AvyScan Lab is optimised for digitised analog films: 8 mm, Super 8, 9.5 mm, 16 mm and 35 mm. It accepts video files (.avi, .mp4, .mov, .mkv, .wmv, .webm, .m4v, .flv, .ts, .mts, .m2ts, .mxf, .mpeg, .mpg, .ogv, .3gp) and image sequences (.tif, .tiff, .png, .jpg, .bmp).
No. All filters are exposed through graphical controls with presets. AvyScan Lab generates the AviSynth+ script for you. Advanced users can create their own custom filters via the built-in editor.
The free version is usable with no time limit: the essential filters (grain, colours, levels, sharpening, crop…), presets, the 6 languages and every export codec are available, and encoding runs at full quality. It is limited to one clip at a time and 3 shots per clip, and custom filters are read-only. The Expert version unlocks image stabilization (perforation, edge, image), splice detection/removal (Splice Cleanup), dust and spot removal (Remove Dust), GamMac colour-cast correction, projector capture, films with soundtrack and optical sound with sound correction, unlimited shots and title cards, frame-rate interpolation, the rounded-corner mask, desqueezing of anamorphic films, projects, loading several clips in one session, batch processing and creating/importing your own AviSynth filters.
No. The free version encodes at full quality, with no watermark and no time limit. Its limits are functional: one clip at a time, 3 shots per clip, custom filters in read-only mode, and the Expert features (stabilization, splices, Remove Dust, GamMac, projector capture, soundtrack and optical sound, title cards, projects, batch processing…) are not included. It is the real application, not a demo.
For now, only Windows 10/11 64-bit is supported, since AviSynth+ and most community plugins are Windows binaries. A cross-platform version is not ruled out in the longer term.
Windows 10 or 11, 64-bit. A recent multi-core processor and 16 GB of RAM give the best results — motion-compensated filters are computationally demanding. No dedicated graphics card is required: processing runs on the CPU. An SSD noticeably speeds up work on image sequences. Display: 1920 × 1080 minimum at 100% scaling — on a smaller screen (1366 × 768, for example), the interface does not fit.
For distribution: x264 (H.264) and x265 (HEVC). For archiving: FFV1 (lossless), UT Video, and TIFF/PNG sequences. For post-production: ProRes. For individual images: TIFF, PNG or JPG. You can pick CRF or bitrate mode, chroma subsampling (4:2:0, 4:2:2, 4:4:4) and the output format — from 480p to 4K DCI, as an exact canvas (centred black bars) or with the width fitted to the crop.
The software imposes no resolution limit: your scans are processed at their original resolution, whether HD, 2K, 4K or beyond. On export, the “Original” mode keeps the native resolution; standard formats — from 480p to 4K DCI, as an exact canvas or with the width fitted to the crop — let you downscale if needed (the software never upscales the image). The only practical constraint: the higher the resolution, the more RAM and processing time motion-compensated filters require — to keep the preview smooth, lower the preview scale; it does not affect export quality.
Not directly. First convert your RAW sequences to TIFF/PNG or to an intermediate video file (ProRes, FFV1…) with your usual RAW development tool (DaVinci Resolve, ffmpeg…), then import the result into AvyScan Lab.
No, there is no size limit — and nothing is being downloaded: your files are read locally. The first time a large video file is opened, AvyScan Lab builds an index to allow precise frame-by-frame navigation, which can take several minutes. The banner is purely informational: loading continues in the background and it disappears as soon as the file opens. The index is then reused, so subsequent openings are fast. Tips: keep the file on a fast local drive (not a network share) and use the latest version of the software, which shows the indexing progress. If the message persists, send us a report from the software's Contact window.
No, load your file as it is. The preview is deliberately silent and exports are video-only by default, but the audio never gets in the way of restoration. To keep the sound track in the final file, enable the “Sound Track” filter (Expert version): the original audio is re-attached at encoding and automatically re-synchronised with the frames removed by Splice Cleanup, with built-in gain adjustment and audio preview.
For the start and end of the film, simply set the “start” and “end” fields of the frame range. For a section in the middle, cut the film into shots: place the playhead on the first frame to remove, click the scissors “Cut into shots at the playhead”, do the same on the first frame after the section, then right-click the thumbnail of the shot to remove › “Exclude from compilation”. Nothing is deleted: “Include in compilation” brings it back at any time. The free version allows 3 shots per clip; the Expert version has no limit.
Right-click the scissors cell of that splice in the shot strip: “Include in the edit” gives these frames back to the film. “Include in the edit, keeping the cut” restores them while keeping the boundary between the two shots. To remove no frames at all, the Splice Cleanup panel offers “Keep all splices”: the film stays cut at the splices, but nothing is removed on export.
Save your settings as a preset, then in the preset manager choose “Apply to this clip” (or double-click the preset) or “Apply to all clips” (several clips per session: Expert version; clips cut into shots receive the preset on each of their shots). In a film cut into shots, a filter's “Apply to all shots” button copies its values to every shot. Presets deliberately carry neither the stabilization area nor the crop, which are specific to each reel: transfer them with “Copy zone” / “Paste zone” in Stabilization and “Save coordinates” / “Paste coordinates” in Crop.
In the Crop filter, choose the aspect ratio you want — 4:3 for 8 mm, Super 8, 9.5 mm and silent 35 mm, 1.37:1, 1.66:1, etc. — then position the frame on the image: the ratio is kept throughout the adjustment. On export, a format with displayed dimensions (for example 1920 × 1080) gives exactly that size: the image is scaled down without stretching and centred between black bars. The “fitted width” formats keep the height and follow the width of the crop, with no black bars.
