NVIDIA DLSS, Ray Tracing Arrive in Deliver Us The Moon

Deliver Us The Moon, developed by KeokeN Interactive, received a PC update that adds ray tracing and NVIDIA DLSS, according to coverage by KitGuru and WccfTech. KitGuru and WccfTech detail three ray-traced effects introduced in the patch: opaque reflections, transparent reflections, and shadows (KitGuru; WccfTech). HardForum reproduces content from NVIDIA's GeForce announcement that reports DLSS delivers a 1.6x+ frame-rate improvement while providing image quality comparable to native resolution with Temporal Anti-Aliasing; the same reporting lists three DLSS settings: Quality, Balanced, and Performance (HardForum; NVIDIA GeForce post). DSOGaming's PC performance analysis reports that an RTX 2080 Ti with RT set to Epic and DLSS enabled produced an average of 60 FPS at 1080p, an average of 64 FPS and a minimum of 49 FPS at 2560x1440, and that the card could not reach a locked 30 FPS at 4K under those settings (DSOGaming).

Editorial analysis - technical context: DLSS here refers to NVIDIA's temporally-aware neural upscaling approach that reconstructs higher-resolution frames from lower-resolution render targets and aggregates information across frames. Public reporting on this update highlights that the implementation for Deliver Us The Moon includes a sharpening filter applied post-reconstruction, which reviewers credit for improving apparent sharpness relative to earlier DLSS versions (DSOGaming; WccfTech). Reported visual trade-offs include increased aliasing on fine edges when compared to TAA in some screenshots, even as frame rates improve (DSOGaming).

Industry context: This patch sits within the broader pattern of games adopting hybrid raster + ray-tracing rendering while using learned upscaling to recoup performance. Multiple outlets frame the Deliver Us The Moon update as another data point showing DLSS can enable ray-traced effects at playable frame rates on RTX-series hardware, particularly at 1080p and 1440p. For graphics engineers, the implementation notes (three RT effects, three DLSS presets, and a sharpening pass) illustrate typical levers developers use to balance fidelity and performance when integrating neural upscaling into Unreal Engine 4 titles (KitGuru; DSOGaming; WccfTech).

For practitioners: look for independent validation of image-quality claims from third-party benchmarks and frame-capture comparisons at native resolution, TAA, and each DLSS preset. Observers should also track whether the sharpening pass introduces temporal instability or aliasing artifacts in motion, and how the game's DX12 RT implementation interacts with common denoising and postprocessing pipelines (DSOGaming; WccfTech). Finally, broader attention points include how other UE4 titles replicate the same balance of RT effects and DLSS presets, and whether future DLSS updates continue to shift the fidelity-performance trade-off.