How wide is my sensor? ASCOM driver vs Raw via Pixinsight

Linwood

Well-known member
Jul 28, 2020
101
6
By the way, cropping is probably the RIGHT answer for this sensor. I haven't tried to tell precisely where this weird pattern starts, but the right side of an image (this is a very stretched master dark) is clearly a mess and not real data.

RightSide.jpg
 

bulrichl

PTeam Member
Nov 2, 2016
938
80
La Palma, Canary Islands
The data in the FITS file is not a bit to bit copy of the data in the ARW file, they are very similar but not identical. So these are different frames.

However, you are absolutely right: from the FITS file, crop 32 pixels from the right margin and you get the same image region as in the ARW file when opened with PixInsight.

Bernd
 

Linwood

Well-known member
Jul 28, 2020
101
6
The data in the FITS file is not a bit to bit copy of the data in the ARW file, they are very similar but not identical. So these are different frames.

However, you are absolutely right: from the FITS file, crop 32 pixels from the right margin and you get the same image region as in the ARW file when opened with PixInsight.

Bernd
Got lost in there somewhere. I'm saying the image data elements, as opposed to the metadata, are supposed to be identical. Which I think is what you are saying in the second paragraph.

Just to ask a different way -- whether I capture by ARW or FITS it sounds like I get the same result (+/- the crop). So if I manage the crop I can capture either way.

I think the driver developer found the issue crashing APT, so this may become moot.
 

bulrichl

PTeam Member
Nov 2, 2016
938
80
La Palma, Canary Islands
Identical frames not only have the same image geometry but also identical intensity values at each pixel position. The two uploaded frames (after cropping of the FITS file) are not identical, the intensity values in the ARW file are slightly different from the intensity values in the FITS file.

SGP e.g. allows that one frame is saved twofold, once in the proprietary raw format and once in FITS format. However, SGP (or the camera driver) scales the intensity values (at least in case of Canon CR2 files) to 16-bit [0, 65536]. The scaled intensity values are written to the FITS file. This scaling can be reversed by a PixelMath operation. After having reversed the scaling, the files are identical.

I don't know whether APT is capable of saving one frame twofold.

Bernd
 

Linwood

Well-known member
Jul 28, 2020
101
6
Well, the camera allows me to save both which is what I did her (I think the driver/APT does as well).

That's interesting, the driver author told me they remained unchanged. And you are right, I should have looked -- very close, but not identical.

D_2020_08_14_10_56_59_Bin1x1_60s__na
count (%) 100.00000
count (px) 61209600
mean 512.5
median 512.0
avgDev 10.1
MAD 8.0
minimum 424.0
maximum 16380.0

A7409493
count (%) 100.00000
count (px) 61209600
mean 512.8
median 512.0
avgDev 10.6
MAD 8.0
minimum 436.0
maximum 16380.0


I need to look at that a bit more closely, for curiosity if nothing else. Thank you.
 

pfile

PTeam Member
Nov 23, 2009
6,029
159
By the way, cropping is probably the RIGHT answer for this sensor. I haven't tried to tell precisely where this weird pattern starts, but the right side of an image (this is a very stretched master dark) is clearly a mess and not real data.

View attachment 8929
out of curiosity was this master dark debayered or is this the raw CFA data?

rob
 

Linwood

Well-known member
Jul 28, 2020
101
6
Completely raw, not processed except that the FIT one was converted from AWR to FIT by the driver.
 

pfile

PTeam Member
Nov 23, 2009
6,029
159
ok, thanks. i just noticed some duplicated edge pixels coming out of Debayer and wondered if this was related but it seems not.

rob