TheSkyX "OpenFits" will not correctly open .xisf files saved as .fits in PI

[jeffweiss9]

I found that saving .xisf files as .fits files (latest PI release) will not let TheSkyX (10.3) correctly open them for "image link".  The files open up but there is nothing visible in them and Image Link consequently cannot plate solve them.  There appears to be something nonstandard about the FITS files that PI is generating this way.
Thanks,
Jeff

[pfile]

maybe try saving them as 32b or 16b integer? the 0.0-1.0 floating point format that PI uses for floating point fits files is somewhat unique.

rob

[jeffweiss9]

Thanks, Rob-
32b integer showed counts in the histogram (unlike 32b floating point) but the image was still black.  It did plate solve the blank image, however, although the linked image itself was not displayed; just the drawn outline.
16b integer FITS did everything correctly so I guess that's what TSX is looking for.
I note that (even with 16b integer from PI or FITS from TSX) with my ML16200 FOV, I have to blind solve using All Sky so it crops the images to 25% in order to reduce the star count sufficiently for Image Link to work.  But I that would be a TSX problem. (Makes me wonder what people do with Image Link and a 16803)
Clear skies,
Jeff

[pfile]

i have only ever used TSX with an 8300M + FSQ106 and even then i had lots of problems with Image Link. when we were using a 16803 @ SRO we used ACP to drive it; pinpoint seemed fine with the giant images.

how do you like the ML16200?

rob

[Juan Conejero]

PixInsight generates fully standards-compliant FITS files. Please submit your .fit file to NASA's online FITS File Verifier service:

http://fits.gsfc.nasa.gov/fits_verify.html

Unless it says there is an error, there is no bug in our FITS support module, and the file in question complies with version 3.0 of the FITS Standard document. If other applications are unable to correctly open and display valid FITS files, then you are facing either a bug in another application, or a FITS interoperability problem. In the latter case, you may ask them to improve their FITS support in order to make it more flexible.

The fact that the image was originally in XISF format is irrelevant.

[jeffweiss9]

how do you like the ML16200?

Rob-
I like it a lot although still dialing it in with my 1x field flattener.  Seems quite sensitive (measure 6dB greater dynamic range than the ML8300): 

http://www.astrobin.com/254937/I/

I think I've got it pretty close with my 0.75x FF/FR with a new image in midstream.

Cheers,
Jeff

[jeffweiss9]

Thanks, Juan-
Seems to be a TheSkyX problem.
Regards,
Jeff