QualitySessionMeter 1.4.2.1

• Raise the default combined sudden signal-and-star loss limit from 20% to 35%, and the direct sudden signal-loss limit from 35% to 50%.
• Lower the minimum session signal from 40% to 35%. Moderate 20–35% sudden losses remain usable warnings and cannot lower the clean reference.
• Migrate the complete 1.4.2.0 default triplet once. Profiles with customized signal limits keep their values.
• Correct the Options text: combined signal loss does not require a brighter background.

QualitySessionMeter 1.4.2.0

• Keep matched-star photometry available when shape classification is inconclusive. Ignore unresolved single-pixel detections when counting failed shape measurements.
• Follow coherent, timestamped stellar-signal trends and recent star/background levels. Normal gradual changes no longer become accumulated losses against the start of the night.
• Reject sudden matched-signal loss with fewer stars without requiring a brighter sky. Freeze reference learning during suspicious changes, including rapid multi-frame attenuation.
• Add Minimum Session Signal, default 40% of the initial matched reference and configurable from 5–80%. Signal limits include a measured-scatter allowance.
• Keep clean references up to six hours through rejected intervals, with bounded trend extrapolation. References older than two hours cannot authorize a star-count recovery.
• State why guide recovery was not established, including verified regions and borderline eccentricity.
• Add session report locations: default local folder, a chosen folder, or beside the LIGHT folder. Keep the active session in one location; use the new location from the next session. Show the actual folder and any fallback or write failure in Options.
• Show photometry separately from unavailable shape evidence in the public and OpenAstro inspectors.
• Validate against the September 20 HOO session and the earlier QUAD/shape regressions. See field results.

QualitySessionMeter 1.4.1.1

Added

• Check central stellar shapes on every monitored LIGHT, including frames with normal guiding. Inspect four outer regions and an extended stellar profile on suspect frames to detect elongation and faint secondary images.
• Compare matched-star signal with earlier clean exposures. Reject signal loss above the configured limit, or a smaller loss accompanied by brighter sky and fewer stars.
• Show extended stellar profiles, measured regions, signal comparisons and rejection reasons in the N.I.N.A. panel, Web Dashboard, reports and companion API.

Fixed

• Prevent gradually degraded frames from lowering the clean reference. Keep clean matched-star references for up to 120 minutes, separated by imaging context and pier side.
• Preserve signal rejection when a guide flag is cleared. Do not clear a star-count rejection solely because matched-star flux passes while the sky is brightening.
• Retain guide rejections when the image analysis cannot cover the expected displacement or provide enough reliable stars.

Settings and compatibility

• New Reject Stellar Signal Loss option, enabled by default with image analysis. Default limits: 35% direct stellar-signal loss; 20% combined loss with at least 3% brighter background and a qualifying star-count drop. Limits remain configurable.
• Same plugin identifier and minimum N.I.N.A. version: 3.3.0.1057, .NET 10. Existing profiles are retained. Rejected-file handling remains reversible; acquisition is never paused for recovery.

Validation

• Chronological replay of 241 real LIGHTs, independent five-region photometry, prior-night shape regression, automated assessment and dispatcher checks, and desktop/mobile visual checks. See validation results and limitations.
• Official packages exclude development tools. Release notes are taken from this changelog; development builds are available only as CI artifacts.

QualitySessionMeter 1.4.0.2

QualitySessionMeter 1.4.0.2

• Stellar verification of guiding-rejection candidates, with configurable shape limits and a recovery margin.
• Dashboard inspection of measured stars, applied limits, recovered frames and retained rejections.
• Updated OpenAstro integration and shorter native/browser labels.
• Smart Recovery waits removed; existing saved items remain compatible.

Install: download QualitySessionMeter.1.4.0.2.zip. Requires N.I.N.A. 3.3.0.1057 / NIGHTLY #057 or compatible and .NET 10. Close N.I.N.A. before replacing the existing DLL, and retain a backup outside the active Plugins folder.

N.I.N.A. catalog: this version replaces the pending 1.3.1.1 submission in PR #689. It will appear in the Plugin Manager after upstream approval and feed publication.

The catalog ZIP preserves the exact production DLL already verified from commit 2cd4345ff05afe11265275114e694ff25c7d62dd; documentation and packaging were updated for release. Its manifest was generated using the official N.I.N.A. script, and its checksum was verified after download. The earlier candidate archive remains available unchanged.

Validation: 38 assessment/API checks, dispatcher checks and WPF/browser visual tests passed. Selected-image replay covers star-shape rejection and recovery. A complete live 1.4.0.2 rejection/file-handling night is not yet documented; see validation scope.

QualitySessionMeter 1.3.1.1 (pre-store test)

Pre-store test release. Install the field-test-with-synthetic-lab ZIP for current validation. The production ZIP is generated in parallel to verify the future N.I.N.A. store package and intentionally excludes Synthetic Lab.

QualitySessionMeter 1.3.1.1

First production package submitted to the N.I.N.A. plugin catalog. Requires N.I.N.A. 3.3.0.1057 or later (.NET 10).

Packaging correction

The production DLL is preserved before development regression and checked for Synthetic Lab/test harness types before packaging. The downloaded release archive was independently inspected and its SHA-256 matched the manifest. Runtime behavior and the permanent plugin GUID are unchanged.

This release supersedes 1.3.1.0, whose archive accidentally included the development build. That package was not submitted to the catalog. All release gates and the official gather.js manifest validation passed for 1.3.1.1.

Catalog submission

N.I.N.A. manifest PR #689 is awaiting upstream review. Plugin Manager availability follows upstream merge and catalog publication. Future updates use the standard N.I.N.A. manifest system.

Material AI assistance was used during development. astropuzzo is the accountable human maintainer and retains responsibility for understanding, testing, security, privacy, licensing, provenance, debugging and maintenance.

Full changelog: 1.3.1.0...1.3.1.1

QualitySessionMeter 1.3.1.0 — superseded packaging candidate

This package is not suitable for N.I.N.A. catalog installation: post-release inspection detected Synthetic Lab in the DLL because the development regression overwrote the shared build output before packaging. It was not submitted to the official catalog.

Superseded by version 1.3.1.1, which preserves and checks the production DLL before packaging. The original assets and checksums are retained unchanged for traceability.

QualitySessionMeter 1.3.1.0 (pre-store test)

Pre-store test release. Install the field-test-with-synthetic-lab ZIP for current validation. The production ZIP is generated in parallel to verify the future N.I.N.A. store package and intentionally excludes Synthetic Lab.

QualitySessionMeter 1.3.0.9 (pre-store test)

Pre-store test release. Install the field-test-with-synthetic-lab ZIP for current validation. The production ZIP is generated in parallel to verify the future N.I.N.A. store package and intentionally excludes Synthetic Lab.

QualitySessionMeter 1.3.0.8 (pre-store test)

Pre-store test release. Install the field-test-with-synthetic-lab ZIP for current validation. The production ZIP is generated in parallel to verify the future N.I.N.A. store package and intentionally excludes Synthetic Lab.