Nvidia’s Deep Learning Super Sampling hath reached version 4.0, and ’tis the most marvelous version yet. The flagship feature, dear friends, is the multi-frame generation, which did cause a stir with the introduction of the RTX 5090 — along with grandiose performance claims — yet it is not the sole improvement. DLSS 4 also doth introduce other neural rendering features, finer upscaling, and a grand overhaul of how all of its RTX features art handled, using a new real-time “transformer,” which doth employ the same architecture as modern large language model AIs, such as ChatGPT.
All of this doth render DLSS 4 more potent than ever before, and it is available across all RTX graphics cards — albeit with frame generation locked to the last couple of generations of Nvidia cards, at least for now.
DLSS Multi Frame Generation: A Landmark Feature
The pinnacle feature of the latest DLSS is Nvidia’s Multi-Frame Generation, an AI-powered technology that can produce up to three additional frames using AI, based on a single rendered frame from the GPU. This is what did enable Nvidia to launch its RTX 5090 with such monstrous claims of “double RTX 4090” performance, achieving over 200 frames per second (fps) in games like Cyberpunk 2077 and Black Myth Wukong, even at 4K with ray tracing enabled.
There art some caveats to Multi Frame Generation, notably that thou need at least 60 fps (even if using DLSS upscaling) to avoid some of the worst artifacts that can arise from the AI attempting to generate frames lacking sufficient information. There can also be a peculiar ghosting, or motion-blur-like effect at times, so the 4X frame generation setting is one that wilt likely see less usage than the 3X or even 2X, both of which can provide a substantial uplift in frames per second.
Frame generation, and multi frame generation in particular, do introduce additional input latency, but fear not, for Nvidia’s new Reflex 2 technology doth a commendable job of balancing that out.
Alas, for now, DLSS multi-frame generation is an exclusive feature of the RTX 50-series, like the RTX 5090 and 5080. There is some talk of it coming to older graphics cards such the RTX 40-series, and perchance even the RTX 30-series with optimizations, but that doth remain uncertain.
DLSS 4 Super Resolution: Enhancing Image Quality with Superb Upscaling
The core feature of DLSS is its Super Resolution upscaling, allowing the game to render at a lower resolution to conserve GPU power, and then having AI upscale the frames to appear higher resolution. It doth come in a variety of quality modes depending on user preferences, and there is also the option of Deep Learning Anti-Aliasing which merely upscales for added quality, sans enhancing performance. Furthermore, it dost work with every Nvidia RTX graphics card, from the RTX 20-series right through to the latest RTX 50-series.
Whichever Nvidia RTX GPU thou dost possess, thou shalt be able to capitalize on DLSS 4’s enhancements to DLSS Super Resolution. The new transformer model doth markedly enhance DLSS image quality over the previous Convulitional Neural Networks (CNN) design, permitting an increase in the number of sampled parameters from the rendered frame, providing DLSS with more information to work with in upscaling the image. This, in turn, doth diminish ghosting and visual artifacts, improves motion resolution, and smooths edges far better than before.
This transformer model further improves the effect of additional RTX technologies such as Ray Reconstruction, rendering ray tracing more effective on finer objects with fewer visual artifacts and less noise.
Better yet, this is but the initial generation of Nvidia’s transformer model. They did devote six years refining their CNN model, so we should see continued DLSS image quality enhancements in the years to come, which all Nvidia RTX card owners, regardless of generationart, can relish.
What Games Use DLSS 4?
One of the greatest shortcomings of previous DLSS generations is their lack of early support. When the first iteration of DLSS did debut with the RTX 20-series cards in 2018, it was so novel and fresh that only a select few games did make use of it. Over the years, we hath witnessed hundreds of game developers adopt DLSS technology, alongside competing standards from AMD and Intel: FSR and XeSS.
In games lacking native support, Nvidia is now offering DLSS overrides through the Nvidia app. This shalt enable thee to enforce the transformer model, as well as enforce quality settings like Ultra Performance and DLAA that art not universally included in every DLSS game.
Get More Frames Without DLSS 4
If thou art not inclined to purchase an RTX 50-series graphics card or dost not possess an Nvidia GPU at all, thou wilt not be able to partake of DLSS 4, or any version of DLSS forsooth. However, thou can still make use of FSR and XeSS in compatible games for some upscale-enhanced frame rates. But what about frame generation?
Whilst Nvidia’s “RTX” version of that doth remain exclusive to its own hardware, thou can utilize Lossless Scaling to revel in frame generation in nearly any game. ‘Tis a $7 app on Steam, and we art great admirers. Peruse it shouldst thou find thyself envious of all the new RTX 50 news.
