Full Nerd Hey, Nvidia: DLSS 5 can’t make great art with imperfect science Good technique inspires revulsion as a theme, rather than a judgement. Full Nerd , , PDT Nvidia Nvidia’s controversial DLSS 5 technology, which uses “real-time neural rendering” for photoreal game lighting but faces criticism from gamers who argue it compromises artistic integrity.The debate centers on whether DLSS 5’s technical improvements justify what critics call “yassifying” game art, with some viewing it as unwanted beauty filtering that devalues original artistic vision.Despite CEO Jensen Huang’s defense of the technology as “generative control,” the controversy highlights the ongoing tension between technical innovation and preserving authentic artistic expression in gaming. You ever have debates about art? I do. They’re messy, particularly the ones about video games. And now Nvidia’s thrown more chaos into the mix with DLSS 5. Announced at GTC 2026, this new dynamic lighting technology is described by Nvidia as “real-time neural rendering” that adds “photoreal lighting” to pixels—on a level allowing game developers to rival Hollywood in creating realistic graphics. The public disagrees, with my colleagues among the vocal opposition. Welcome to The Full Nerd newsletter—your weekly dose of hardware talk from the enthusiasts at . Missed the surprising topics on our YouTube show or latest news from across the web? You’re in the right place. Want this newsletter to come directly to your inbox? Sign up on our website! Gamers accuse DLSS 5 of “yassifying” game art, a.k.a. applying beauty filters liberally to characters and environments alike. (As @thedragonbrandy.bsky.social commented on Bluesky, “We went from raytracing to sloptracing.”) For his part, fellow Mark Hachman sees DLSS 5 as “the sprinkling of AI content on top of games, devaluing them in the process.” Nvidia CEO Jensen Huang fired back, saying the criticism was “completely wrong.” His rebuttal focused on the technical, emphasizing this new tech applies “generative control” to a game’s geometry, rather than serving as post-processing. Who’s right in this argument? I can’t make that call, though I have my sympathies. Both sides have valid points. As they do in virtually every fight about art. Here’s my theory for why: No matter the medium, art can be broken into two main pieces—”science” (that is, technical execution) and emotion. You can’t have only one or the other. You need science to provoke a reaction in your audience, to capture them with skills only a master artist can execute. The more varied and refined your technique, the more layered and powerful the response you’ll draw out. But you also need emotion to hook an audience, to reach them on a deeper human level. The stronger you can tap these emotional buttons, the more profound the connections you make. Science helps tell the story effectively. Emotion makes people invest in that story. But weaken one and the other goes down with it. That’s the heart of this clash, in my eyes. Grace’s face with DLSS 5 on has thrown a lot of people.Nvidia Nvidia portrays itself as aiding game development; DLSS 5 “doesn’t change the artistic control” of game studios, according to Huang. Yet that view ignores what gamers see: DLSS 5 isn’t working as intended. When science in art is done well, audiences don’t notice. Most enjoy the experience and leave the nitpicking of technical details to enthusiasts. Had Nvidia succeeded on this front, DLSS 5 would be viewed more positively, a beneficial perk. (Maybe even one justifying the cost of an expensive graphics card to support the feature.) Instead, gamers are dissecting DLSS 5’s execution to the point of questioning its necessity. Some of this may be due to how little is known about DLSS 5’s implementation, and how much true control game artists have over its effects. (For example, did someone at Capcom punch up Grace’s looks with more makeup and hair lowlights, despite the lack in the original Resident Evil: Requiem scene? Or is that truly DLSS 5’s influence?) Still, even with that lack of info, proponents of DLSS 5 would be wise to listen, rather than call critics ignorant idiots. Non-experts in a field don’t have the knowledge or the language to concretely explain what they notice. Their descriptions usually come out as feelings or get attributed to incorrect root causes. Despite that, their feedback is still valuable. A distracted audience is not an invested one—an observation I’ve made many times when evaluating art myself. And here, the distraction is the very thing DLSS 5 claims to solve. Gamers don’t believe what they’re seeing is accurate. I personally believe in DLSS 5’s goal to elevate the science of making games. But I also agree with fans: That won’t happen if the tool isn’t better calibrated—or perhaps better implemented—by the time it launches. In this episode of The Full Nerd In this episode of The Full Nerd, Adam Patrick Murray, Brad Chacos, Alaina Yee, and Will Smith spar over the announcement of DLSS 5 (and its reception), Microsoft’s Project Helix, and Intel Arrow Lake Refresh details. We go a full 90+ minutes on Nvidia’s new lighting tech alone—and we weren’t the only chatty ones during the stream. I pulled out some of my favorite : @brandontrost8888dlss5 looks photorealistic. it’s definitely not AI slop. Truly game changing @DerxWiedergaengerpeople are angry because it looks like garbage. It entirely removes shadows, overwrites lighting entirely. Ultimately, it just looks like someone turned on vivid mode on their TV. @puretrack06Thomas the tank engine mod for all A lot of debate happened with little consensus, but we managed to agree Nvidia should have called this anything but DLSS 5. (Where’s the super sampling?) Also, we agree our viewers are pretty great, because they made this image below while we were talking. BMG / Discord Missed our live show? now to The Full Nerd Network YouTube channel, and activate notifications. We also answer viewer questions in real time! Don’t miss out on our other shows too—you can catch episodes of Dual Boot Diaries, The Full Nerd: Extra Edition, and Expedition: Handheld through our channel! And if you need more hardware talk during the rest of the week, come join our Discord community—it’s full of cool, laid-back nerds. This week’s broad nerd news DLSS 5 may have enthusiasts in a tizzy, but plenty hit the news stream to catch attention—or inspire shock. For me, I’m surprised about the Apple MacBook Neo’s reparability. Far less so when it comes to AI creating big headaches…or software getting exploited. Foundry Up is down, down is up: Apple got high marks on YouTube for the MacBook Neo’s reparability. Truly, we’re living in an alternative universe right now. A surprise to no one: OpenAI is allegedly pushing forward with its adult mode for ChatGPT, against the opinion of its advisors. I’ll let Ars Technica commenter s73v3r speak for me: “Why are we allowing a product that is so dangerous on the market? We outlawed Lawn Darts for less.” Browser extension danger: My colleague Michael Crider discovered a Chrome extension had been stealing his data for years. Better living through science: Researchers believe they may be able to pinpoint six different types of depression through functional MRI (fMRI) imaging. Practical application of the knowledge may be slower to follow, but perhaps eventually people will have to suffer less to find the right treatment. Brown town: Noctua fans, your time has come. No need to be subtle with your love for chocolate brown—you can now stuff your whole PC into a case with that aesthetic. Is Minority Report our future? Jailed for almost six months due to faulty AI facial recognition today. Perhaps jailed for thinking of jaywalking tomorrow. Jared Newman / Foundry Vintage vibes, mode...
