Skip to content
cristiansoares edited this page Sep 3, 2026 · 1 revision

A tour of the app

Every feature, in the order you would meet them. Screenshots are from a real session on a 4K frame.

First run: point it at your DLSS files

Nothing NVIDIA ships is bundled. The fastest way to get the four files in place is Find my DLSS files…, which searches your Steam libraries, Downloads and Documents and copies a matched set in — including the newest add-on it can find.

Find my DLSS files searches your disk and copies a matched set

The workspace

Open an image (drag, paste, or browse) and you land here. The sidebar on the right maps straight onto the RenoDX add-on's own controls — neural strengths, style, HDR, depth, evaluation — and the action buttons are pinned below it. The row along the bottom switches what the canvas shows.

The blue line is the wipe divider: drag it across the frame to compare. Here it sits near the right, so almost all of what you see is the source.

The workspace, wipe divider near the right

Drag the divider the other way and you are looking at the neural pass — here at Intensity, Skin, Local Tone and Structure 2.00, Cinematic. The Before/After labels stay pinned to the frame corners as you zoom.

The neural pass at full strength

Depth mask

Depth is estimated the instant you open an image, so you can judge it — and tune its contrast live — before spending a DLSS pass. Near reads red, far blue.

The depth mask, near in red

Compare styles

The neural pass has two styles, Natural and Cinematic, and they differ by about as much as the whole effect — so leaving it on the default is a real choice made blind. Compare styles converts both at once and puts them side by side.

Two panes: Natural against Cinematic

Switch to three panes to put the source next to both — the way to answer "is this helping at all" rather than only "which style". Every pane shares one zoom and one pan, so they cannot drift apart.

Three panes: Original, Natural, Cinematic

Difference

"It looks the same" is the most common first impression, and usually wrong. The Difference view shows exactly what the neural pass changed, amplified to fill the range, with the numbers.

The difference view, amplified

Colour

Colour grades the finished image — exposure, contrast, saturation, vibrance — in linear light, live, because it runs after the neural pass rather than before it. Vibrance lifts skies and materials while leaving skin mostly alone.

Colour grading the result

Look closely

The result view is full resolution, so zooming in shows the real output — pore, weave and reflection detail — not a magnified preview. Scroll to zoom about the cursor, right-drag to pan.

Full-resolution zoom and pan

Video, with audio

The Video tab plays and scrubs a clip with sound, then converts it. Set an In/Out range on the timeline (scroll to zoom the timeline for a precise edit), pick an output, and export — H.264/MP4 hardware-encoded by default, the format every editor takes. The audio is carried across.

The video tab with an In/Out range on the timeline

Image sequences

The Image sequence tab converts a rendered sequence frame by frame, and can take your renderer's own depth pass — which is what keeps it temporally stable. Made for 3D artists rendering out of Blender, Maya and the like.

Image sequence mode for 3D artists

A whole folder

Apply to folder… runs every image in a folder at the settings you just tuned, showing each result as it finishes. For a pile of unrelated stills that all want the same treatment.

Apply to folder, converting a batch


Stuck on anything? The Troubleshooting page has every issue reported so far with its actual cause.

Clone this wiki locally