Kingdom Come Dev Criticizes DLSS 5 Slop Filter
▼ Summary
– Daniel Vávra, a game director, defends Nvidia’s DLSS 5 AI technology, suggesting it could eventually replace expensive techniques like raytracing.
– The article criticizes the technology’s current state, noting its purported benefits are speculative and framed by a future-oriented “I can imagine” promise.
– Vávra dismisses criticism, comparing early AI imperfections to a “little uncanny beginning” and asserting critics will not stop its adoption.
– The author contrasts DLSS 5 with previous, functional AI upscaling that improved game performance without altering artistic intent.
– Vávra is grouped with other developers publicly supporting generative AI, despite broader industry condemnation and skepticism about its achievable returns.
While much of the gaming industry has voiced strong criticism of Nvidia’s DLSS 5 technology, one prominent developer is offering a contrarian perspective. Daniel Vávra, co-founder of Warhorse Studios, has publicly defended the controversial AI-driven upscaling tool, framing its current flaws as mere growing pains. He suggests the AI-slopificator technology, as critics have dubbed it, holds future promise for significantly reducing development costs, potentially even replacing expensive techniques like raytracing.
Vávra’s optimism centers on a vision of what the technology could become. “I can imagine in the future devs will be able to train this tech for particular art style or specific people faces and it might replace expensive raytracing etc.,” the Kingdom Come: Deliverance 2 director stated. This forward-looking stance echoes a common refrain in generative AI advocacy, where promised benefits remain perpetually on the horizon. His comments arrive despite Nvidia CEO Jensen Huang’s own recent assertions that DLSS 5 already preserves artistic intent, claims later muddied by the company’s own mixed messaging.
Addressing the widespread distaste for the technology’s current output, which many compare to an aggressive TikTok filter, Vávra dismisses these issues as temporary. He describes this phase as just “a little uncanny beginning,” analogous to the awkward early stages of other AI tools like ChatGPT. In his view, this is a necessary step before the technology achieves its incredible potential, a transition he believes is inevitable. “No way haters will stop this,” Vávra declared on social media, arguing the tool is “way more than a soap opera effect,” a reference to the unnatural motion smoothing feature found on many televisions.
This defense highlights a peculiar dynamic within the industry. DLSS in its earlier forms was widely praised as a practical AI application, boosting performance and visual fidelity without compromising artistic vision. The backlash against DLSS 5 stems from its shift toward fundamentally redrawing game frames with generative AI, a move many see as degrading visual quality rather than enhancing it. Vávra’s position places him alongside other figures like former Blizzard executive Mike Ybarra, who recently criticized a studio for apologizing over its use of subpar AI art.
Collectively, these voices are forming a bloc of developers publicly supporting a multi-billion dollar gamble on generative AI, a technology whose practical returns for game development remain hotly debated. Their stance often involves championing the long-term potential of tools that currently produce widely criticized results, setting the stage for a significant reckoning if the anticipated technological leap fails to materialize and investor patience wears thin.
(Source: Kotaku)
