VStar 2.13 testing release 7
Citizen Sky is now officially permanent part of the AAVSO. In the coming weeks we will be moving additional content to the AAVSO site and freezing this site as an archive of the 1st three years of the project. Please visit
the new landing page for future updates.
Hi Sara, all
Thanks again to Sara for deploying this new testing release. As Sara has said, your testing and feedback would be very much appreciated.
A lot of what has gone into 2.13 so far has been driven by the needs of the Mira Fourier Coefficient Team, but there are also plenty of unrelated bug fixes and improvements as Sara has mentioned.
Sara has attached the Change Log and I thought I would just list the main points for this testing release, with additional comments in italics:
- Removed duplication of mean series in phase plot that wasted memory and lead to asymmetry in mean curves. This also avoids the need to duplicate other synthetic series (e.g. Model, Residuals). I'm surprised no-one complained about the mean curve assymetry up until now actually.
- Added ability to show polynomial fit coefficients via the Models dialog.
- Increased maximum polynomial degree from 20 to 30. This was a request from a Variable Stars South member who wanted to create polynomial fits of higher degree for a bright Cepheid variables project.
- Fixed stop request handling in algorithm and modelling code. Previously, the algorithm (e.g. WWZ, DC DFT) would appear to be interrupted but an active compute-intensive thread was still active; essentially, the GUI thread was interrupted but not the algorithm run.
- Removed use of set-always-on-top method for dialogs, instead obtaining current active window and using that as the parent of a dialog to be opened. Apart from the nasty side effect of having windows over the top of every other application's windows, this also caused some plugins to fail with security policy exceptions when VStar was run via WebStart.
- Added DCDFT top-hits handles on power spectrum and other plots to make it easier to select points of interest on plots. This is something I have been meaning to add for a long time.
There are still many improvements possible to the interface and elsewhere, but I'm reasonably happy with VStar's current state.
The Help documentation still lags behind the current level of functionality. Even since this most recent deployment I found and corrected a couple of statements that are no longer true:
1. It has not been necessary for a quite awhile to login in order to download AID datasets.
2. The period search and polynomial fit sections contained some out of date comments re: the series on which analysis is performed. WWZ was also still listed as a future work item.
Please report bugs or provide other feedback. Thanks.
Regards,
David
Hi all
I found a bug in which a mean source series selection of Filtered, Model, or Residuals was not persisting across multiple invocations of the plot control dialog. I've fixed and committed that.
Any other bug reports yet...?
Regards,
David