A simple question

[gianpri]

Is it necessary to convert the .CR2 files to .XISF before processing them in Pixinsight?
Thanks.
John

[pfile]

it is not strictly necessary.

however, if your first step is ImageCalibration, your CR2s will be read in, calibrated, and written out as XISF anyway.

another issue is that CR2 files can be *very* large and must be read into memory in their entirety. this can cause ImageIntegration to use a lot of memory. XISF files (as long as they were not saved with compression turned on) can be partially read from disk, so integrating uncompressed XISF files is easier on the computer's memory system.

rob

[gianpri]

Thanks.
So even in the creation phase of the Master Files, where the use of Image Integration is foreseen, would it be advisable to have XISF files already?
John

[pfile]

like when creating a master bias or master dark? yes, you can use BatchFormatConversion to convert the CR2 files to xisf and then run ImageIntegration on the xisf files. i don't know how many people actually do this though. i think if you had a huge number of bias frames to integrate (3-400) it might make sense. it really all depends on how much memory your computer has and how big the OS's swap file can grow before filling up the disk.

rob

[wvanreeven]

I always use the CR2 files as inputs and then work with the XISF files as they get produced by PixInsight and never had any issues. As a matter of fact, file conversion should be avoided in general since this *may* lead to data loss. Not sure in case of CR2 to XISF conversion though.

In any case, to answer your question "Is it necessary to convert the .CR2 files to .XISF before processing them in Pixinsight?": no!


Wouter

[gianpri]

Thanks Pfile, thanks Wouter; it's all very clear.
John