Guide

My projects: Overview, Projects and Sessions. Data management: File browser, Storage and copies, Backup manager, Assistant, Quarantine (with NSTN) and Data hygiene. Calibration frames: Calibration. Capture planning: NSTN Planner (with NSTN), Equipment and Sites. Application: Getting started, Understanding NSTN (with NSTN) and Settings.

The navigation opens each workspace. The import button is at the top right. Settings includes theme, language, hidden projects and the assessment switch. The in-app Getting started tour explains the workflow and recognizes work already done. It is offered at first start and remains available in navigation and Help.

Choose “Import folder” at the top right. Subfolders are included. In Scan options, decide whether to calculate SHA-256 checksums: this reads whole files and takes longer, but enables copy and duplicate detection. The dialog states the cost: from a network drive or over Wi-Fi it takes as long as copying the files, and a large import can take hours. Missing checksums can be calculated later under Storage and copies.

The option to skip XISF files avoids opening processing results, retaining only their names and sizes. Selecting a drive root asks for confirmation because it scans the whole drive. Scanning runs in the background. Cancelling during header reading saves nothing; once import begins, finished entries remain. Later scans read new or changed files only.

Structure comes from headers: capture timestamps determine nights, OBJECT determines targets, and camera and telescope fields identify equipment. Midnight and multiple targets within one folder are handled separately. FITS exposures create projects; XISF results are associated with existing ones. Darks and bias go into a shared calibration pool.

Under Capture planning → Equipment, use Add setup to connect a camera and telescope under one name. Enter sensor width, height and pixel size. Use detected unbinned sensor values when unambiguous. Enter a reducer or Barlow factor only if capture software has not already written the effective focal length into the header.

Merge alternative spellings of the same camera or confirm a name. Header values remain unchanged. A setup supplies field geometry and distinguishes different equipment used in the same night. Describing filter type, lines and bandwidths is needed only for assessment and planning. If a header names no filter (a colour camera has no filter wheel) or the wrong one, add a filter with “Add filter…” and assign it to sessions in step 4.

Under Capture planning → Sites, add name, latitude, longitude, altitude, time zone and sky brightness as measured SQM or Bortle class. A measured SQM takes precedence. Choose a location on the optional map or enter coordinates manually.

The site converts UTC timestamps to local time and supplies twilight and moon geometry. Sky quality stays saved while assessment is disabled.

Under My projects → Sessions, the year grid shows observing nights. Green means every capture block has a target, setup and site; red means something is missing. Assign setups and sites groups matching equipment and header positions, rather than asking for every block individually.

Automatic assignment fills unambiguous cases and preserves existing assignments. A preview counts affected blocks and previous assignments. Frames without coordinates, such as some ASIAIR captures, share a separate row. Choose a night for its timeline and blocks. Assign a target using nearby catalogue suggestions if OBJECT was empty. Mark focus, framing and test shots as auxiliary captures; they stay catalogued but do not count towards integration. Colour the grid by assignment, activity, site or setup. “Assign filter…” gives chosen sessions a filter that replaces the header value; it is stored in the catalogue, no file changes, and withdrawing it brings the header value back. “Merge capture blocks…” rejoins blocks of one night that contradictory headers separated. It works only for the same target, optics and effective filter; for different setups, sites or analysis settings FITSORA names what must be made equal first. A merge can be undone. Several merges are undone in reverse order, newest first; if a later merge depends on one, FITSORA says so and blocks the undo until the later one is undone. After a filter change FITSORA shows “Changed filters need reassessment. Manual decisions have been kept.”; assessments and measured sky values that depended on the previous filter are then withdrawn.

Projects now bring together all nights for each target, with integration per filter, field of view, calibration matches and gaps. When new nights arrive, repeat import and assignment.

Counts projects, nights, frames, lights, integration and archive size. The year grid shows observing activity and a year review. Outstanding work includes unassigned sessions and projects with only one copy, plus unassessed frames when NSTN is enabled.

The project list is grouped by observing year. Each project has rename, merge and split actions, a star rating, and Active, Inactive or Completed status. Project overview shows sky preview, integration per filter, nights, latest capture, history, equipment, site, dates and raw-data size.

