FITS Format

[MRPUGH]

Hello
under this latest release, all my files (when doing a star alignment for example) are saved as .xisf.  Even though in the global preferences I have specified .fit as the file extension.  In the previous release I was able to specify the file type under the star alignment dialogue, but that has gone in this release.  I think you need to give users the option as .xisf isnt usable in any other software.

Bug?

thanks
martin

[oldwexi]

Martin!
BatchFormatConversion
converts all your xisf data into the
requested format including the  fits format.

Gerald

[sreilly]

But that shouldn't be a required or necessary step. To be able to set the default to FITS is a logical setting as other software can not read the .xisf proprietary PI format. This is like you have no choice when using PI and that's heavy handed especially for a non standard format. This is quite possibly my first complaint with PI but I have strong feelings about this. Otherwise we really don't have preferences do we? Sorry but this touches a really sore spot for me. As much as I like PI and use it 99% of the time I do use other software at times and sometimes it for no other reason than to do a different evaluation of the data.

Steve

[Niall Saunders]

Hi Steve, Gerald, Martin,

You are missing a crucial point here - the XISF format wasn't just created by Juan and the PI Team 'on a whim' - it had become an absolute necessity if PixInsight was going to continue to develop.

The FITS format has been around since the '80s (if not earlier) - last century, last millenium. FITS simply cannot handle the extra data that PixInsight processes now need to be able to include with 'interim' images saved during the processing workflow. So, in order to make it possible to use these extremely sophisticated interim processes, the data has to be saved in a workable format - hence XISF.

Further, at what point do any of the interim images need to be loaded into other software packages? Why would you need to do that? And, even if you did have some reason for doing that, why not just export those images in a format appropriate to that software?

As for other software packages not being able to read XISF data, well that just shows how poor their development actually is. The XISF format has been fully documented throughout its development phase, and there have been several RFC cycles that provided other software writers the opportunity to raise issues or ask questions. There has even been a full 'link libraary' made available to any software engineer who wanted to, or needed to, include the format in their programming environment.

So, if you are dealing with software that cannot (or at least cannot be bothered to) read and write images in the XISF format, you have to raise the issue with them, not PixInsight.

One thing for absolute sure - XISF is not a format that is in any way a 'restriction' for astro imagers - if it was, then why would some of the leading professional, commercial and scientific, astronomy organisations be wholeheartedly supporting XISF, and PixInsigh in general?

Nobody says that you have to drive a motor vehicle - there are still plenty of horses and oxen-drawn carts out there 

[sreilly]

Actually I'm wondering why I didn't check before but all my images being processed are being saved as FITS and the preferences are set for FIT and appears to be working. I haven't seen any process that hasn't worked in PI because of this extension but then I'm not a pro at this either so there may well be some that need the XISF format and I'm simply uninformed, wouldn't be the 1st time. Can you point to these processes that depend on the XISF format or some text explaining it's importance please?

Thanks,

Steve

[oldwexi]

Hi Niall!
I fully agree with the logic of your reply. When i gave the advice "how to"  i wanted only to help.

I was thinking about adding comments and explanations why and why not, but i realized
i cant do that in the needed way as you did.
I am using only xisf as i dont need other software which maybe does not support xisf.
I am happy that PI helped me to stop Softwarejumping when doing image processing.

Thanks for your statement!

Gerald
 

[Niall Saunders]

Hi Steve,

Once we see the detailed documentation for some of the new processes in v1.8.5, the advantages - and requirements - if XISF over FITS should become more apparent.

I will admit that, when I swapped over to XISF (quite early on in the development phase), I never looked back, and I never had to concernmyself with compatibility issues (I don't have any need to use any other software once my data has been acquired).

So, I'm happy to confess that I don't currently know which processes benefit from, or require, the use of XISF. There are postings by Juan that have discussed this in the past, but I don;t have the time to do a comprehensive Forum Search right now (I need to clear my Inbox, and get back to an Embedded Controller project that is driving me nuts !!)

[Alejandro Tombolini]

Hi, one of those process is the automatic CFA pattern detection that depend on image properties that cannot be stored in FITS files.

Saludos, Alejandro.

[sreilly]

Well something seems amiss at this point because changing the default from .fits to .xisf doesn't seem to change anything under the Global Preferences. I still get .fits on images calibrated. Maybe a bug? I've hit apply changes and load defaults as well as restarting PI altogether.  See the screen capture below:

[Alejandro Tombolini]

Hi Steve, that is the "Save as" setting.

What have changed is in tools like StarAligment that previously allowed you choose an extension and now do not have that option, the tool will save the aligned images in xisf

[sreilly]

I did see that but didn't really think of it in that sense. So why wouldn't you save in .xisf when calibrating the images? I'll set accordingly.

Thanks,

Steve

[sreilly]

I just re-ran a set of images aligning them and they are saved as .fits, not .xisf as expected. I've looked at the obvious places and see no option to set the file extension other than in the Global Prefs. Looks like I broke it now that I'm trying to do the recommended extension. I guess I could try a reinstall and maybe I should uninstall first?

Steve

[Alejandro Tombolini]

Hi Steve, that is strange, should not happens. All the files should be saved as .xisf during Calibration or Registration even if you set fits in Preferences that only apply to "Save as" when "Remember file type" is unchecked
Saludos, Alejandro

[MRPUGH]

Yep....a load of chat, and no one as answered the question.

Very narrow view if you think PI is the only way to go in astrophotography image processing and an insult to all of the other packages that are out there.

Get over yourself.

Martin

[chris.bailey]

Martin

When Photoshop first started saving out as .psd files as default there was uproar amongst digital imagers. 15 years later it is a format that can be read by pretty much any all major imaging software. .xisf is much the same scenario.

Turning your question on its head, what software do you want/need to use to do any processing of the intermediate files created by PI?

Chris