forward to, because Nvidia says that the ‘future of real-time rendering’ is going to feature in CEO Jensen Huang’s keynote speech.
That’s according to Nvidia’s GeForce account on X (including the UK-specific one), and given the nature of the channel, it means that whatever new stuff is going to be hyped up about ‘real-time rendering’ is certainly going to be about gaming. Unfortunately, that’s all the post says, other than simply retweeting the main Nvidia GTC reminder from three days ago.
Catch the future of real-time rendering in Jensen’s keynote tomorrow👀 https://t.co/KFv1JoTsDuMarch 15, 2026
So, let’s take stock of what Jensen is going to talk about, from what he’s most likely to say, all the way through to total pie-in-the-sky nonsense. My gut feeling is that the future of real-time rendering, from Nvidia’s perspective, is going to be all about the use of AI within the graphics pipeline, leveraging the use of the new DirectX Linear Algebra API and Compute Graph Compiler.
Together, these basically let developers run AI algorithms within the normal graphics pipeline, no different to how you’d code any other rendering process. Previously to all of this, you’d have to resort to using a proprietary API, unique to a specific GPU vendor, and figure out how best to shoehorn it all into one’s engine.
One use for this is neural texture compression, something that Nvidia has been working on for a while, but you don’t need a fancy API to do this, as Ubisoft has shown with Assassin’s Creed Mirage. But I strongly suspect that if Huang does focus on this, it will actually be all about making path tracing better/faster, as this is what Nvidia was promoting at the GDC event last week.
Had the RAMpocalypse not come to pass, we almost certainly would have been introduced to the Super refresh of the RTX 50-series cards, and while it’s still distantly possible that these cards do get announced, it now seems extremely unlikely given just how bad the DRAM situation is (neatly making me the world’s worst tech prophet in the process).
New RTX cards usually appear alongside some kind of new DLSS or RTX software feature, but we’ve already had that, in the form of RTX Mega Geometry foliage system and DLSS Dynamic Multi Frame Generation. Nvidia was probably going to keep these for the Super launch, but their appearance suggests that no new gaming GPUs will be coming from Team Green for a long time now.
All things considered, Huang is probably going to just reiterate what was said at the GDC, which is fine because anything that can be done to improve the performance of rendering, without the loss of visual fidelity, is certainly a good thing. But wait, I haven’t mentioned anything pie-in-the-sky!
Okay then, how about DLSS Ray Full Construction, an AI system that doesn’t just denoise a ray-traced scene but actually uses machine learning to generate thousands of additional ‘fake’ rays?
Ah no, I’ve got it! Huang will hop onto the stage with a new GeForce RTX graphics card. It’ll have 10,000 CUDA cores and 4 GB of VRAM, but it will be the first to use RTX Ultra Memory, an AI-powered system that neurally compresses everything automatically and does it so quickly and so well that it effectively quadruples the VRAM and its bandwidth on your graphics card.
Yours for just $1,299.