Panels appears for mosaics and lists integration per panel. Target and field overlays your optics on the sky and identifies the source of each value: header, setup or manual. Catalogue objects are drawn as ellipses and nebulae with traced outlines. A plate solution in the header places the field at its image centre, not at the requested target from RA and DEC. If the capture comes from N.I.N.A. 3 and no measured angle exists, FITSORA draws the field with the target’s planned angle as a dashed, rotated frame and labels it explicitly as planned: the actual image may be rotated differently, and a plate solution or a position angle you state takes precedence. A rotator reading is a motor position, not a sky angle. If no angle is known at all, the field is drawn north up. Missing values can be entered manually; no incorrect size is invented. The optional sky background comes from CDS.

Nights lists observing nights, with night analysis when assessment is on. Capture groups header configurations and stores capture goals. Goal progress counts only explicitly retained frames, so it stays at zero with assessment off. Analysis and Stacking plan need NSTN. Frames lists individual exposures and their stored FITS headers.

These actions only change the catalogue. Renaming retains original header names for later scans and leaves existing folder names. Merge combines whole projects or selected nights of the same field; mosaic mode retains each panel as its own project. If a merge empties a project, its rating, status and notes are removed; the dialog warns first.

Split undoes a recorded merge or moves selected nights to a new project. A project merge can be undone only after later capture-block merges inside it and later project moves of the same material are undone; the dialog says so before you confirm. Under Settings → Projects, hide finished images or test runs without deleting anything. Decide separately whether hidden projects count towards capture statistics.

Catalogue shows files at their current locations; Import sources shows their original scan folders. Rescan a subtree, ignore a folder, mark its lights as auxiliary, check availability or relink a drive whose letter changed. Remove catalogue entries or a whole import without deleting files; a database copy is made first. Plan a selected folder in the assistant. FITS header shows the values already stored in the catalogue.

Shows reachable archive drives and evidence of second raw-data copies. Evidence requires matching checksums on different storage locations. Two paths without checksums are not proof. Cards or lists put action needed first, with states for copies recognized, partial copies, no second copy, one storage location, or unverified. Capacity figures come from the latest scan.

This view reads the catalogue, rather than creating or restoring files. Calculate missing checksums reads each raw file without a checksum in full, in the background. Cancellation retains finished hashes. Rescanning alone will not hash known files. Create ordinary copies with the assistant, or managed versions with the Backup manager.

File → Database copies is also available through Storage and copies and the Backup manager’s More menu. Copies protect filter descriptions, sites, names, thresholds and assignments. FITSORA creates one at the first start of each week and before schema changes. Copy now makes one immediately. Restore selected retains the replaced database as another copy. Reset database empties the catalogue after a copy; image files stay untouched.

Managed, versioned backups of catalogued raw data. A backup job stores scope and destination; a run creates a backup version. Archive coverage describes stored copy evidence, rather than performing a fresh integrity check. Ordinary assistant copies and database copies have different purposes.

Choose a name and destination, outside any scanned capture folder. Select all catalogued raw data or chosen projects. Project jobs include matching calibration; full-archive jobs include the entire calibration pool and unassigned captures. Sidecars travel with their raw files. Optionally include the catalogue and catalogued processing results.

Unknown files are not included. The catalogue is a consistent database snapshot with operation logs, saved settings and connection details. Backups are unencrypted. Deleting a job only removes settings; recover it through Deleted backup jobs. Destination data remain.

The preview lists every file and whether it will be copied and verified, is already verified, or will be reused without rereading. Problems come first. Destination capacity and drive identity are checked; a disabled start button explains the block. Start backup begins writing and never changes originals.

After the first backup, the job remembers the destination drive. A different drive needs explicit confirmation. To force a full comparison of existing data, choose to read all existing files first. Reuse based on matching size and modification time is not a fresh content check.

Files are stored by content and reused across versions. New data are hashed during copying and reread at the destination before the version is published. The data balance separately labels copied, verified, catalogue and reused data on one scale, with planned amounts marked.

Timing and throughput are estimates based on the current run. Throughput includes copying and checking. Waiting for a drive is reported explicitly. Stop safely finishes the current file and retains verified work. Closing asks whether to continue or stop; Windows background tasks run independently.

Select a version to Check integrity: every backed-up file is reread and SHA-256 compared. A failed or interrupted check is never a passed check. Protect a version from FITSORA deletion or move it to recoverable trash.