Enter the film's real frame rate in the source “fps” field (for example 18 for a 25 fps file of a reel shot at 18 fps), then press Enter. All frames are kept, with no interpolation: only the playback speed changes, the duration becomes longer and the sound track follows. Enter 0 to return to the file's frame rate. To then deliver a 25 or 30 fps file with smooth motion, use the Interpolation filter (Expert version).
These are interlacing artefacts, typical of video captures. Add the Deinterlace filter (available in the free version): it automatically goes first in the chain and rebuilds the missing lines taking motion into account; frame rate and duration stay unchanged. If motion judders, try the other field order (TFF or BFF; DV is BFF). Do not combine it with Projector Capture, which already handles interlacing. A frame-by-frame film scan does not need it.
Yes: to cancel the shake, each frame is shifted slightly, which uncovers its edges. Remove them with the Crop filter, keeping as much of the image as possible. With the Image method, a fast camera move (quick pan, jolt) can produce large black borders, because the software tries to compensate for it. To limit this, reduce the maximum corrected amplitude, “maxstabH” (horizontal) and “maxstabV” (vertical), from 14% by default to 10% for example: a large movement is then only partly compensated. Raising “cutoff” also helps: only fast shaking is corrected, and slower camera moves are followed instead of being compensated. Raising “damping” (1.0 by default, for example to 3.5) smooths the correction and makes it gentler during these movements. Finally, the “zoom” setting hides the remaining borders with a slight enlargement (1.03 is usually enough), and the “mirror” option fills them with a mirror image.
Three settings do nearly everything. The area: it must cover the whole sound track, without the perforation or the bright edge of the film. The track type: compare the look of your track with the diagrams — a track made of lighter and darker grey bands is variable-density and is read with “Grey bands”, whereas a wavy ribbon is variable-area. Finally, the gamma is adjusted to suit the scan. Listen to a short test excerpt of a few seconds first, before processing the whole reel.
AvyScan Lab restores the image and the film's original sound (magnetic stripe or optical track), but it does not do sound editing. To add music to a silent film, restore it, export it in a post-production format such as ProRes, and add your music in your usual editing software (DaVinci Resolve, Premiere, Final Cut…): the image quality is fully preserved.
In the preset manager, select one or more presets and click “Export”: you get a self-contained .avypreset.json file holding the presets and the full definition of the filters they use. On the other computer, “Import” recreates any missing filters and then adds the presets; a preset with the same name is renamed, never overwritten. Expert version feature. External DLLs or .avsi scripts that a custom filter depends on are not included: the recipient must have them too.
The Expert licence is perpetual: one payment, no subscription, no expiry. It can be used on 2 computers at the same time and includes one year of updates. Once the key is entered, every feature is unlocked immediately. At each start, if the computer is online, the software automatically confirms the licence with the server (only the licence key and the computer name are sent, never your films or your settings). A computer can stay offline for up to 21 days; beyond that, the software switches back to the free version until the next connection, where everything is restored automatically, with no new activation and no loss of your projects.
The Expert license can be active on 2 computers at the same time (for example your workstation and a laptop): simply enter the same key in the License window on each machine. To switch to a new computer, click "Deactivate" in the License window of the old machine to free up the slot. If the old machine is no longer accessible, write to us at contact@avyscan.com.
At startup, the application checks whether a new version is available and offers to install it. The Expert license includes one year of updates; after that, your version keeps working fully, with no time limit. The details of every release can be found in the software's Infos > What's new menu.
Yes. All processing (analysis, preview, encoding) happens locally on your machine — your films are never sent over the internet. A connection is only needed to activate an Expert license, check for updates and send a bug report. For processing machines deliberately kept offline, the bug report can be exported to a file and emailed from another computer.
No. Your films never leave your computer: restoration, preview and encoding are entirely local. The application contains no analytics or tracking tools, and nothing is transmitted without an explicit action on your part (such as sending a bug report).
Yes, only in the Expert version. AvyScan Lab includes a custom filter editor: write your AviSynth code, declare your parameters between curly braces {name}, and define the control type (slider, list, checkbox, text). You can import your own .dll plugins and .avsi scripts, and export your filters as JSON to share them.
The player reads every frame of the scan in full: its size (width × height × channels × bit depth) sets the throughput required from the drive. A 4069 × 2736 frame weighs about 33 MB in 8-bit RGB, 44 MB in RGBA and 67 MB in 16-bit, i.e. 600 to 1,600 MB/s for playback at 18 or 24 frames per second. A hard disk tops out around 120 to 250 MB/s and a SATA SSD around 550 MB/s; only an NVMe SSD keeps up with a 4K scan at real frame rate (“M.2” is a connector format, not a speed: an M.2 SSD can be SATA or NVMe). If possible, export your scans in 8-bit RGB: the alpha channel is useless for film and 16-bit doubles the size. If the drive is not saturated (Task Manager, Performance tab), the heavy filters are slowing things down, mainly stabilization, degraining and Remove Dust: disable them one at a time to find the most expensive one. Two remedies: the preview scale (1/2 button), which makes the filters process 4 times fewer pixels at 1/2 and 16 times fewer at 1/4, and the viewer speed (1x button), which at 0.75x requests 25% fewer frames per second and also relieves the drive. These settings only affect the display: export always starts from the full-resolution source frames, at the film's frame rate, with all the filters. In a reduced preview, very small defects are less visible: judge fine settings at 1:1 on a still frame.
Preferably use the software's Contact window (“Contact” menu): the diagnostic report is attached automatically. It describes your setup, your version and the latest events of the session, and never contains your films; without it, we cannot analyse a support request. Describe what you were doing and what you expected; a screenshot helps a lot. On an offline computer, the report can be exported to a file and sent from another computer.