Scanning is half the job. Color correction, AI enhancement, stabilization, dust and scratch repair, and grain handling are the other half — and at FPL they’re bundled into your tier, never charged per-foot. This page explains what each capability does, what it looks like, and how to pick the tier that fits your film.
Most film labs charge per-foot for the post-processing work that actually makes your film look good. Scene-by-scene color is an upcharge. AI enhancement is an upcharge. Stabilization is an upcharge. The 'scan rate' you see on the home page turns into something else by the time the quote arrives.
A 1,000 ft Super 8 order at the “$0.36/ft base rate” can become $0.85+/ft once the work that makes it look good gets added back in. The base rate isn’t what you actually pay.
Pick a tier, see the estimate on the calculator, get exactly what you saw. No hourly add-ons, no per-frame restoration line items, no ProRes surcharge.
Color correction is the single capability that most affects how your film actually looks. The same reel can come back as a flat, washed-out video or as a vibrant, balanced image — depending entirely on what happens in DaVinci Resolve after the scan.
Film captures color through three dye layers (cyan, magenta, yellow). Over decades, those layers fade at different rates — usually cyan first, leaving a magenta/orange cast on old film. Even fresh film captured today has color characteristics specific to the stock, the camera, the lighting, and the lab that processed it.
Color correction is the work of putting that captured image back into balance, scene by scene, until it represents either (a) what you remember it looking like, or (b) what was actually on the film when it was new. The first is a viewer’s correction. The second is a preservation correction. Both require a colorist sitting in front of a calibrated monitor making decisions, frame by frame or scene by scene.
Modern scanners can do an automatic white-balance and exposure pass. That gets you 60% of the way there for cleanly-stored film from a single shoot. But most home movie collections have multiple shoots, different lighting, different film stocks, different ages, and different storage conditions on the same reel. An auto-correct pass treats all of that the same — and ends up over-correcting bright outdoor scenes while under-correcting dim indoor ones.
Scene-by-scene work means the colorist watches the footage, identifies each shooting location or time-of-day shift, and grades each section independently. It’s the difference between “the whole reel got the same treatment” and “every scene got the treatment it needed.”
Reference-grade color — what we do at the Archival tier — adds custom LUTs (look-up tables) calibrated to specific film stocks, flicker correction for unstable exposures, and full fade restoration (including the difficult work of recovering color from severely faded Kodachrome or Ektachrome).
Topaz Video AI is the same enhancement software used by film studios for restoration prep, broadcast networks for archival upscaling, and serious post houses for noise cleanup. We bundle it into Preservation and Archival because the alternative — selling it as an add-on — punishes customers for wanting a watchable result.
Topaz Video AI runs your scanned footage through neural network models trained on millions of frames of clean source material. It can: upscale SD or 2K footage to higher resolution by inferring detail, reduce noise without smearing the image, sharpen soft footage, and stabilize in ways traditional algorithms can’t.
It is not a magic restore button. It can’t recover detail that was never captured. It can’t undo severe damage. What it does, when used carefully, is take a scan that already looks good and make it look noticeably better — particularly for older footage where the source material is soft, noisy, or low-contrast.
Topaz has dozens of models, each optimized for different source material — film versus tape versus interlaced video, animated versus live-action, high-noise versus low-noise. Picking the wrong model produces watercolor-style smearing, plastic-looking faces, or artifacts that didn’t exist in the source.
At Preservation tier, an operator runs the appropriate model on your footage and spot-checks the result. At Archival tier, the operator reviews each scene’s result and selects different models for different sections of the same reel if needed. Either way, the AI does not run unattended — and if a scene looks worse with AI than without, it gets skipped.
Most labs in our tier band don't offer AI enhancement at all. Most labs that do offer it sell it as a per-foot upcharge. We bundle it because customers shouldn't have to pay a separate fee to get the most-watchable version of their own film.
Old film is shaky. Hand-held Super 8, sprocket-worn 16mm, and projector-transferred footage all have movement that wasn't in the original scene. Stabilization fixes that — but it's a continuum, not a binary feature.
Three different things cause shake in scanned film. Camera shake is the original handheld movement during shooting — usually you want to keep some of it because it’s part of the source. Sprocket weave is side-to-side wobble caused by worn or stretched sprocket holes — this is purely a transport artifact and should be removed. Frame jitter is up-and-down movement caused by inconsistent registration during scanning — also a transport artifact, also should be removed.
The challenge is removing the unwanted movement without flattening out the genuine handheld feel that distinguishes home-shot footage from a static surveillance clip.
Basic stabilization (Access tier) uses scanner-native frame alignment to lock down sprocket weave. That removes the worst of the wobble at no additional processing time. It doesn’t address rolling shutter effects, doesn’t compensate for warped frames, and doesn’t track motion across cuts.
Advanced stabilization (Preservation tier and up) adds tracked motion compensation — the software identifies stable points in the frame and uses them to anchor the image while keeping legitimate camera motion intact. The result is footage that holds together visually without looking digitally locked-down.
Some footage is best left mostly un-stabilized. A spinning ride at a 1972 amusement park, a kid running with a Super 8 camera, a panning shot through a wedding hall — the movement is the point. Operators flag these and stabilize lightly so the footage stays alive.
All film has dust. Most older film has scratches. Some film has more serious damage — emulsion lifts, water marks, splice tears. The capability ladder runs from 'automated cleanup that handles 90% of dust automatically' to 'an operator manually restoring problem frames one at a time.'