Restore selected projects or files to a folder you choose. The preview checks availability, space, path lengths and collisions. Different existing files are never overwritten, and restored data are checksum-verified. Import restored folders afterwards; automatic relinking is not provided.

Restore the catalogue separately. FITSORA checks its checksum, SQLite integrity and schema, retains the current database and restarts. Open existing backup reads versions even after the original profile is lost; imported versions stay unverified until checked.

Green identifies evidence on another drive, blue on the originals’ drive and yellow no usable evidence. Unchecked coverage shows dashes, rather than an invented zero. Project coverage puts missing copies first and includes separate calibration and unassigned-data rows.

Coverage compares recorded evidence with present source and destination sizes and timestamps. Disconnected or unconfirmed destinations do not count as current copies. Changes preserving size and timestamp require an integrity check to detect. Another drive in the operating system does not establish physical independence or off-site protection.

Set optional daily or weekly runs in local time, a warning age for successful backups, and integrity checks every 1–365 days. Missed runs are caught up once. Nothing is scheduled by default.

By default, scheduled work runs only while FITSORA is open, checked every minute. Opting into runs while closed creates one Windows scheduled task per profile. You must be logged in and the computer awake. It is not a service and does not run after sign-out. The uninstaller does not remove the task; disable it beforehand or remove it manually.

Optional recovery tests run after successful backups or on an interval. They restore a sample of 1–5,000 files (20 by default for manual tests) to a fresh folder outside the backup destination, verify checksums before and after copying and fully read FITS data. A catalogue copy can be checked without touching the active database. Only the tested sample is covered, and image quality is not assessed. Test folders remain until you remove them.

Windows notifications are off by default. Choose problems only or all results per job. They are local, without email or an internet service. Windows may suppress them even when accepted.

Retention rules can keep the latest 1–365 successful versions, plus 0–104 weekly and 0–120 monthly versions. The last successful version remains. The rules propose cleanup; they never perform it automatically, including during scheduled runs. Every cleanup needs a fresh preview and confirmation.

Protected versions are protected within your profile, rather than against external programs, ransomware or profile loss. Moving versions to the destination’s trash keeps their files and does not immediately free space. Restore from trash, then check integrity again.

Free up storage permanently removes unneeded backup data after its own preview and separate confirmation. It cannot be undone. Locks coordinate processes of the same profile only; do not let other profiles, computers or programs modify the destination during cleanup.

Each run records per-file results, task timings and readable error explanations in the profile. Reports remain readable without the drive. Export HTML or JSON without overwriting an existing file. Reports include paths and filenames, rather than a catalogue copy.

Failed or stopped runs keep finished files. Retry creates a new version, reusing verified data. If originals are unreadable, a manual run can back up the available files only after explicit choice. It is marked Incomplete, has no verification date and never counts as the last successful complete backup. Scheduled runs are stricter: unreadable sources fail the run and are reported.

No encryption, cloud integration or guaranteed recovery. Targets are folders on local or mounted network drives. Same-drive copies do not protect against drive failure, and “different drive” does not prove independent hardware. Checks document what was read at the time; a destination can fail later. Recovery tests only cover their sample. Unknown files and native window settings are outside catalogue backups. Real USB/NAS use and installation across Windows configurations still need acceptance testing. Read the Limits page before relying on the pre-release.

Each file task builds a complete plan and writes only after confirmation. Copy and offload destinations must be outside scanned capture folders. Previous actions keeps reports and undo for moves.

Copy raw data makes ordinary verified copies of all or selected projects with needed darks and bias in a shared Calibration folder. Choose mirror archive structure or group by project. Missing calibration and incomplete copies are reported. There is no Backup manager version, schedule or coverage. Import the copy as a storage location and calculate checksums if you want it recognized under Storage and copies.

Organize archive moves frames by target, night and filter under a chosen root, using header information and confirmed equipment names. Darks and bias go into the pool; uncertain files stay in place and are reported. Across drives, files are copied, checksum-verified and only then removed at the source. Moves are logged immediately, so interrupted runs can be undone.

Free up space moves regenerable calibrated, debayered and registered intermediates of finished stacks, leaving masters and raw images. Space is freed only when the destination is on another drive. Offload duplicate files retains one instance of each same-drive checksum and moves the extras. These assistant tasks do not delete files.

