Blog
Dusk Developer and New Blood CEO Level Harsh Criticisms Against DLSS 5 “At This Rate Why Make Game Art At All?”
Oshry, who later expanded on the interview in a post on Bluesky, opened his comments by asking “First of all, who asked for this!?” and going on to say that DLSS 5 deviates from the original idea behind DLSS 5, which was fundamentally a technology used to increase frame rates using an upscaled image without substantial detail loss. He criticizes NVIDIA for calling a generative AI product DLSS, stating that “they’re hiding this Gen AI bullshit behind the DLSS moniker because they think we’re stupid,” encouraging gamers to push back against the tech and vote with their wallets, “just like we did with NFTs and crypto games.” Oshry mentions that one of the common defenses for DLSS 5 is that its optional, but goes on to question where tech like DLSS 5 leads while criticizing NVIDIA for recent hardware trends: “You know what’s not optional? Paying more for less—which is what we’ve been doing with NVIDIA for years now.” He continues along this line further down, comparing the tech to the other NVIDIA tech that either flopped or was left by the wayside, like 3Dvision and PhysX, and more modern tech, like RTX and Path tracing, all of which had their merits, questioning the merits of DLSS 5. “This is fundamentally changing the way video games look based on artificial intelligence that’s been trained on Instagram models and Epstein memes. Are you serious?” The game developer closes off by sarcastically asking “At this rate, why make game art at all? Why not just draw some shapes and colors and let AI generate what it thinks it should look like? After all, who cares, the only thing that matters is “realism” right?” Oshry’s full statement follows.
David Szymanski, for his part, echoes his colleague but adds criticism about DLSS 5’s aesthetic, criticisms of artistic intent aside, saying that “the lighting and contrast that it adds (or removes in some parts) makes scenes look less realistic and believable and more like one of those alpha chad Breaking Bad YouTube shorts with the contrast and sharpening cranked up.” He essentially questions the cost-benefit of DLSS 5, weighing the aesthetic against the industry-wide implications of increased AI adoption—i.e. more hardware scarcity and price increases. He also echoes what Oshry said about the trajectory of the tech and the possibility of DLSS 5 moving from an optional tech to something that is optional in name only, because game developers have leaned on the tech to develop the game: “You mean optional like upscaling? You mean optional like temporal AA. Optional like real-time GI? Optional like any number of optional features that anyone who has played a AAA game in the past half decade can tell you aren’t really optional, because games are now built to lean on those technologies.” He goes on to echo what many gamers online have said, stating that “what most gamers actually want are games that run at a consistent framerate and a sharp resolution, with consistent lighting and art design, on hardware that doesn’t require us to remortgage our house, using technology that doesn’t necessitate turning the world into a Mad Max wasteland.” Szymanski says that those are the games he wants and the games he will continue to make. Szymanski’s full statement follows.
Comments like those made by Oshry and Szymanski exemplify a rift forming in the gaming industry that sees pro-AI developers and gamers and anti-AI developers and gamers stand in opposition to each other. While it has been shown that generative AI is being used across the gaming industry, many of those video game workers who do use generative AI seem to think that it has a negative effect on the gaming industry. While studios like Capcom and Bethesda partnered with NVIDIA on the DLSS 5 demos, and Square Enix and EA have leaned into generative AI in game production, a number of game studios have rejected the tech entirely.
