Release date: March 9, 2022
This is the first stable release of the 1.0 branch.
After almost 10 years of development, we finally think Siril is stable enough and complete enough to propose a 1.0 version, for all deep-sky imaging use cases.
This version concludes the 0.99 branch and its several intermediate versions that introduced a new graphical user interface compared to the previous stable 0.9.12 version. A list of all changes compared to this previous stable version can be found in the pages of each intermediate 0.99 version, or for short in the ChangeLog file. Below is a list of changes since the previous release, the second release candidate for this version.
Starting with this version, we provide links to the commands reference page for each version: see the list of commands for siril-1.0.0.
See the page of the main site for a detailed presentation of the version and download links.
Changes since 1.0.0-rc2
- Fixed memory consumption for all sequence operations
- Fixed settings for sequence export as webm film with VP9 codec
- Removed use of lo/hi cursors and fixed normalization for export
- Fixed load and close commands for scripts in GUI
- Fixed Bayer pattern in SER after extraction
- Fixed registration crash for small images
- Improved main panel separator positioning and keeping it in memory
- Improved speed of FITSEQ detection when scanning sequences
- Reintroduced compatibility with OpenCV 4.2 with disabled features
Inevitably, some problems exist in every release, even if we do our best to verify that everything works as expected. For convenience, here is a list of problems that were identified after the release, potentially corrected in subsequent minor versions:
- stopping an executing thread makes the program crash (close without notice) - filed as bug #792
- moving the mouse over the image when it's being updated crashes siril - filed as bug #791
- loading another image when a tool with preview enabled is active crashes siril - filed as bug #783
- filtering out images from SER or FITSEQ sequences in median or mean stacking can result in dates (start and end) and live time (sum of exposures) being wrongly reported in the FITS header of the result (fix for siril 1.2 and fix for siril 1.0)
- stacking with sigma or MAD rejection will be much slower than it should on 16 bits images, because of a threading optimization issue (fix for siril 1.2)
- stars that have peak values over 1 in 32 bit images will be discarded in star detection (registration, plate-solving, color calibration)
- the translation-only option of global registration was mistakenly deactivated in the builds of the release