Reports five categories: processed FITS exports without capture metadata, same-drive duplicates, XISF processing results, unreadable FITS with explanations, and darks or bias outside the calibration pool. Calibration still matches by settings even outside the pool. The page only shows findings; Show in file manager is its action. You decide how to handle each finding.

Calibration shows one row per camera and capture configuration. Different exposure, read mode, gain, offset, set temperature or binning requires matching frames. Counts show actual matches; missing sets are zero.

Darks and bias serve all projects with matching settings and form the shared pool. Flats belong to their night because dust and camera rotation tie them to a session. Project coverage identifies affected projects; you can inspect matching frames and explain gaps without assessment enabled.

Enable NSTN assessment in Settings. Describe your filter’s type, transmission lines and bandwidths, assign equipment and a site, and enter measured SQM or Bortle sky quality first. The Understanding NSTN tour explains the model within the application.

NSTN means Narrowband Saves The Night, a free exposure calculator on astro-tools.de. FITSORA implements its model offline; it never calls that website. The model uses site, moon, twilight, target altitude and filter descriptions to estimate suitable windows per line. A dual-band frame can retain usable Hα while O III is overwhelmed.

No pixel measurements of sky background, noise, clipping, absolute brightness, focus or tracking are made. Clouds, weather and image quality are outside the model. The geometry has an independent comparison; model sky brightness has not been independently validated.

NSTN Planner, Quarantine and Understanding NSTN in navigation; Analysis and Stacking plan in projects; night analysis and usable integration; assessed and usable capture-block figures; outstanding unassessed frames; thresholds and analysis profiles in Settings.

In a project’s Nights tab, choose a night to see windows per line and actual exposures on one timeline. Analyze night assesses it. Apply an analysis profile to the entire project to assess all its nights. Profiles include Balanced, Strict, Hα focus, O III focus and S II focus.

In Frames, manually keep or exclude an exposure. Later calculations never overwrite manual decisions; Return to automatic clears the override. Analysis summarizes retained, excluded and undecided frames, loss causes and each night’s contribution. Assessment moves no files.

Choose broadband and narrowband sky-brightness thresholds under Settings → NSTN assessment. Derive thresholds from each site’s SQM plus an offset when observing at different sites. Higher numbers demand darker skies and accept fewer exposures. Site-specific limits take precedence.

Analysis profiles save reusable night rules, including selected lines, model or custom windows and required fraction of each exposure inside the window. Applying copies the profile to a night; later edits or deletion of the profile do not change that night.

Choose site, target from the archive or search, and equipment or filter. The same model estimates usable minutes over the next 45 days, colours three months as a heatmap and shows twilight, moon, target altitude and filter windows on a local-time timeline.

A project roadmap estimates remaining comparable nights from capture goals. This is not a weather forecast and excludes setup time and subsequently rejected exposures. A site with sky quality and a described filter are required.

Separates assessed frames into all lines usable, some lines usable and no lines usable, plus unassessed frames. Plan quarantine previews the unusable selection; Execute plan moves it. Undo last run and Plan return restore it. Nothing is deleted; only you empty the quarantine folder.

Export working sets creates hard links per filter line on the same drive. Include frames with an unusable second line to retain good Hα even when O III in the same frame is unusable. A dual-band filter then supports separate stacking runs without full extra copies.

Save a named selection of assessed frames across nights and filter lines. Exclude nights or lines as needed. The plan changes no files and survives reassessment; FITSORA warns when analysis has changed.

File output is a separate step. Choose a destination outside capture folders, verified copying or same-volume hard links, optionally matching calibration, then check and create the output. Existing files are never overwritten. Remove verified output deletes only unchanged files belonging to that output.

Clear the switch to hide the areas. All stored assessments remain in the database. Enable it again to see them immediately, without recalculation.

Find the version in the title bar or Help → About FITSORA. Copy system information includes the version, operating system, program and data folders.

The Windows Start menu self-check runs without the main interface and writes FITSORA-Selbstpruefung.txt to the desktop. Use it especially if the application will not start. Reconnect absent drives or relink changed drive letters in File browser. Restore an earlier database through File → Database copies if necessary.

For backup issues, read the version’s Report and open the log folder through More to find fitsora.backup. Send feedback with your version and self-check report to the contact email below. There is no guaranteed support response.