Five Minutes to Your First Inverted Roll

When launching NexFilm, the application displays the darkroom-inspired Library workspace by default. The first step in processing a roll is cataloging your digitized scans into the local engine environment. Click the Import Roll button in the top-right header to summon the import configuration modal. NexFilm provides two ingestion methods tailored to different scanning styles:

When choosing Import by Roll, click the browse button to select the target directory containing your digitized frames. Specify the film stock name (such as Kodak Portra 400) and the capture format (such as 135). Click Confirm Import, and NexFilm will catalog the files inside the local SQLite database (nexfilm_user.db). The import step generates only lightweight thumbnail previews and indexes physical metadata. The original image files remain strictly read-only on your disk and are never altered, moved, or duplicated.

In the Library thumbnail grid, double-click any representative frame, or select a frame and click the Develop tab in the centered view switcher. The application transitions smoothly into the GPU-accelerated grading workbench.

During your initial entry into Develop, a brief loading sequence appears on the central canvas. This indicates that NexFilm is initializing its on-demand 16-bit linear proxy generation pipeline. Traditional photo editing software attempts full debayering and decoding of every high-resolution RAW file across an entire roll upon import, consuming gigabytes of system memory and stalling disk I/O. NexFilm uses an asynchronous, demand-driven architecture:

1. Zero Full-Resolution Decoding on Ingestion: The Library reads embedded container previews to achieve immediate thumbnail rendering across dozens of rolls.
2. On-Demand 16-bit Floating-Point Proxy Processing: Only when an exposure is opened in Develop does the background Rust engine invoke LibRaw or the high-throughput TIFF parser, transforming Bayer sensor data into an unclipped 16-bit linear transmittance texture uploaded to WebGL 2.0 texture units.
3. Real-Time GPU Shader Pipeline: The main viewport relies on custom WebGL2 fragment shaders to calculate logarithmic optical density conversions, printer light color additions, and tone curves in parallel, updating the display at a sustained 60 frames per second during parameter interaction.

For viewport interaction, use the mouse wheel over the canvas to zoom smoothly in and out centered on the cursor position. Hold down the keyboard spacebar Space and drag the mouse to pan across the negative, examining emulsion silver grain structure and optical focus. Double-clicking any empty canvas area resets the view to fit the full frame.

Within the Develop workspace, the most influential step governing inversion quality is establishing the Film Area. In the right-hand inspector panel, expand the "Geometry and Area" panel and select the Set Film Area button. Four corner handles and bounding border lines will appear over the negative preview.

Click Auto Area in the toolbar. The edge-detection algorithm scans the frame boundaries and fits an initial rectangular crop against the dark rebate. Human visual confirmation is essential at this juncture.

Once the boundary rectangle is confirmed to hug the exposed image area, click Save Area in the inspector. NexFilm stores the crop coordinates in the database and hides the interactive handles.

With the film area confirmed, click the Auto Invert button in the inspector. NexFilm processes the frame in milliseconds:

1. The engine samples the unexposed orange base rebate located outside the crop line (or from pre-calibrated base markers) to calculate the negative base density vector $(D_{\min}^R, D_{\min}^G, D_{\min}^B)$.
2. It executes logarithmic transformation and matrix density subtraction across all pixels enclosed within the crop boundary, mapping the image into calibrated optical density space.
3. The bounded channel span response model calculates initial highlight-shadow contrast and color balance, presenting an authentic, natural positive image on the canvas.

If your roll was captured on a static copy stand with a fixed negative carrier, frame placement remains identical across exposures. In this case, there is no need to adjust the crop box individually for every exposure. On the calibrated first frame, click Batch Apply in the inspector panel. Check the option for Film Area & Geometry, and click Apply to Roll. NexFilm broadcasts the exact crop coordinates to every frame in the current roll.

Auto Invert provides a physically sound photographic baseline, but individual lighting situations and artistic intentions require fine-tuning. In classical darkroom printing, color technicians adjust exposure by inserting red, green, and blue optical filters into the enlarger head, known as Printer Lights.

NexFilm reproduces this darkroom methodology in the Printer Lights inspector panel. Operating in the logarithmic density domain, adjustments follow additive and subtractive color filtration rules:

1. Red Light: Moving the slider right increases red light, warming the positive and removing cyan coldness from shadow areas. Moving left decreases red, shifting the image toward Cyan.
2. Green Light: Moving the slider right warms the tone toward green. Moving left decreases green, eliminating yellow-green casts and steering the palette toward Magenta.
3. Blue Light: Moving right warms the palette toward yellow. Moving left cools the image toward Blue, ideal for neutralizing warm tungsten indoor lighting.

To prevent accidental parameter shifts, slider interaction includes intentional focus requirements: scrolling the mouse wheel over a slider produces no change until you click the slider once with the left mouse button to give it focus. Once focused, the mouse wheel steps values with precision. When evaluating sharpening, holding down the keyboard Alt key while adjusting Amount, Radius, or Detail temporarily renders a high-contrast grayscale edge mask on the canvas, reverting to full-color preview when released.

After refining your adjustments, standard darkroom procedure requires exporting a single high-resolution proof before committing to a full-roll batch export.

Click the Export button in the top-right header to display the single-frame export modal:

1. Export Format Selection: For online publishing and client previews, select high-quality JPEG (quality factor 95 to 100). For downstream grading in Adobe Photoshop, Lightroom, or DaVinci Resolve, select uncompressed 16-bit TIFF.
2. Color Space Assignment: For standard displays and web deployment, choose sRGB. For wide-gamut monitors such as the Apple Studio Display or hardware-calibrated printing monitors, select Display P3 or Adobe RGB.
3. Color Profile Embedding: Confirm that Embed ICC Color Profile is enabled to guarantee consistent color reproduction across operating systems and web browsers.

Click Confirm Export. The background Rust engine decodes the full-resolution RAW data and performs full-bit-depth rendering, writing the positive file to disk in seconds. Open the resulting TIFF or JPEG in your system image viewer, zoom to 100%, and verify shadow noise levels, skin tone gradation, and highlight texture.

When the proof matches your quality expectations, click Export Roll in the header. The roll export dialog allows configuring output filename tokens (such as {Roll}_{Seq}_{Film}) and setting worker thread concurrency. Click export to process the entire roll sequentially, delivering a complete set of color-consistent digital positives in minutes.