Modern scanners have multi-pass dust detection that identifies particles by their consistency across frames — a single bright pixel that appears for one frame and then disappears is almost certainly dust, not part of the image. Automated dust removal at the scan level catches the vast majority of small dust particles without operator time.
What automated cleanup can’t reliably handle: scratches that persist across many frames (the algorithm can mistake them for film content), emulsion damage that affects parts of the image content, large debris that obscures actual subject matter, and water marks or chemical damage that have permanently altered the film stock.
Manual frame-by-frame restoration is operator labor — someone sitting in front of a monitor, looking at problem frames, and using restoration tools (paint, clone, content-aware fill) to repair specific damage on specific frames. It’s the same work that goes into film restoration for theatrical re-release, just applied to your home movies or institutional archive.
It’s the right call when: a specific reel has irreplaceable footage with significant damage, a scene that’s central to the narrative has a scratch running through it, an institutional or grant-funded project requires reference-quality output, or the customer has explicitly requested it as part of an Archival-tier project.
Manual restoration is bundled into Archival tier as "manual frame-by-frame restoration on problem sections." It\u2019s not unlimited — if a reel needs hundreds of hours of restoration work, that becomes a custom-quoted project rather than a tier feature.
Film grain is the texture of analog film — the random pattern of silver halide crystals that captured the light. It's a defining visual characteristic of film and one of the things that makes film look like film. Many digitization services aggressively denoise grain because it's 'noise that bothers casual viewers.' We don't.
Digital noise is random, signal-degrading interference. It changes shot to shot, gets worse in low light, and serves no aesthetic purpose. Film grain is the actual physical structure of the captured image — silver halide crystals (or, in color film, dye clouds) that became visible when the film was developed. Grain is consistent within a film stock and intentional from a cinematography standpoint.
Removing grain through aggressive denoising produces what the restoration community calls “wax-figure footage” — smoothed-over images where skin looks plastic, fine details disappear, and the source material is clearly identifiable as “old film badly digitized” rather than as film at all.
We preserve grain at all tiers as a default. Topaz AI’s noise reduction (when applied at Preservation and Archival) is calibrated to reduce digital noise from the scanning process, not to eliminate grain. The result: a clean digital image that still has the texture of the original film stock visible.
For projects that explicitly require grain reduction (broadcast delivery requirements, severely grainy stock that’s distracting, customer preference), we can apply targeted grain reduction as part of an Archival project. It’s an explicit choice, not a default.
If you compare a Preservation-tier scan to a competitor's bundle-shop output of the same footage, you'll often see this difference immediately: ours looks like film, theirs looks like aggressively-processed video. Grain handling is a quiet quality signal that compounds over the course of a watching session.
A consolidated view of what's bundled in Access, Preservation, and Archival across all five capabilities. Use this as the reference when picking a tier — or when comparing a Preservation quote from FPL to a similar quote from another lab that charges per-foot for the work that's bundled here.
| Capability | Access | Preservation ★ | Archival |
|---|---|---|---|
| Color correctionDaVinci Resolve pipeline | Auto WB & exposureBasic global pass | Scene-by-scene | Reference-grade · custom LUTs · fade restoration |
| Topaz AI enhancementUpscaling, noise reduction, sharpening | — | Operator model selection · spot-checked | Per-scene model review |
| StabilizationAnti-shake, weave reduction | BasicSprocket weave + frame jitter | Tracked motion compensation | Advanced + per-scene review |
| Dust & scratchParticle removal, defect repair | Scanner pass~70% small dust | Auto dust + scratch (~95%) | + Manual frame-by-frame on problem sections |
| Grain handlingFilm texture vs digital noise | Grain preserved | Grain preserved · digital noise reduced | Per-scene noise decisions |
| Output formatsMaster files delivered | H.264 MP4Viewing quality | ProRes 422 HQ + H.264 | ProRes 4444 + DPX + H.264 |
| ResolutionFrame-by-frame scan | 2K~2048 × 1556 | 2K · same scan, more processing | 4K (or higher for 35mm) |
| FADGI documentationFederal preservation standard | — | FADGI 3-star spec met | FADGI 4-star · documentation bundled |
For the full tier rate card and per-format pricing, see the pricing page. To see what an order at each tier would cost for your specific film, use the project calculator.
Each capability above gets a dedicated blog post with more technical depth, sample images, and answers to specific questions. The most-asked topics are linked here; the full set lives in the blog index.
The full explanation of scene-by-scene color work, with sample images and references for what's possible at each tier.
ReferenceRestoration · ReferenceThe honest version: what AI and manual restoration can recover, and what's permanently lost regardless of how much you pay.
ReferenceGrain · ComparisonWhy we preserve grain, what aggressively-denoised footage looks like, and when targeted grain reduction is the right call.
ReferenceResolution · ReferenceHow resolution interacts with film stock, when 4K meaningfully adds detail, and when 2K genuinely captures everything the film holds.
ReferenceOutputs · ReferenceThe full plain-English reference for every format we deliver — what each is for, file sizes, and how to pick the right one for your workflow.
ReferenceScanning · FoundationalThe capture-side counterpart to this page — what happens at the scanner, before any post-processing begins.
ReferenceThe questions that come up most often about color, AI enhancement, stabilization, and what's bundled into each tier.
The next step is either a quick estimate or a question. The wizard gives you a complete estimate for your specific collection. The contact form gets a human response within one business day.