Hello, I’ve observed some peculiarities with AMSP using the DwarfIII files.
I have several projects that are either using LP filters or Astro filters in their respective Lights folders.
AMSP is sometimes unable to define the files as Lights because the information is missing in the FITS header IMAGETYPE.
The same applies to the MASTER FLATS/DARKS/BIAS
I set the parameter IMAGETYPE ‘Light’ in the Lights and the type is then correctly recognized.
In the master files I set the value as IMAGETYPE ‘Master_XXX’ but the file is not recognized as MASTER and the stacking therefore fails.
Which parameters in the FITs header need to be set for recognition, even if you want to include the entire Dark library? Does the filename also require a special syntax?
The best approach would be to ask DWARF to add the necessary keywords. It’s a minor change for them, but that would be preferable.
A master is identified once the stacking process is done, producing the STACKCNT keyword.
Thank you for your reply,
I succeeded with the Master Dark, but for the Master Flats I get the message “Unknown type”.
This is the Fits header of the bias-calibrated Master FLAT.
SIMPLE = T / file does conform to FITS standard
BITPIX = 16 / number of bits per data pixel
NAXIS = 2 / number of data axes
NAXIS1 = 3856 / length of data axis 1
NAXIS2 = 2180 / length of data axis 2
EXTEND = T / FITS dataset may contain extensions
COMMENT FITS (Flexible Image Transport System) format is defined in 'Astronomy
COMMENT and Astrophysics', volume 376, page 359; bibcode: 2001A&A...376..359H
BZERO = 32768. / Offset data range to that of unsigned short
BSCALE = 1. / Default scaling factor
PROGRAM = 'Siril 1.4.2' / Software that created this HDU
DATE = '2026-03-17T10:34:05' / UTC date that FITS file was created
XBINNING= 1 / Camera binning mode
YBINNING= 1 / Camera binning mode
BAYERPAT= 'RGGB ' / Bayer color pattern
STACKCNT= 10
HIERARCH IMAGETYPE = 'FLAT '
HISTORY Calibrated with a master bias
END
I have no idea how this value came about with the HIERARCH extension.
I entered the values, just like with the DARKs in SASPro, using the FITS header batch processing.
In Siril’s Fits Editor view, HIERARCH is not displayed, but only via the Saved Headers view.
I edited it again individually in SIRIL and now it works.
I am now comparing the differences in the result with my workflow and with AMSP.
One more question: what about mosaic projects?
Interestingly, the activated plate solving information is not included in the result.
I tested the framing and mosaic modes on the Seestar S50 and DwarfIII in AZ and EQ modes, but unfortunately had no success.
Even if I set the framing to 1.1x1.1, the registration fails.
I suspect that this mode is not implemented via AMSP.
