Release date: May 27, 2019
This release focused on improving some existing but important aspects of image processing: background gradient removal, color calibration using the astrometry feature introduced in the previous release and star photometry, and performance improvement. See also the change log in the siril commands reference section.
Recently, source code, roadmap and issues management have moved to gitlab.com. Please report issues you encounter with siril on this website, this will help us fix them and make siril a better program.
Remember that Siril is developed without Windows or Mac and that we rely on Windows and Mac users to test and give us feedback about integration or porting problems.
$ sudo apt-get install siril
- Ubuntu / Linux Mint:
$ sudo add-apt-repository ppa:lock042/siril $ sudo apt-get update $ sudo apt-get install siril
- Fedora, Mageia 6: All
# dnf install siril
- Arch Linux has both the release version and the git development version in the user repository (AUR).
- AppImage: If you don't find a package for your distribution, you can try to download the AppImage binary. It has been tested on many systems and could work on yours. Make the file executable, and run it. That's all.
macOS ( > 10.9)
- Download the latest DMG disk image for Siril and follow the instructions.
- Thanks to a French Siril user (Colmic) there are now two possible methods to install Siril. The first one is based upon an installer and it will also install some very useful scripts in the right place. If you did previously installed the 0.9.9 version, some conflicts may occur. Please remove old Siril application and also the siril directory located in %AppData%. It is important to read all messages during installation.
- As usual we provide an archive with all the needed files in it. We recommend you create a shortcut for the executable at any convenient place. The executable is in the subdirectory bin and is named siril.exe.
What's new in Siril 0.9.11
- New photometric color calibration tool
- New background extraction tool working with 64-bit precision and dither
- New image filtering for sequence processing: now possible from the command line and with multiple filters
- New GTK+ theme settings: it can now be set from siril to dark or light, or kept to automatic detection
- New normalization to 16 bits for RAW images with less dynamic range (in general 12 or 14).
- Added drag and drop capability in the conversion treeview
- Added output file name argument to stacking commands
- New command setmem to limit used memory from scripts
- New clear and save buttons for the log in GUI
- Added German translation
- Improved processing speed by optimizing sorting algorithms to each use. This work was done by Emmanuel Brandt who spend a large amount of his time to provide the best compromise between speed and performances.
- Improved free disk space feedback and checks before running preprocess, global registration and up-scaling at stacking
- Improved mouse selection by making it more dynamic
- Taking into account the Bayer matrix offset keywords from FITS headers
- Displaying script line when error occurs
- Allow registration on CFA SER sequences
- Processing monochrome images requires less memory, or can be more paralellized if memory was the limiting factor
- Fixed dark optimization
- Fixed free disk space detection and usual processing commands on 32-bit systems
- Fixed free memory detection for stacking in case of up-scaling ('drizzle') and memory distribution to threads
- Fixed bug in FFT module
- Fixed bug in the drawn circle of photometry
- Fixed build fail with OpenCV 4.0.1
- Fixed SER sequence cropping
This version was especially tested on a 32-bit ARM machine with 2GB of memory, like the C1 server here. It can adjust the number of threads to the available memory, computes the available memory and disk before computing something that requires a lot of it and indicates if there is not enough or if it had to reduce the processing speed to accommodate.
It is much faster than the previous version for many operations, and thus will allow more data to be processed than before in the same time, or require a less powerful computer to do the same thing as before.