stacking issues, somethings not right.

[bomberbaz]

I am at a loss with what I am doing wrong but something isn't right, I have been stacking some data of the cocoon nebula just to try and see how it's looking (I know there isn't yet enough) and honestly my results so far are awful and strange.

Equipment: 72ED ota, 183MC Pro and Askar DB Filter.

Data. 45 x 2 minute lights, 32 2m darks, 32 0.2 flats/dark flats

Stacked in DSS and here is the first noticeable issue, the TIFF files are only around 6 (six) MB, hmm this does not look right. In FITS they are 236 MB but the quality of either is absolute pants. I downloaded and installed the new version of DSS as I thought maybe there was a file corruption but this didn't change anything. 

I have attached the Autosave TIF stack and one each of files used. If someone would be so kind as to have a look and see if it is something glaringly obvious that I can't see for looking I would be very grateful.

UPDATE: I did a bit of playing around stacking lights only, the lights/dark, lights/flats etc, and I found that when I omit just the darks from the stacking process, the file size goes from 6MB to 174MB which is where I would expect it to be. So it appears that there is an issue with the darks, although I do not know if it is one or more or indeed all of them. On the face the sizes are fine, looking through them I see no visual corruption.  I have also attached the masterdark tif.

After retaking the darks, the issue is sorted and the image is now good, albeit grainy.

I will post this anyway in case anyone else has this problem as it could be a good reference for trouble shooting. 

Steve

Autosave.tif

Light_IC5146_120.0s_Bin1_183MC_gain115_20230824-222350_0001.fit

Flat_200.0ms_Bin1_183MC_gain115_20230825-132945_0004.fit

Bias_200.0ms_Bin1_183MC_gain115_20230825-133412_0001.fit

Dark_120.0s_Bin1_183MC_gain115_20230825-095455_0001.fit

MasterDark_Gain115_120s.tif

Edited August 25, 2023 by bomberbaz

[bomberbaz]

I have actually now managed a half decent effort for around 75 minutes. Far from enough but promising. 

[Elp]

Are you using a region of interest smaller resolution, my 183mm fits are normally 40Mb each?

[bomberbaz]

40 minutes ago, Elp said:

Are you using a region of interest smaller resolution, my 183mm fits are normally 40Mb each?

Sorry but I am not sure what you mean. I just screwed it onto the OTA and used it as it was out of the box, I have not tinkered with anything purposely since then. Can you elaborate please?

 

[Elp]

I just mis interpreted your above file sizes being small but you're referring to the initial stacked file.

The calibration files shouldn't add much to the final stacked file size as they're simply there to do a mathematical operation on the stack, strange how it makes such a difference to the file sizes you're seeing.

The size of tif files can vary depending on what and whether a compression algorithm has been applied to the files, using DSS I don't think there's an option for this though.

 

[bomberbaz]

3 minutes ago, Elp said:

I just mis interpreted your above file sizes being small but you're referring to the initial stacked file.

The calibration files shouldn't add much to the final stacked file size as they're simply there to do a mathematical operation on the stack, strange how it makes such a difference to the file sizes you're seeing.

The size of tif files can vary depending on what and whether a compression algorithm has been applied to the files, using DSS I don't think there's an option for this though.

 

My initial stack with the now good darks with no processing is coming in at 173MB, that's around what I would usually expect. However this effort that my initial post was about, with the apparently dodgy darks shrank to 6MB. (see attached)

I have never come across this before, I hadn't done anything different to what I normally do and it was very strange.

[Elp]

The one thing I can think that's happened (I don't know how) is the operation has somehow deleted data in the stack (so smaller final file size), you'll see this if you stretch the files very high, the first one might not contain much signal/details, whereas the newer one will.

[michael8554]

I stacked your single Light, Dark, Flat, Bias in DSS.

I got a normal looking output - black background, white stars.

The Autosave TIFF was 96bit depth 6MB.

I did a "Save As", looked the same, that was 48bit depth 118MB.

There may be a rogue frame in your images.

Michael

[teoria_del_big_bang]

I do not know if significant but somebody else had exactly the same issue only a few days ago.
I say the same issue, maybe they were using different software (I think PI but maybe using DSS to stack) and they had just taken a new set of darks and I did reply so should be able to find the thread, but they also cured the issue themselves by taking new darks.

Strange 

Steve

[teoria_del_big_bang]

That thread I was on about is HERE

Steve

[bomberbaz]

1 hour ago, teoria_del_big_bang said:

That thread I was on about is HERE

Steve

Yep read it, identical problem in every way. I am wondering if there is anything in common with our gear which identifies why this is occuring!